{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Table of contents\n", "\n", "This notebook is split into 7 steps:\n", "1. Get Stats Perform data from AWS Data Exchange into an S3 bucket.\n", "2. Join and massage the Stats Perform data so that it can be used to train our models.\n", "3. Setup the hyper parameters for our models.\n", "4. Train our models against 2015 - 2018 data and verify against 2019 data.\n", "5. Get the data from the 2019 season so that we can use it to generate lineups.\n", "6. Compare the size of the full player universe to the upside player universe.\n", "7. Optimize our lineups and see how our models perform against the full universe of players. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Before you attempt to run this notebook, read this\n", "\n", "Thanks for downloading this notebook! The code here will \"just work\" if you do three things:\n", "\n", "1. Point this notebook at your S3 bucket in the cell of this notebook.\n", "2. Run this notebook with an IAM Role that has `AmazonSageMakerFullAccess`, `AWSDataExchangeSubscriberFullAccess`, and `[ \"s3:GetObject\", \"s3:PutObject\", \"s3:DeleteObject\", \"s3:ListBucket\" ]` on whichever S3 bucket you're going to use.\n", "3. An existing subscription to [Stats Perform Fantasy Player Data](https://console.aws.amazon.com/dataexchange/home?region=us-east-1#/products/prodview-tte3yvctdjs7a)." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "# Replace this bucket with your own S3 bucket\n", "bucket = 'lineup-optimizer-demo-226bbd09-5c06-42d7-acc4-ce34b3c0c3e6'" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Install boto3-1.16 and botocore-1.19.15 which have support for `dataexchange` ." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Collecting botocore==1.19.15\n", " Using cached botocore-1.19.15-py2.py3-none-any.whl (6.7 MB)\n", "Installing collected packages: botocore\n", " Attempting uninstall: botocore\n", " Found existing installation: botocore 1.19.15\n", " Uninstalling botocore-1.19.15:\n", " Successfully uninstalled botocore-1.19.15\n", "Successfully installed botocore-1.19.15\n", "\u001b[33mWARNING: You are using pip version 20.0.2; however, version 20.2.4 is available.\n", "You should consider upgrading via the '/home/ec2-user/anaconda3/envs/python3/bin/python -m pip install --upgrade pip' command.\u001b[0m\n", "Collecting boto3==1.16.15\n", " Using cached boto3-1.16.15-py2.py3-none-any.whl (129 kB)\n", "Installing collected packages: boto3\n", " Attempting uninstall: boto3\n", " Found existing installation: boto3 1.16.15\n", " Uninstalling boto3-1.16.15:\n", " Successfully uninstalled boto3-1.16.15\n", "Successfully installed boto3-1.16.15\n", "\u001b[33mWARNING: You are using pip version 20.0.2; however, version 20.2.4 is available.\n", "You should consider upgrading via the '/home/ec2-user/anaconda3/envs/python3/bin/python -m pip install --upgrade pip' command.\u001b[0m\n" ] } ], "source": [ "import sys\n", "!{sys.executable} -m pip install --force --no-deps botocore==1.19.15\n", "!{sys.executable} -m pip install --force --no-deps boto3==1.16.15" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 1 Start: Export Stats Perform data from AWS Data Exchange to your S3 Bucket" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Set the resource IDs that refer to the Stats Perform data we need. If you have a subscription to the [Stats Perform product](https://console.aws.amazon.com/dataexchange/home?region=us-east-1#/products/prodview-tte3yvctdjs7a), you'll be able to export these data sets." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "stats_perform_predictions_data_set_id = 'bbbaa790d1fb4eb1e90ccc5d4e74f774'\n", "stats_perform_salary_data_set_id = '3a0049b13e2018d20898334c7bb0c636'\n", "stats_perform_odds_data_set_id = '3b4fcf7d131ebef8584a8b8088417424'\n", "stats_perform_box_score_data_set_id = '20557179573f3af759e70358077edcb7'\n", "stats_perform_player_reference_data_set_id = '4ebbb5702ae5cd66eebbd58f72210ccf'" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "import boto3\n", "\n", "dx = boto3.client('dataexchange', region_name = 'us-east-1')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Get all of the Assets from the first Revision of each of Stats Perform's data sets." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "predictions_revision_id = dx.list_data_set_revisions(DataSetId = stats_perform_predictions_data_set_id)['Revisions'][0].get('Id')\n", "predictions_assets = dx.list_revision_assets(DataSetId = stats_perform_predictions_data_set_id, RevisionId = predictions_revision_id)['Assets']\n", "\n", "salary_revision_id = dx.list_data_set_revisions(DataSetId = stats_perform_salary_data_set_id)['Revisions'][0].get('Id')\n", "salary_assets = dx.list_revision_assets(DataSetId = stats_perform_salary_data_set_id, RevisionId = salary_revision_id)['Assets']\n", "\n", "odds_revision_id = dx.list_data_set_revisions(DataSetId = stats_perform_odds_data_set_id)['Revisions'][0].get('Id')\n", "odds_assets = dx.list_revision_assets(DataSetId = stats_perform_odds_data_set_id, RevisionId = odds_revision_id)['Assets']\n", "\n", "box_score_revision_id = dx.list_data_set_revisions(DataSetId = stats_perform_box_score_data_set_id)['Revisions'][0].get('Id')\n", "box_score_assets = dx.list_revision_assets(DataSetId = stats_perform_box_score_data_set_id, RevisionId = box_score_revision_id)['Assets']\n", "\n", "player_reference_revision_id = dx.list_data_set_revisions(DataSetId = stats_perform_player_reference_data_set_id)['Revisions'][0].get('Id')\n", "player_reference_assets = dx.list_revision_assets(DataSetId = stats_perform_player_reference_data_set_id, RevisionId = player_reference_revision_id)['Assets']" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A helper function that we'll use to export Assets from AWS Data Exchange into our S3 Bucket." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "import time\n", "\n", "def export_assets(assets):\n", " asset_destinations = []\n", "\n", " for asset in assets:\n", " asset_name_with_out_path = asset.get('Name').split('/')[-1]\n", " \n", " asset_destinations.append({\n", " \"AssetId\": asset.get('Id'),\n", " \"Bucket\": bucket,\n", " \"Key\": 'statsperform/{}'.format(asset_name_with_out_path)\n", " })\n", " \n", " job = dx.create_job(Type = 'EXPORT_ASSETS_TO_S3', Details = {\n", " \"ExportAssetsToS3\": {\n", " \"RevisionId\": assets[0].get(\"RevisionId\"), \"DataSetId\": assets[0].get(\"DataSetId\"),\n", " \"AssetDestinations\": asset_destinations\n", " }\n", " })\n", " \n", " job_id = job.get('Id')\n", " \n", " dx.start_job(JobId = job_id)\n", "\n", " while True:\n", " job = dx.get_job(JobId = job_id)\n", "\n", " if job.get('State') == 'COMPLETED':\n", " break\n", " elif job.get('State') == 'ERROR':\n", " raise Exception(\"Job {} failed to complete - {}\".format(\n", " job_id, job.get('Errors')[0].get('Message'))\n", " )\n", "\n", " time.sleep(1)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "export_assets(predictions_assets)\n", "export_assets(salary_assets)\n", "export_assets(odds_assets)\n", "export_assets(box_score_assets)\n", "export_assets(player_reference_assets)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 1 Complete: Stats Perform data has been exported to your S3 Bucket" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "statsperform/boxScoreByPlayer.csv\n", "statsperform/boxScoreByPlayer.json\n", "statsperform/fantasyPredictions.csv\n", "statsperform/fantasyPredictions.json\n", "statsperform/fantasySalaries.csv\n", "statsperform/fantasySalaries.json\n", "statsperform/odds.csv\n", "statsperform/odds.json\n", "statsperform/players.csv\n", "statsperform/players.json\n" ] } ], "source": [ "import boto3\n", "for file in boto3.resource('s3').Bucket(bucket).objects.all():\n", " print(file.key)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 2 Start: Join Stats Perform data (odds, historical performance, and player reference data) to train the ML model" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "odds_location = 's3://{}/statsperform/odds.json'.format(bucket)\n", "history_location = 's3://{}/statsperform/boxScoreByPlayer.json'.format(bucket)\n", "player_reference_location = 's3://{}/statsperform/players.json'.format(bucket)\n", "salary_location = 's3://{}/statsperform/fantasySalaries.json'.format(bucket)\n", "fantasy_predictions_location = 's3://{}/statsperform/fantasyPredictions.json'.format(bucket)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Create dataframe for training\n", "\n", "Now that we have the Stats Perform data from AWS Data Exchange, we merge the historical data, odds data, and player reference data so that we know how each player did in their games by season and week, including the position they played, and the odds for that game.\n" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "import pandas as pd" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "historical_df = pd.read_json(history_location)" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "odds_df = pd.read_json(odds_location)" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "players_df = pd.read_json(player_reference_location).rename(columns = { 'playerid': 'player_id' })" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "joined_df = pd.merge(historical_df, odds_df, how = 'left', left_on = [ 'week', 'fixture_id' ], right_on = [ 'week', 'fixture_id' ])" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "joined_df = joined_df.set_index([ 'player_id', 'season' ]).join(players_df.set_index([ 'player_id', 'season' ])).reset_index()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Drop rows that are missing data." ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "joined_df = joined_df.dropna(subset = [ \n", " 'player_id', \n", " 'team_id', \n", " 'positionname', \n", " 'season', \n", " 'week', \n", " 'line', \n", " 'favorite_points',\n", " 'favorite_team_id'\n", "]).fillna(0)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Map long position names to short position abbreviations and rename the column. For example, Tight End to TE. This makes it easier to join with predictions and salary data from Stats Perform later on in the notebook." ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "joined_df['positionname'] = joined_df['positionname'].replace([ 'Tight End', 'Running Back', 'Wide Receiver', 'Quarterback' ], [ 'TE', 'RB', 'WR', 'QB' ])" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [], "source": [ "joined_df = joined_df.rename(columns = { 'positionname': 'position' })" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Filter down to offensive players. Note this model doesn't optimize lineups including DSTs but, the Stats Perform data contains individual defensive player and this can be calculated by aggregating defensive box score data." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "joined_df = joined_df[joined_df['position'].isin([ 'TE', 'RB', 'WR', 'QB' ])]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Rename the odds columns to match the terms we used in our slides." ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [], "source": [ "joined_df = joined_df.rename(columns = { 'line': 'ou', 'favorite_points': 'spread' })" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Turn the O/U and Spread into an implied score for the player's team and an implied score for the opponent's team as this is an easy way to represent the odds to our ML models." ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "def implied_game_score(x):\n", " ou = x['ou']\n", " abs_spread = abs(x['spread'])\n", " \n", " favored_score = ((ou - abs_spread) / 2) + abs_spread\n", " underdog_score = ((ou - abs_spread) / 2)\n", " \n", " return favored_score if x['team_id'] == x['favorite_team_id'] else underdog_score" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "def opponent_implied_game_score(x):\n", " ou = x['ou']\n", " abs_spread = abs(x['spread'])\n", " \n", " favored_score = ((ou - abs_spread) / 2) + abs_spread\n", " underdog_score = ((ou - abs_spread) / 2)\n", " \n", " return underdog_score if x['team_id'] == x['favorite_team_id'] else favored_score" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [], "source": [ "joined_df['implied_game_score'] = joined_df.apply(implied_game_score, axis = 1)" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [], "source": [ "joined_df['opponent_implied_game_score'] = joined_df.apply(opponent_implied_game_score, axis = 1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "With all of the data joined in a single dataframe, filter down to the columns we need:\n", "* `player_id`: So we can uniquely identify each player.\n", "* `season`: This splits the data up by year. This way we can run our trained model against one of the years in this dataframe so we can see how the model performed.\n", "* `week`: Tells which week the game was played.\n", "* `fixture_id`: Uniquely IDs each game which is useful later on when we join against Predictions data from Stats Perform.\n", "* `position`: The position played.\n", "* `rush_yds`: The yards rushed by the played. Used for scoring.\n", "* `rush_tds`: The number of rushing touchdowns. Used for scoring.\n", "* `pass_yds`: The yards thrown by the QB. Used for scoring.\n", "* `pass_tds`: The touchdowns thrown by the QB. Used for scoring.\n", "* `recs`: The number of catches made by the player. Used for scoring.\n", "* `rec_yds`: The number of receiving yards. Used for scoring.\n", "* `rec_tds`: The numer of receiving touchdowns. Used for scoring.\n", "* `pass_atts`: The number of times the QB passed the ball. We use this field when training our model to filter out QBs that aren't starters and as a feature when predicting upside from odds.\n", "* `rush_atts`: The number of times the player had a chance to rush with the football. We use this field when training our model to filter out RBs that don't get a lot touches as a feature when predicting upside from odds.\n", "* `rec_targets`: The number of times the player was targeted for a catch. We use this field when training our model to filter out WRs and TEs that don't get a lot of targets and as a feature when predicting upside from odds.\n", "* `implied_game_score`: Total points the odds predict the player's team will score. Used as a feature for our ML models.\n", "* `opponent_implied_game_score`: Total points the odds predict the player's opponent will score. Used as a feature for our ML models." ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [], "source": [ "scoring_df = joined_df.reset_index()[[\n", " 'player_id',\n", " 'team_id', \n", " 'season', \n", " 'week',\n", " 'fixture_id',\n", " 'position',\n", " 'rush_yds', \n", " 'rush_tds',\n", " 'pass_yds', \n", " 'pass_tds',\n", " 'recs', \n", " 'rec_yds', \n", " 'rec_tds',\n", " 'pass_atts',\n", " 'rush_atts',\n", " 'rec_targets',\n", " 'implied_game_score',\n", " 'opponent_implied_game_score'\n", "]]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Score player points\n", "\n", "We use a standard PPR (point per reception) scoring function.\n", "* Each passing yard gives the player 0.04 points (`pass_yds`).\n", "* Each passing touchdown gives the player 4 points (`pass_tds`).\n", "* Each rushing yard gives the player 0.1 points (`rush_yds`).\n", "* Each rushing touchdown gives the player 6 popints (`rush_tds`).\n", "* Each receiving yard gives the player 0.1 points (`rec_yds`).\n", "* Each reception gives the player a point (`recs`).\n", "* Each receiving touchdown gives the player 6 points (`rec_tds`)." ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [], "source": [ "def points_scored(player):\n", " total = 0.0\n", " \n", " total += player['pass_yds'] * 0.04\n", " total += player['pass_tds'] * 4\n", " total += player['rush_yds'] * 0.1\n", " total += player['rush_tds'] * 6\n", " total += player['recs']\n", " total += player['rec_yds'] * 0.1\n", " total += player['rec_tds'] * 6\n", " \n", " return total" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [], "source": [ "scoring_df['actual_points'] = scoring_df.apply(points_scored, axis = 1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 2 Complete: Your data is ready to train the ML model" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
player_idteam_idseasonweekfixture_idpositionrush_ydsrush_tdspass_ydspass_tdsrecsrec_ydsrec_tdspass_attsrush_attsrec_targetsimplied_game_scoreopponent_implied_game_scoreactual_points
144845467423562018102040922RB0.00.00.00.02.017.00.00.00.02.027.2523.753.7
171825632933642019172142153TE0.00.00.00.00.00.00.00.00.00.015.7529.250.0
257006960803232017141744730RB32.00.00.00.00.00.00.00.09.00.027.0024.503.2
222156094963262019112142174WR0.00.00.00.04.015.00.00.00.06.017.5022.505.5
18981598989366201741744870RB7.00.00.00.06.037.00.00.02.06.019.2522.7510.4
16523557954324201842040939WR0.00.00.00.02.08.00.00.00.03.017.5026.502.8
157975562943482019122142245RB5.00.00.00.01.0-6.00.00.02.03.025.0019.500.9
25218694041347201992142116WR12.00.00.00.01.04.00.00.01.05.025.5020.002.6
282737423873272019102142056WR0.00.00.00.06.062.00.00.00.08.016.7527.2512.2
7556402162326201651635697TE0.00.00.00.00.00.00.00.00.00.021.7525.750.0
\n", "
" ], "text/plain": [ " player_id team_id season week fixture_id position rush_yds \\\n", "14484 546742 356 2018 10 2040922 RB 0.0 \n", "17182 563293 364 2019 17 2142153 TE 0.0 \n", "25700 696080 323 2017 14 1744730 RB 32.0 \n", "22215 609496 326 2019 11 2142174 WR 0.0 \n", "18981 598989 366 2017 4 1744870 RB 7.0 \n", "16523 557954 324 2018 4 2040939 WR 0.0 \n", "15797 556294 348 2019 12 2142245 RB 5.0 \n", "25218 694041 347 2019 9 2142116 WR 12.0 \n", "28273 742387 327 2019 10 2142056 WR 0.0 \n", "7556 402162 326 2016 5 1635697 TE 0.0 \n", "\n", " rush_tds pass_yds pass_tds recs rec_yds rec_tds pass_atts \\\n", "14484 0.0 0.0 0.0 2.0 17.0 0.0 0.0 \n", "17182 0.0 0.0 0.0 0.0 0.0 0.0 0.0 \n", "25700 0.0 0.0 0.0 0.0 0.0 0.0 0.0 \n", "22215 0.0 0.0 0.0 4.0 15.0 0.0 0.0 \n", "18981 0.0 0.0 0.0 6.0 37.0 0.0 0.0 \n", "16523 0.0 0.0 0.0 2.0 8.0 0.0 0.0 \n", "15797 0.0 0.0 0.0 1.0 -6.0 0.0 0.0 \n", "25218 0.0 0.0 0.0 1.0 4.0 0.0 0.0 \n", "28273 0.0 0.0 0.0 6.0 62.0 0.0 0.0 \n", "7556 0.0 0.0 0.0 0.0 0.0 0.0 0.0 \n", "\n", " rush_atts rec_targets implied_game_score \\\n", "14484 0.0 2.0 27.25 \n", "17182 0.0 0.0 15.75 \n", "25700 9.0 0.0 27.00 \n", "22215 0.0 6.0 17.50 \n", "18981 2.0 6.0 19.25 \n", "16523 0.0 3.0 17.50 \n", "15797 2.0 3.0 25.00 \n", "25218 1.0 5.0 25.50 \n", "28273 0.0 8.0 16.75 \n", "7556 0.0 0.0 21.75 \n", "\n", " opponent_implied_game_score actual_points \n", "14484 23.75 3.7 \n", "17182 29.25 0.0 \n", "25700 24.50 3.2 \n", "22215 22.50 5.5 \n", "18981 22.75 10.4 \n", "16523 26.50 2.8 \n", "15797 19.50 0.9 \n", "25218 20.00 2.6 \n", "28273 27.25 12.2 \n", "7556 25.75 0.0 " ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "scoring_df.sample(n = 10)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 3 Start: Setup to train your models" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Describe the models\n", "\n", "Here we create an object that we reuse throughout the remainder of this notebook to tell us the features used as inputs for each player by position, along with qualifiers for inclusion in the model, and inference later on when generating lineups.\n", "\n", "For all positions, we use `implied_game_score` and `opponent_implied_game_score` as features to the ML model. As discussed during the presentation, we're using these features to predict if a player is likely to score points.\n", "\n", "We filter out QBs that have fewer than 10 `pass_atts`. We also use `pass_atts` and `rush_atts` as features.\n", "\n", "We filter our WRs that have fewer than 5 `rec_targets`. We also use `rec_targets` as a feature.\n", "\n", "We filter out RBs that have fewer than 8 `rush_atts`. We also use `rec_targets` and `rush_atts` as features.\n", "\n", "We filter out TEs that have fewer than 3 `rec_targets`. We also use `rec_targets` as a feature.\n", "\n", "For all positions, we label the player for consideration in our lineups if they're in the upper quantile (per position) of scoring for all of the data points we have over the 2015-2018 seasons." ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [], "source": [ "predictions = {\n", " 'QB': { \n", " 'features': [\n", " 'implied_game_score', \n", " 'opponent_implied_game_score',\n", " 'rush_atts',\n", " 'pass_atts'\n", " ], \n", " 'start_quantile': .5, \n", " 'qualifier_field': 'pass_atts', \n", " 'qualifier_value': 10 \n", " },\n", " 'WR': { \n", " 'features': [\n", " 'implied_game_score', \n", " 'opponent_implied_game_score',\n", " 'rec_targets'\n", " ], \n", " 'start_quantile': .3, \n", " 'qualifier_field': 'rec_targets', \n", " 'qualifier_value': 5 \n", " },\n", " 'RB': { \n", " 'features': [\n", " 'implied_game_score', \n", " 'opponent_implied_game_score',\n", " 'rush_atts',\n", " 'rec_targets'\n", " ], \n", " 'start_quantile': .4, \n", " 'qualifier_field': 'rush_atts', \n", " 'qualifier_value': 8 \n", " },\n", " 'TE': { \n", " 'features': [\n", " 'implied_game_score', \n", " 'opponent_implied_game_score',\n", " 'rec_targets'\n", " ], \n", " 'start_quantile': .5, \n", " 'qualifier_field': 'rec_targets', \n", " 'qualifier_value': 3 \n", " }\n", "}" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We continue to fill out the object created in the previous step.\n", "\n", "1. We remove players that don't qualify based on the qualifiers above.\n", "2. We get the points scored for the quantile specified above.\n", "3. We label the player as someone that we should start in our lineup if they're above our quantile cutoff.\n", "4. We convert all fields to float32 as required by Amazon SageMaker.\n", "5. Finally, we set aside data from the 2019 season for verification later on to calculate Recall, Precision, and the F1 Score.\n", "\n", "Note, we use 40% and above for RBs and 30% and above for WRs otherwise the model aggressively labels players as not having upside from the odds, leaving us with too few players later on to fill a lineup." ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [], "source": [ "for position in predictions:\n", " features = predictions[position]['features']\n", "\n", " fields = features[:]\n", " fields.append('actual_points')\n", " fields.append('season')\n", "\n", " qualifier_field = predictions[position]['qualifier_field']\n", " qualifier_value = predictions[position]['qualifier_value']\n", "\n", " predictions[position]['predictions_df'] = scoring_df[\n", " (scoring_df['position'] == position) &\n", " (scoring_df[qualifier_field] > qualifier_value)\n", " ][fields]\n", "\n", " start_quantile = predictions[position]['start_quantile']\n", " predictions[position]['minimum_points_wanted'] = predictions[position]['predictions_df'][['actual_points']].quantile([start_quantile]).values[0][0]\n", "\n", " predictions[position]['predictions_df']['has_upside'] = predictions[position]['predictions_df'].apply(\n", " lambda x: 1 if x['actual_points'] >= predictions[position]['minimum_points_wanted'] else 0, axis=1\n", " )\n", "\n", " predictions[position]['predictions_df'] = predictions[position]['predictions_df'].astype('float32')\n", "\n", " training_data_df = predictions[position]['predictions_df'][\n", " predictions[position]['predictions_df']['season'] != 2019\n", " ]\n", " \n", " verification_data_df = predictions[position]['predictions_df'][\n", " predictions[position]['predictions_df']['season'] == 2019\n", " ]\n", " \n", " predictions[position]['training_vectors'] = training_data_df[features].values\n", " predictions[position]['training_labels'] = training_data_df['has_upside'].values\n", " \n", " predictions[position]['verification_vectors'] = verification_data_df[features].values\n", " predictions[position]['verification_labels'] = verification_data_df['has_upside'].values" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 3 Complete: You're ready to train the models\n", "\n", "Print out the minimum points we want the models to target for each position. For example, if the model says a QB has upside, then it means the model believes they'll score more then 17.32 points." ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Position: QB, start if above: 17.32\n", "Position: WR, start if above: 9.8\n", "Position: RB, start if above: 10.84\n", "Position: TE, start if above: 9.6\n" ] } ], "source": [ "for position in predictions:\n", " print('Position: {}, start if above: {}'.format(position, round(predictions[position]['minimum_points_wanted'], 2)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 4 Start: Train your model" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Train our models\n", "\n", "The step may look a little intimidating but, it's mostly boiler plate. For each position, we're going to create a model and stand up an endpoint that we can use for inference when we generate our lineups. We run these steps for each position:\n", "\n", "1. Take our training data, pack it into a buffer, and upload it to S3 so our ML model can train on it.\n", "2. Our goal is to classify players as upside or no upside based on odds so, we create a [linear learner](https://docs.aws.amazon.com/sagemaker/latest/dg/linear-learner.html) which is useful for classification problems.\n", "3. Train the model using the features for that position.\n", "4. Deploy an endpoint that we can use for inference." ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "2020-11-16 00:07:00 Starting - Starting the training job...\n", "2020-11-16 00:07:03 Starting - Launching requested ML instances.........\n", "2020-11-16 00:08:35 Starting - Preparing the instances for training......\n", "2020-11-16 00:10:02 Downloading - Downloading input data...\n", "2020-11-16 00:10:23 Training - Downloading the training image..\u001b[34mDocker entrypoint called with argument(s): train\u001b[0m\n", "\u001b[34mRunning default environment configuration script\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Reading default configuration from /opt/amazon/lib/python2.7/site-packages/algorithm/resources/default-input.json: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'auto', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'1000', u'huber_delta': u'1.0', u'num_classes': u'1', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Merging with provided configuration from /opt/ml/input/config/hyperparameters.json: {u'feature_dim': u'4', u'mini_batch_size': u'200', u'predictor_type': u'binary_classifier'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Final configuration: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'4', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'200', u'huber_delta': u'1.0', u'num_classes': u'1', u'predictor_type': u'binary_classifier', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 WARNING 140334821308224] Loggers have already been setup.\u001b[0m\n", "\u001b[34mProcess 1 is a worker.\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Using default worker.\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Checkpoint loading and saving are disabled.\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:43.570] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 0, \"duration\": 15, \"num_examples\": 1, \"num_bytes\": 12000}\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Create Store: local\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:43.634] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 1, \"duration\": 63, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Scaler algorithm parameters\n", " \u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Scaling model computed with parameters:\n", " {'stdev_weight': \u001b[0m\n", "\u001b[34m[3.616295 3.7118077 2.3965478 9.128197 ]\u001b[0m\n", "\u001b[34m, 'stdev_label': None, 'mean_label': None, 'mean_weight': \u001b[0m\n", "\u001b[34m[22.742779 22.671947 2.8649998 33.84389 ]\u001b[0m\n", "\u001b[34m}\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] nvidia-smi took: 0.0251190662384 secs to identify 0 gpus\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:43 INFO 140334821308224] Number of GPUs being used: 0\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Total Batches Seen\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Total Records Seen\": {\"count\": 1, \"max\": 2119, \"sum\": 2119.0, \"min\": 2119}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Reset Count\": {\"count\": 1, \"max\": 2, \"sum\": 2.0, \"min\": 2}}, \"EndTime\": 1605485443.731294, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"init_train_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1605485443.731257}\n", "\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:44.023] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 4, \"duration\": 291, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6784317949083116, \"sum\": 0.6784317949083116, \"min\": 0.6784317949083116}}, \"EndTime\": 1605485444.023344, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023249}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6823264736599393, \"sum\": 0.6823264736599393, \"min\": 0.6823264736599393}}, \"EndTime\": 1605485444.023436, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023416}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6899109819200304, \"sum\": 0.6899109819200304, \"min\": 0.6899109819200304}}, \"EndTime\": 1605485444.023507, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023487}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6850500657823351, \"sum\": 0.6850500657823351, \"min\": 0.6850500657823351}}, \"EndTime\": 1605485444.023575, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023556}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6601708009507921, \"sum\": 0.6601708009507921, \"min\": 0.6601708009507921}}, \"EndTime\": 1605485444.023647, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.02363}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6601351081000434, \"sum\": 0.6601351081000434, \"min\": 0.6601351081000434}}, \"EndTime\": 1605485444.023718, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023706}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6607632149590387, \"sum\": 0.6607632149590387, \"min\": 0.6607632149590387}}, \"EndTime\": 1605485444.023774, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023757}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6599577458699545, \"sum\": 0.6599577458699545, \"min\": 0.6599577458699545}}, \"EndTime\": 1605485444.023833, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023821}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.695781504313151, \"sum\": 0.695781504313151, \"min\": 0.695781504313151}}, \"EndTime\": 1605485444.023894, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023876}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6845144314236111, \"sum\": 0.6845144314236111, \"min\": 0.6845144314236111}}, \"EndTime\": 1605485444.023961, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.023944}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6795155334472657, \"sum\": 0.6795155334472657, \"min\": 0.6795155334472657}}, \"EndTime\": 1605485444.024034, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024007}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6901559448242187, \"sum\": 0.6901559448242187, \"min\": 0.6901559448242187}}, \"EndTime\": 1605485444.024095, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024077}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.658349346584744, \"sum\": 0.658349346584744, \"min\": 0.658349346584744}}, \"EndTime\": 1605485444.024158, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024141}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6626794815063477, \"sum\": 0.6626794815063477, \"min\": 0.6626794815063477}}, \"EndTime\": 1605485444.024219, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024201}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6577384482489692, \"sum\": 0.6577384482489692, \"min\": 0.6577384482489692}}, \"EndTime\": 1605485444.024267, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024256}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6524687279595269, \"sum\": 0.6524687279595269, \"min\": 0.6524687279595269}}, \"EndTime\": 1605485444.024321, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024306}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6984898885091145, \"sum\": 0.6984898885091145, \"min\": 0.6984898885091145}}, \"EndTime\": 1605485444.024376, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024359}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.685986565483941, \"sum\": 0.685986565483941, \"min\": 0.685986565483941}}, \"EndTime\": 1605485444.024427, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024411}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6740711042616102, \"sum\": 0.6740711042616102, \"min\": 0.6740711042616102}}, \"EndTime\": 1605485444.024491, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024474}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6719803704155816, \"sum\": 0.6719803704155816, \"min\": 0.6719803704155816}}, \"EndTime\": 1605485444.02456, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024542}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6918115743001302, \"sum\": 0.6918115743001302, \"min\": 0.6918115743001302}}, \"EndTime\": 1605485444.024603, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024588}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6934740405612522, \"sum\": 0.6934740405612522, \"min\": 0.6934740405612522}}, \"EndTime\": 1605485444.024654, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024638}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6919698121812609, \"sum\": 0.6919698121812609, \"min\": 0.6919698121812609}}, \"EndTime\": 1605485444.024726, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024708}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6934444173177083, \"sum\": 0.6934444173177083, \"min\": 0.6934444173177083}}, \"EndTime\": 1605485444.024791, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024774}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6992276085747613, \"sum\": 0.6992276085747613, \"min\": 0.6992276085747613}}, \"EndTime\": 1605485444.024855, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024838}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7060904947916666, \"sum\": 0.7060904947916666, \"min\": 0.7060904947916666}}, \"EndTime\": 1605485444.024901, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024891}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6942655266655816, \"sum\": 0.6942655266655816, \"min\": 0.6942655266655816}}, \"EndTime\": 1605485444.024956, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.02494}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6882593621148003, \"sum\": 0.6882593621148003, \"min\": 0.6882593621148003}}, \"EndTime\": 1605485444.025006, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.024996}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7231230926513672, \"sum\": 0.7231230926513672, \"min\": 0.7231230926513672}}, \"EndTime\": 1605485444.025061, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.025045}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7388735877143012, \"sum\": 0.7388735877143012, \"min\": 0.7388735877143012}}, \"EndTime\": 1605485444.025118, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.025103}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7223830922444662, \"sum\": 0.7223830922444662, \"min\": 0.7223830922444662}}, \"EndTime\": 1605485444.025175, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.025159}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7238291337754992, \"sum\": 0.7238291337754992, \"min\": 0.7238291337754992}}, \"EndTime\": 1605485444.025246, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485444.025227}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #quality_metric: host=algo-1, epoch=0, train binary_classification_cross_entropy_objective =0.678431794908\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #early_stopping_criteria_metric: host=algo-1, epoch=0, criteria=binary_classification_cross_entropy_objective, value=0.65246872796\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Epoch 0: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Saving model for epoch: 0\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Saved checkpoint to \"/tmp/tmpmtS0PR/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #progress_metric: host=algo-1, completed 6 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Total Batches Seen\": {\"count\": 1, \"max\": 21, \"sum\": 21.0, \"min\": 21}, \"Total Records Seen\": {\"count\": 1, \"max\": 4038, \"sum\": 4038.0, \"min\": 4038}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Reset Count\": {\"count\": 1, \"max\": 3, \"sum\": 3.0, \"min\": 3}}, \"EndTime\": 1605485444.039583, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605485443.731542}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #throughput_metric: host=algo-1, train throughput=6227.16273514 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:44.293] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 6, \"duration\": 253, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6638083140055339, \"sum\": 0.6638083140055339, \"min\": 0.6638083140055339}}, \"EndTime\": 1605485444.293631, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.293541}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6671506585015191, \"sum\": 0.6671506585015191, \"min\": 0.6671506585015191}}, \"EndTime\": 1605485444.293714, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.293695}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.674196523030599, \"sum\": 0.674196523030599, \"min\": 0.674196523030599}}, \"EndTime\": 1605485444.293786, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.293767}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6698526424831814, \"sum\": 0.6698526424831814, \"min\": 0.6698526424831814}}, \"EndTime\": 1605485444.293866, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.293845}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6212687047322591, \"sum\": 0.6212687047322591, \"min\": 0.6212687047322591}}, \"EndTime\": 1605485444.293941, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.293921}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6116782082451715, \"sum\": 0.6116782082451715, \"min\": 0.6116782082451715}}, \"EndTime\": 1605485444.294015, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.293996}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.611635390387641, \"sum\": 0.611635390387641, \"min\": 0.611635390387641}}, \"EndTime\": 1605485444.294086, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294067}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.611515358818902, \"sum\": 0.611515358818902, \"min\": 0.611515358818902}}, \"EndTime\": 1605485444.294152, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294134}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6796170383029514, \"sum\": 0.6796170383029514, \"min\": 0.6796170383029514}}, \"EndTime\": 1605485444.294213, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294197}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6698052215576172, \"sum\": 0.6698052215576172, \"min\": 0.6698052215576172}}, \"EndTime\": 1605485444.294279, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.29426}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6647329627143013, \"sum\": 0.6647329627143013, \"min\": 0.6647329627143013}}, \"EndTime\": 1605485444.294352, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294333}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6741779412163629, \"sum\": 0.6741779412163629, \"min\": 0.6741779412163629}}, \"EndTime\": 1605485444.294425, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294409}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6056050025092231, \"sum\": 0.6056050025092231, \"min\": 0.6056050025092231}}, \"EndTime\": 1605485444.294486, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.29447}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6213341267903646, \"sum\": 0.6213341267903646, \"min\": 0.6213341267903646}}, \"EndTime\": 1605485444.29455, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294532}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6086685307820638, \"sum\": 0.6086685307820638, \"min\": 0.6086685307820638}}, \"EndTime\": 1605485444.294613, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294596}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6069440417819553, \"sum\": 0.6069440417819553, \"min\": 0.6069440417819553}}, \"EndTime\": 1605485444.294673, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294656}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6825121646457248, \"sum\": 0.6825121646457248, \"min\": 0.6825121646457248}}, \"EndTime\": 1605485444.294735, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294718}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6735102420383029, \"sum\": 0.6735102420383029, \"min\": 0.6735102420383029}}, \"EndTime\": 1605485444.294795, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294778}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6636746724446615, \"sum\": 0.6636746724446615, \"min\": 0.6636746724446615}}, \"EndTime\": 1605485444.294858, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.29484}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6627296193440755, \"sum\": 0.6627296193440755, \"min\": 0.6627296193440755}}, \"EndTime\": 1605485444.294918, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294902}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6682962714301215, \"sum\": 0.6682962714301215, \"min\": 0.6682962714301215}}, \"EndTime\": 1605485444.294979, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.294962}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6731328074137369, \"sum\": 0.6731328074137369, \"min\": 0.6731328074137369}}, \"EndTime\": 1605485444.29504, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295023}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6668603515625, \"sum\": 0.6668603515625, \"min\": 0.6668603515625}}, \"EndTime\": 1605485444.295104, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295087}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.666353374057346, \"sum\": 0.666353374057346, \"min\": 0.666353374057346}}, \"EndTime\": 1605485444.295181, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295163}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6919816674126519, \"sum\": 0.6919816674126519, \"min\": 0.6919816674126519}}, \"EndTime\": 1605485444.295234, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295218}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6932007175021702, \"sum\": 0.6932007175021702, \"min\": 0.6932007175021702}}, \"EndTime\": 1605485444.295294, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295276}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6920465172661675, \"sum\": 0.6920465172661675, \"min\": 0.6920465172661675}}, \"EndTime\": 1605485444.295352, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295335}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6925116729736328, \"sum\": 0.6925116729736328, \"min\": 0.6925116729736328}}, \"EndTime\": 1605485444.295416, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295397}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6969509718153212, \"sum\": 0.6969509718153212, \"min\": 0.6969509718153212}}, \"EndTime\": 1605485444.295486, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295467}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7294702657063802, \"sum\": 0.7294702657063802, \"min\": 0.7294702657063802}}, \"EndTime\": 1605485444.295558, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295539}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6957673899332683, \"sum\": 0.6957673899332683, \"min\": 0.6957673899332683}}, \"EndTime\": 1605485444.295627, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.295608}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6967657131618924, \"sum\": 0.6967657131618924, \"min\": 0.6967657131618924}}, \"EndTime\": 1605485444.295698, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.29568}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #quality_metric: host=algo-1, epoch=1, train binary_classification_cross_entropy_objective =0.663808314006\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #early_stopping_criteria_metric: host=algo-1, epoch=1, criteria=binary_classification_cross_entropy_objective, value=0.605605002509\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Epoch 1: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Saving model for epoch: 1\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Saved checkpoint to \"/tmp/tmpDVmxFU/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #progress_metric: host=algo-1, completed 13 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Total Batches Seen\": {\"count\": 1, \"max\": 31, \"sum\": 31.0, \"min\": 31}, \"Total Records Seen\": {\"count\": 1, \"max\": 5957, \"sum\": 5957.0, \"min\": 5957}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Reset Count\": {\"count\": 1, \"max\": 4, \"sum\": 4.0, \"min\": 4}}, \"EndTime\": 1605485444.302656, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605485444.039877}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #throughput_metric: host=algo-1, train throughput=7299.3568191 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:44.657] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 8, \"duration\": 354, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6531811565823025, \"sum\": 0.6531811565823025, \"min\": 0.6531811565823025}}, \"EndTime\": 1605485444.657169, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657088}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6561436928643121, \"sum\": 0.6561436928643121, \"min\": 0.6561436928643121}}, \"EndTime\": 1605485444.657255, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657241}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6626490698920355, \"sum\": 0.6626490698920355, \"min\": 0.6626490698920355}}, \"EndTime\": 1605485444.657318, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657301}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6583168623182508, \"sum\": 0.6583168623182508, \"min\": 0.6583168623182508}}, \"EndTime\": 1605485444.657384, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657365}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6234595531887478, \"sum\": 0.6234595531887478, \"min\": 0.6234595531887478}}, \"EndTime\": 1605485444.657476, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657459}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6020988167656792, \"sum\": 0.6020988167656792, \"min\": 0.6020988167656792}}, \"EndTime\": 1605485444.657531, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657516}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.602011227077908, \"sum\": 0.602011227077908, \"min\": 0.602011227077908}}, \"EndTime\": 1605485444.657585, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657569}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6017567613389757, \"sum\": 0.6017567613389757, \"min\": 0.6017567613389757}}, \"EndTime\": 1605485444.657653, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657634}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6673575931125217, \"sum\": 0.6673575931125217, \"min\": 0.6673575931125217}}, \"EndTime\": 1605485444.65773, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657709}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6581902821858724, \"sum\": 0.6581902821858724, \"min\": 0.6581902821858724}}, \"EndTime\": 1605485444.657798, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.65778}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6539380009969076, \"sum\": 0.6539380009969076, \"min\": 0.6539380009969076}}, \"EndTime\": 1605485444.657864, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657846}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6625129699707031, \"sum\": 0.6625129699707031, \"min\": 0.6625129699707031}}, \"EndTime\": 1605485444.657935, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657908}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5976203325059679, \"sum\": 0.5976203325059679, \"min\": 0.5976203325059679}}, \"EndTime\": 1605485444.657999, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.657981}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6205638292100695, \"sum\": 0.6205638292100695, \"min\": 0.6205638292100695}}, \"EndTime\": 1605485444.658063, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658045}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.598615345425076, \"sum\": 0.598615345425076, \"min\": 0.598615345425076}}, \"EndTime\": 1605485444.658125, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658107}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5973096381293402, \"sum\": 0.5973096381293402, \"min\": 0.5973096381293402}}, \"EndTime\": 1605485444.658187, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.65817}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6720979817708334, \"sum\": 0.6720979817708334, \"min\": 0.6720979817708334}}, \"EndTime\": 1605485444.658251, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658233}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6678617604573568, \"sum\": 0.6678617604573568, \"min\": 0.6678617604573568}}, \"EndTime\": 1605485444.658313, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658296}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6599347517225478, \"sum\": 0.6599347517225478, \"min\": 0.6599347517225478}}, \"EndTime\": 1605485444.658376, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658359}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6600136650933159, \"sum\": 0.6600136650933159, \"min\": 0.6600136650933159}}, \"EndTime\": 1605485444.658439, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658421}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6689459482828776, \"sum\": 0.6689459482828776, \"min\": 0.6689459482828776}}, \"EndTime\": 1605485444.6585, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658484}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6666383531358507, \"sum\": 0.6666383531358507, \"min\": 0.6666383531358507}}, \"EndTime\": 1605485444.658561, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658545}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6707583448621962, \"sum\": 0.6707583448621962, \"min\": 0.6707583448621962}}, \"EndTime\": 1605485444.658633, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658614}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.669315677218967, \"sum\": 0.669315677218967, \"min\": 0.669315677218967}}, \"EndTime\": 1605485444.658701, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658684}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6911116027832032, \"sum\": 0.6911116027832032, \"min\": 0.6911116027832032}}, \"EndTime\": 1605485444.658759, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658745}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6896459706624349, \"sum\": 0.6896459706624349, \"min\": 0.6896459706624349}}, \"EndTime\": 1605485444.658826, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658808}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6929984622531467, \"sum\": 0.6929984622531467, \"min\": 0.6929984622531467}}, \"EndTime\": 1605485444.658898, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.65888}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6939664374457465, \"sum\": 0.6939664374457465, \"min\": 0.6939664374457465}}, \"EndTime\": 1605485444.658961, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.658944}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.701903813680013, \"sum\": 0.701903813680013, \"min\": 0.701903813680013}}, \"EndTime\": 1605485444.659011, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.659001}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6952207862006293, \"sum\": 0.6952207862006293, \"min\": 0.6952207862006293}}, \"EndTime\": 1605485444.659041, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.659033}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.702326897515191, \"sum\": 0.702326897515191, \"min\": 0.702326897515191}}, \"EndTime\": 1605485444.659094, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.659078}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7024954393174914, \"sum\": 0.7024954393174914, \"min\": 0.7024954393174914}}, \"EndTime\": 1605485444.659163, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.659145}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #quality_metric: host=algo-1, epoch=2, train binary_classification_cross_entropy_objective =0.653181156582\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #early_stopping_criteria_metric: host=algo-1, epoch=2, criteria=binary_classification_cross_entropy_objective, value=0.597309638129\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Epoch 2: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Saving model for epoch: 2\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] Saved checkpoint to \"/tmp/tmpj4BEDa/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #progress_metric: host=algo-1, completed 20 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Total Batches Seen\": {\"count\": 1, \"max\": 41, \"sum\": 41.0, \"min\": 41}, \"Total Records Seen\": {\"count\": 1, \"max\": 7876, \"sum\": 7876.0, \"min\": 7876}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Reset Count\": {\"count\": 1, \"max\": 5, \"sum\": 5.0, \"min\": 5}}, \"EndTime\": 1605485444.666705, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605485444.30291}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:44 INFO 140334821308224] #throughput_metric: host=algo-1, train throughput=5273.26516391 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:45.007] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 10, \"duration\": 340, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6435852898491754, \"sum\": 0.6435852898491754, \"min\": 0.6435852898491754}}, \"EndTime\": 1605485445.007411, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.007315}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6461045032077365, \"sum\": 0.6461045032077365, \"min\": 0.6461045032077365}}, \"EndTime\": 1605485445.007511, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.00749}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6521672821044922, \"sum\": 0.6521672821044922, \"min\": 0.6521672821044922}}, \"EndTime\": 1605485445.007585, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.007564}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6478642188178169, \"sum\": 0.6478642188178169, \"min\": 0.6478642188178169}}, \"EndTime\": 1605485445.007665, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.007643}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6005900022718641, \"sum\": 0.6005900022718641, \"min\": 0.6005900022718641}}, \"EndTime\": 1605485445.007736, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.007717}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.601122203403049, \"sum\": 0.601122203403049, \"min\": 0.601122203403049}}, \"EndTime\": 1605485445.007815, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.007794}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6004152891370985, \"sum\": 0.6004152891370985, \"min\": 0.6004152891370985}}, \"EndTime\": 1605485445.007879, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.007862}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6005581198798285, \"sum\": 0.6005581198798285, \"min\": 0.6005581198798285}}, \"EndTime\": 1605485445.007953, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.007923}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6563023291693794, \"sum\": 0.6563023291693794, \"min\": 0.6563023291693794}}, \"EndTime\": 1605485445.008015, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.007998}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6477538172403972, \"sum\": 0.6477538172403972, \"min\": 0.6477538172403972}}, \"EndTime\": 1605485445.008078, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008061}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6442941580878364, \"sum\": 0.6442941580878364, \"min\": 0.6442941580878364}}, \"EndTime\": 1605485445.008134, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008119}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6518721856011285, \"sum\": 0.6518721856011285, \"min\": 0.6518721856011285}}, \"EndTime\": 1605485445.008191, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008174}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.599718746609158, \"sum\": 0.599718746609158, \"min\": 0.599718746609158}}, \"EndTime\": 1605485445.008252, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008235}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6018477461073134, \"sum\": 0.6018477461073134, \"min\": 0.6018477461073134}}, \"EndTime\": 1605485445.008314, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008298}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6018326229519314, \"sum\": 0.6018326229519314, \"min\": 0.6018326229519314}}, \"EndTime\": 1605485445.008376, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008359}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6013473680284288, \"sum\": 0.6013473680284288, \"min\": 0.6013473680284288}}, \"EndTime\": 1605485445.008436, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008419}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6650596110026041, \"sum\": 0.6650596110026041, \"min\": 0.6650596110026041}}, \"EndTime\": 1605485445.008496, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.00848}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6645776621500651, \"sum\": 0.6645776621500651, \"min\": 0.6645776621500651}}, \"EndTime\": 1605485445.008558, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.00854}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6593319786919488, \"sum\": 0.6593319786919488, \"min\": 0.6593319786919488}}, \"EndTime\": 1605485445.008625, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008609}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6598762003580729, \"sum\": 0.6598762003580729, \"min\": 0.6598762003580729}}, \"EndTime\": 1605485445.008691, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008666}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6700510067409939, \"sum\": 0.6700510067409939, \"min\": 0.6700510067409939}}, \"EndTime\": 1605485445.008754, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008736}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6731247287326388, \"sum\": 0.6731247287326388, \"min\": 0.6731247287326388}}, \"EndTime\": 1605485445.00882, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008802}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6687855614556206, \"sum\": 0.6687855614556206, \"min\": 0.6687855614556206}}, \"EndTime\": 1605485445.008884, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008866}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6705822414822049, \"sum\": 0.6705822414822049, \"min\": 0.6705822414822049}}, \"EndTime\": 1605485445.008944, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.008928}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6924549357096355, \"sum\": 0.6924549357096355, \"min\": 0.6924549357096355}}, \"EndTime\": 1605485445.009006, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.00899}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6913980441623264, \"sum\": 0.6913980441623264, \"min\": 0.6913980441623264}}, \"EndTime\": 1605485445.00907, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.009052}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6927113172743056, \"sum\": 0.6927113172743056, \"min\": 0.6927113172743056}}, \"EndTime\": 1605485445.00913, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.009114}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6938032531738281, \"sum\": 0.6938032531738281, \"min\": 0.6938032531738281}}, \"EndTime\": 1605485445.009189, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.009172}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7028279368082683, \"sum\": 0.7028279368082683, \"min\": 0.7028279368082683}}, \"EndTime\": 1605485445.009253, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.009236}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6976564788818359, \"sum\": 0.6976564788818359, \"min\": 0.6976564788818359}}, \"EndTime\": 1605485445.009317, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.009299}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7036228434244791, \"sum\": 0.7036228434244791, \"min\": 0.7036228434244791}}, \"EndTime\": 1605485445.009382, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.009365}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.703307351006402, \"sum\": 0.703307351006402, \"min\": 0.703307351006402}}, \"EndTime\": 1605485445.009481, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485445.00946}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #quality_metric: host=algo-1, epoch=3, train binary_classification_cross_entropy_objective =0.643585289849\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #early_stopping_criteria_metric: host=algo-1, epoch=3, criteria=binary_classification_cross_entropy_objective, value=0.599718746609\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] Saving model for epoch: 3\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] Saved checkpoint to \"/tmp/tmpg00zvb/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #progress_metric: host=algo-1, completed 26 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Total Batches Seen\": {\"count\": 1, \"max\": 51, \"sum\": 51.0, \"min\": 51}, \"Total Records Seen\": {\"count\": 1, \"max\": 9795, \"sum\": 9795.0, \"min\": 9795}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Reset Count\": {\"count\": 1, \"max\": 6, \"sum\": 6.0, \"min\": 6}}, \"EndTime\": 1605485445.01631, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605485444.666987}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #throughput_metric: host=algo-1, train throughput=5491.48861806 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:45.387] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 12, \"duration\": 370, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6353276952107747, \"sum\": 0.6353276952107747, \"min\": 0.6353276952107747}}, \"EndTime\": 1605485445.38774, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.387641}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6373595555623373, \"sum\": 0.6373595555623373, \"min\": 0.6373595555623373}}, \"EndTime\": 1605485445.387829, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.387808}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6430488162570529, \"sum\": 0.6430488162570529, \"min\": 0.6430488162570529}}, \"EndTime\": 1605485445.387902, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.387882}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6388365681966146, \"sum\": 0.6388365681966146, \"min\": 0.6388365681966146}}, \"EndTime\": 1605485445.387976, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.387958}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6023484887017144, \"sum\": 0.6023484887017144, \"min\": 0.6023484887017144}}, \"EndTime\": 1605485445.388052, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388033}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6003869205050998, \"sum\": 0.6003869205050998, \"min\": 0.6003869205050998}}, \"EndTime\": 1605485445.388116, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388099}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5998129950629341, \"sum\": 0.5998129950629341, \"min\": 0.5998129950629341}}, \"EndTime\": 1605485445.388184, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388166}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6000931167602539, \"sum\": 0.6000931167602539, \"min\": 0.6000931167602539}}, \"EndTime\": 1605485445.388258, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388239}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6466876051161025, \"sum\": 0.6466876051161025, \"min\": 0.6466876051161025}}, \"EndTime\": 1605485445.388322, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388305}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6388091405232748, \"sum\": 0.6388091405232748, \"min\": 0.6388091405232748}}, \"EndTime\": 1605485445.388395, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388375}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6360220675998264, \"sum\": 0.6360220675998264, \"min\": 0.6360220675998264}}, \"EndTime\": 1605485445.38847, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388452}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6425510448879665, \"sum\": 0.6425510448879665, \"min\": 0.6425510448879665}}, \"EndTime\": 1605485445.388534, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388516}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5990236239963108, \"sum\": 0.5990236239963108, \"min\": 0.5990236239963108}}, \"EndTime\": 1605485445.388598, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388581}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6063285912407769, \"sum\": 0.6063285912407769, \"min\": 0.6063285912407769}}, \"EndTime\": 1605485445.38866, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388642}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6000747934977213, \"sum\": 0.6000747934977213, \"min\": 0.6000747934977213}}, \"EndTime\": 1605485445.388721, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388705}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.599461424085829, \"sum\": 0.599461424085829, \"min\": 0.599461424085829}}, \"EndTime\": 1605485445.388782, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388765}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6613124084472656, \"sum\": 0.6613124084472656, \"min\": 0.6613124084472656}}, \"EndTime\": 1605485445.388843, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388826}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6627257876925998, \"sum\": 0.6627257876925998, \"min\": 0.6627257876925998}}, \"EndTime\": 1605485445.388905, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388888}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6603352525499132, \"sum\": 0.6603352525499132, \"min\": 0.6603352525499132}}, \"EndTime\": 1605485445.388964, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.388947}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.660744145711263, \"sum\": 0.660744145711263, \"min\": 0.660744145711263}}, \"EndTime\": 1605485445.389015, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389004}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.666025865342882, \"sum\": 0.666025865342882, \"min\": 0.666025865342882}}, \"EndTime\": 1605485445.38907, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389058}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6640348307291667, \"sum\": 0.6640348307291667, \"min\": 0.6640348307291667}}, \"EndTime\": 1605485445.389127, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389112}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6666323937310112, \"sum\": 0.6666323937310112, \"min\": 0.6666323937310112}}, \"EndTime\": 1605485445.389183, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389172}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6664860280354817, \"sum\": 0.6664860280354817, \"min\": 0.6664860280354817}}, \"EndTime\": 1605485445.389237, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389221}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6933414713541667, \"sum\": 0.6933414713541667, \"min\": 0.6933414713541667}}, \"EndTime\": 1605485445.389295, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389283}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6931917826334636, \"sum\": 0.6931917826334636, \"min\": 0.6931917826334636}}, \"EndTime\": 1605485445.389349, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389333}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6927628665500217, \"sum\": 0.6927628665500217, \"min\": 0.6927628665500217}}, \"EndTime\": 1605485445.389434, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389414}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6925951978895399, \"sum\": 0.6925951978895399, \"min\": 0.6925951978895399}}, \"EndTime\": 1605485445.389475, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389466}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6954899427625868, \"sum\": 0.6954899427625868, \"min\": 0.6954899427625868}}, \"EndTime\": 1605485445.389503, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389495}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7004153866238064, \"sum\": 0.7004153866238064, \"min\": 0.7004153866238064}}, \"EndTime\": 1605485445.389529, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389522}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6957335493299697, \"sum\": 0.6957335493299697, \"min\": 0.6957335493299697}}, \"EndTime\": 1605485445.389558, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389549}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6953580559624566, \"sum\": 0.6953580559624566, \"min\": 0.6953580559624566}}, \"EndTime\": 1605485445.389614, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.389597}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #quality_metric: host=algo-1, epoch=4, train binary_classification_cross_entropy_objective =0.635327695211\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #early_stopping_criteria_metric: host=algo-1, epoch=4, criteria=binary_classification_cross_entropy_objective, value=0.599023623996\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] Saving model for epoch: 4\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] Saved checkpoint to \"/tmp/tmpfw1Tl5/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #progress_metric: host=algo-1, completed 33 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Total Batches Seen\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Total Records Seen\": {\"count\": 1, \"max\": 11714, \"sum\": 11714.0, \"min\": 11714}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Reset Count\": {\"count\": 1, \"max\": 7, \"sum\": 7.0, \"min\": 7}}, \"EndTime\": 1605485445.396445, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605485445.016536}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #throughput_metric: host=algo-1, train throughput=5049.56127013 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:45.747] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 14, \"duration\": 350, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6284189097086589, \"sum\": 0.6284189097086589, \"min\": 0.6284189097086589}}, \"EndTime\": 1605485445.747645, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.747535}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6299942821926541, \"sum\": 0.6299942821926541, \"min\": 0.6299942821926541}}, \"EndTime\": 1605485445.747829, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.747803}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6353466415405273, \"sum\": 0.6353466415405273, \"min\": 0.6353466415405273}}, \"EndTime\": 1605485445.747902, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.747884}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6312264039781359, \"sum\": 0.6312264039781359, \"min\": 0.6312264039781359}}, \"EndTime\": 1605485445.748021, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.747947}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6010568152533637, \"sum\": 0.6010568152533637, \"min\": 0.6010568152533637}}, \"EndTime\": 1605485445.748099, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.748078}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5973881488376194, \"sum\": 0.5973881488376194, \"min\": 0.5973881488376194}}, \"EndTime\": 1605485445.748212, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.748148}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.597057033114963, \"sum\": 0.597057033114963, \"min\": 0.597057033114963}}, \"EndTime\": 1605485445.748403, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.748377}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5971932644314236, \"sum\": 0.5971932644314236, \"min\": 0.5971932644314236}}, \"EndTime\": 1605485445.748473, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.748453}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6385313118828667, \"sum\": 0.6385313118828667, \"min\": 0.6385313118828667}}, \"EndTime\": 1605485445.748578, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.748521}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6312756771511502, \"sum\": 0.6312756771511502, \"min\": 0.6312756771511502}}, \"EndTime\": 1605485445.748744, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.748722}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6290982140435113, \"sum\": 0.6290982140435113, \"min\": 0.6290982140435113}}, \"EndTime\": 1605485445.748815, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.748796}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6346627044677734, \"sum\": 0.6346627044677734, \"min\": 0.6346627044677734}}, \"EndTime\": 1605485445.748971, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.748863}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5970855246649848, \"sum\": 0.5970855246649848, \"min\": 0.5970855246649848}}, \"EndTime\": 1605485445.749075, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.74906}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6015496953328451, \"sum\": 0.6015496953328451, \"min\": 0.6015496953328451}}, \"EndTime\": 1605485445.74911, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749101}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5969048055013021, \"sum\": 0.5969048055013021, \"min\": 0.5969048055013021}}, \"EndTime\": 1605485445.749139, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749132}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5970354927910699, \"sum\": 0.5970354927910699, \"min\": 0.5970354927910699}}, \"EndTime\": 1605485445.749168, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749161}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6600311279296875, \"sum\": 0.6600311279296875, \"min\": 0.6600311279296875}}, \"EndTime\": 1605485445.749201, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749189}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6616830698649089, \"sum\": 0.6616830698649089, \"min\": 0.6616830698649089}}, \"EndTime\": 1605485445.74926, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749243}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6613296339246961, \"sum\": 0.6613296339246961, \"min\": 0.6613296339246961}}, \"EndTime\": 1605485445.749318, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749302}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6613988918728299, \"sum\": 0.6613988918728299, \"min\": 0.6613988918728299}}, \"EndTime\": 1605485445.749379, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749362}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6641958279079861, \"sum\": 0.6641958279079861, \"min\": 0.6641958279079861}}, \"EndTime\": 1605485445.749468, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749448}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6645449150933159, \"sum\": 0.6645449150933159, \"min\": 0.6645449150933159}}, \"EndTime\": 1605485445.74953, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749513}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6641675652398004, \"sum\": 0.6641675652398004, \"min\": 0.6641675652398004}}, \"EndTime\": 1605485445.74959, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749573}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6645725928412544, \"sum\": 0.6645725928412544, \"min\": 0.6645725928412544}}, \"EndTime\": 1605485445.749646, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749631}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6928930494520399, \"sum\": 0.6928930494520399, \"min\": 0.6928930494520399}}, \"EndTime\": 1605485445.749707, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.74969}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6932785458034939, \"sum\": 0.6932785458034939, \"min\": 0.6932785458034939}}, \"EndTime\": 1605485445.749768, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749751}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6926193067762587, \"sum\": 0.6926193067762587, \"min\": 0.6926193067762587}}, \"EndTime\": 1605485445.749827, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749811}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6920461357964409, \"sum\": 0.6920461357964409, \"min\": 0.6920461357964409}}, \"EndTime\": 1605485445.749886, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749869}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6961241573757595, \"sum\": 0.6961241573757595, \"min\": 0.6961241573757595}}, \"EndTime\": 1605485445.749943, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749927}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6948208872477214, \"sum\": 0.6948208872477214, \"min\": 0.6948208872477214}}, \"EndTime\": 1605485445.750003, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.749986}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6958341217041015, \"sum\": 0.6958341217041015, \"min\": 0.6958341217041015}}, \"EndTime\": 1605485445.750062, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.750045}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6957179599338108, \"sum\": 0.6957179599338108, \"min\": 0.6957179599338108}}, \"EndTime\": 1605485445.750124, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.750108}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #quality_metric: host=algo-1, epoch=5, train binary_classification_cross_entropy_objective =0.628418909709\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #early_stopping_criteria_metric: host=algo-1, epoch=5, criteria=binary_classification_cross_entropy_objective, value=0.596904805501\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] Saving model for epoch: 5\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] Saved checkpoint to \"/tmp/tmp1ydecN/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #progress_metric: host=algo-1, completed 40 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Total Batches Seen\": {\"count\": 1, \"max\": 71, \"sum\": 71.0, \"min\": 71}, \"Total Records Seen\": {\"count\": 1, \"max\": 13633, \"sum\": 13633.0, \"min\": 13633}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Reset Count\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}}, \"EndTime\": 1605485445.759217, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605485445.396719}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:45 INFO 140334821308224] #throughput_metric: host=algo-1, train throughput=5291.77862663 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:46.084] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 16, \"duration\": 324, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6226909340752496, \"sum\": 0.6226909340752496, \"min\": 0.6226909340752496}}, \"EndTime\": 1605485446.084585, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.08449}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6238636991712783, \"sum\": 0.6238636991712783, \"min\": 0.6238636991712783}}, \"EndTime\": 1605485446.084676, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.084655}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.62889646742079, \"sum\": 0.62889646742079, \"min\": 0.62889646742079}}, \"EndTime\": 1605485446.084747, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.084728}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.624873758951823, \"sum\": 0.624873758951823, \"min\": 0.624873758951823}}, \"EndTime\": 1605485446.084815, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.084796}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.597539185418023, \"sum\": 0.597539185418023, \"min\": 0.597539185418023}}, \"EndTime\": 1605485446.084883, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.084864}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5974565760294597, \"sum\": 0.5974565760294597, \"min\": 0.5974565760294597}}, \"EndTime\": 1605485446.084947, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.084929}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5974524815877279, \"sum\": 0.5974524815877279, \"min\": 0.5974524815877279}}, \"EndTime\": 1605485446.085011, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.084994}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5972771708170573, \"sum\": 0.5972771708170573, \"min\": 0.5972771708170573}}, \"EndTime\": 1605485446.085076, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085058}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6316647720336914, \"sum\": 0.6316647720336914, \"min\": 0.6316647720336914}}, \"EndTime\": 1605485446.085143, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085124}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6249888102213542, \"sum\": 0.6249888102213542, \"min\": 0.6249888102213542}}, \"EndTime\": 1605485446.085207, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.08519}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6233619054158529, \"sum\": 0.6233619054158529, \"min\": 0.6233619054158529}}, \"EndTime\": 1605485446.085279, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085262}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6280649354722765, \"sum\": 0.6280649354722765, \"min\": 0.6280649354722765}}, \"EndTime\": 1605485446.085347, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085329}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5974389478895399, \"sum\": 0.5974389478895399, \"min\": 0.5974389478895399}}, \"EndTime\": 1605485446.085434, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085416}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.597164683871799, \"sum\": 0.597164683871799, \"min\": 0.597164683871799}}, \"EndTime\": 1605485446.085494, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085477}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5973171064588758, \"sum\": 0.5973171064588758, \"min\": 0.5973171064588758}}, \"EndTime\": 1605485446.085557, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085539}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5973172251383464, \"sum\": 0.5973172251383464, \"min\": 0.5973172251383464}}, \"EndTime\": 1605485446.085621, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085604}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6600760311550564, \"sum\": 0.6600760311550564, \"min\": 0.6600760311550564}}, \"EndTime\": 1605485446.085681, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085664}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6610665893554688, \"sum\": 0.6610665893554688, \"min\": 0.6610665893554688}}, \"EndTime\": 1605485446.085743, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085726}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6616771613226996, \"sum\": 0.6616771613226996, \"min\": 0.6616771613226996}}, \"EndTime\": 1605485446.085805, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085788}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6615529971652561, \"sum\": 0.6615529971652561, \"min\": 0.6615529971652561}}, \"EndTime\": 1605485446.08587, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085852}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6658141750759549, \"sum\": 0.6658141750759549, \"min\": 0.6658141750759549}}, \"EndTime\": 1605485446.085932, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085913}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6664532979329427, \"sum\": 0.6664532979329427, \"min\": 0.6664532979329427}}, \"EndTime\": 1605485446.085993, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.085975}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6658127424452039, \"sum\": 0.6658127424452039, \"min\": 0.6658127424452039}}, \"EndTime\": 1605485446.086048, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086032}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6660934024386935, \"sum\": 0.6660934024386935, \"min\": 0.6660934024386935}}, \"EndTime\": 1605485446.086115, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086097}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6923685794406467, \"sum\": 0.6923685794406467, \"min\": 0.6923685794406467}}, \"EndTime\": 1605485446.086177, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086159}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6925773620605469, \"sum\": 0.6925773620605469, \"min\": 0.6925773620605469}}, \"EndTime\": 1605485446.086237, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.08622}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6926222313774957, \"sum\": 0.6926222313774957, \"min\": 0.6926222313774957}}, \"EndTime\": 1605485446.086292, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086278}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6926779090033637, \"sum\": 0.6926779090033637, \"min\": 0.6926779090033637}}, \"EndTime\": 1605485446.086343, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086329}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.698605482313368, \"sum\": 0.698605482313368, \"min\": 0.698605482313368}}, \"EndTime\": 1605485446.086394, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086378}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6962572987874349, \"sum\": 0.6962572987874349, \"min\": 0.6962572987874349}}, \"EndTime\": 1605485446.086445, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086431}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6982844628228082, \"sum\": 0.6982844628228082, \"min\": 0.6982844628228082}}, \"EndTime\": 1605485446.086503, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086486}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.69852781507704, \"sum\": 0.69852781507704, \"min\": 0.69852781507704}}, \"EndTime\": 1605485446.086572, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485446.086554}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #quality_metric: host=algo-1, epoch=6, train binary_classification_cross_entropy_objective =0.622690934075\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #early_stopping_criteria_metric: host=algo-1, epoch=6, criteria=binary_classification_cross_entropy_objective, value=0.597164683872\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] Saving model for epoch: 6\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] Saved checkpoint to \"/tmp/tmpwnt6Rx/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] Early stop condition met. Stopping training.\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #progress_metric: host=algo-1, completed 100 % epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Total Batches Seen\": {\"count\": 1, \"max\": 81, \"sum\": 81.0, \"min\": 81}, \"Total Records Seen\": {\"count\": 1, \"max\": 15552, \"sum\": 15552.0, \"min\": 15552}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1919, \"sum\": 1919.0, \"min\": 1919}, \"Reset Count\": {\"count\": 1, \"max\": 9, \"sum\": 9.0, \"min\": 9}}, \"EndTime\": 1605485446.093186, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605485445.759504}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #throughput_metric: host=algo-1, train throughput=5749.29918413 records/second\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 WARNING 140334821308224] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 WARNING 140334821308224] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:46.095] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 18, \"duration\": 1, \"num_examples\": 1, \"num_bytes\": 12000}\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:46.214] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 21, \"duration\": 117, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m[2020-11-16 00:10:46.255] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 23, \"duration\": 30, \"num_examples\": 10, \"num_bytes\": 115140}\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #train_score (algo-1) : ('binary_classification_cross_entropy_objective', 0.5922795504937264)\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #train_score (algo-1) : ('binary_classification_accuracy', 0.675872850442939)\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #train_score (algo-1) : ('binary_f_1.000', 0.6793814432989691)\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #train_score (algo-1) : ('precision', 0.6663296258847321)\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #train_score (algo-1) : ('recall', 0.6929547844374343)\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #train_score (algo-1) : ('roc_auc_score', 0.7445392409903451)\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #quality_metric: host=algo-1, train binary_classification_cross_entropy_objective =0.592279550494\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #quality_metric: host=algo-1, train binary_classification_accuracy =0.675872850443\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #quality_metric: host=algo-1, train binary_f_1.000 =0.679381443299\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #quality_metric: host=algo-1, train precision =0.666329625885\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #quality_metric: host=algo-1, train recall =0.692954784437\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] #quality_metric: host=algo-1, train roc_auc_score =0.74453924099\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] Best model found for hyperparameters: {\"lr_scheduler_step\": 100, \"wd\": 0.01, \"optimizer\": \"adam\", \"lr_scheduler_factor\": 0.99, \"l1\": 0.0, \"learning_rate\": 0.1, \"lr_scheduler_minimum_lr\": 0.0001}\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] Saved checkpoint to \"/tmp/tmpZi__Bh/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:10:46 INFO 140334821308224] Test data is not provided.\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"totaltime\": {\"count\": 1, \"max\": 2963.0990028381348, \"sum\": 2963.0990028381348, \"min\": 2963.0990028381348}, \"finalize.time\": {\"count\": 1, \"max\": 166.82100296020508, \"sum\": 166.82100296020508, \"min\": 166.82100296020508}, \"initialize.time\": {\"count\": 1, \"max\": 175.1728057861328, \"sum\": 175.1728057861328, \"min\": 175.1728057861328}, \"check_early_stopping.time\": {\"count\": 7, \"max\": 1.0309219360351562, \"sum\": 3.591775894165039, \"min\": 0.16999244689941406}, \"setuptime\": {\"count\": 1, \"max\": 25.515079498291016, \"sum\": 25.515079498291016, \"min\": 25.515079498291016}, \"update.time\": {\"count\": 7, \"max\": 377.5210380554199, \"sum\": 2337.1262550354004, \"min\": 260.4329586029053}, \"epochs\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}}, \"EndTime\": 1605485446.265008, \"Dimensions\": {\"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1605485443.553981}\n", "\u001b[0m\n", "\n", "2020-11-16 00:10:56 Uploading - Uploading generated training model\n", "2020-11-16 00:10:56 Completed - Training job completed\n", "Training seconds: 54\n", "Billable seconds: 54\n", "-------------------!2020-11-16 00:20:46 Starting - Starting the training job...\n", "2020-11-16 00:20:47 Starting - Launching requested ML instances......\n", "2020-11-16 00:22:03 Starting - Preparing the instances for training.........\n", "2020-11-16 00:23:20 Downloading - Downloading input data...\n", "2020-11-16 00:24:12 Training - Training image download completed. Training in progress.\u001b[34mDocker entrypoint called with argument(s): train\u001b[0m\n", "\u001b[34mRunning default environment configuration script\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:16 INFO 139985910667072] Reading default configuration from /opt/amazon/lib/python2.7/site-packages/algorithm/resources/default-input.json: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'auto', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'1000', u'huber_delta': u'1.0', u'num_classes': u'1', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:16 INFO 139985910667072] Merging with provided configuration from /opt/ml/input/config/hyperparameters.json: {u'feature_dim': u'3', u'mini_batch_size': u'200', u'predictor_type': u'binary_classifier'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:16 INFO 139985910667072] Final configuration: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'3', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'200', u'huber_delta': u'1.0', u'num_classes': u'1', u'predictor_type': u'binary_classifier', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:16 WARNING 139985910667072] Loggers have already been setup.\u001b[0m\n", "\u001b[34mProcess 1 is a worker.\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:16 INFO 139985910667072] Using default worker.\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] Checkpoint loading and saving are disabled.\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:17.057] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 0, \"duration\": 16, \"num_examples\": 1, \"num_bytes\": 11200}\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] Create Store: local\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:17.134] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 1, \"duration\": 75, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] Scaler algorithm parameters\n", " \u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] Scaling model computed with parameters:\n", " {'stdev_weight': \u001b[0m\n", "\u001b[34m[3.6492198 3.6978192 2.5404634]\u001b[0m\n", "\u001b[34m, 'stdev_label': None, 'mean_label': None, 'mean_weight': \u001b[0m\n", "\u001b[34m[22.810356 22.779465 8.678215]\u001b[0m\n", "\u001b[34m}\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] nvidia-smi took: 0.0251319408417 secs to identify 0 gpus\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] Number of GPUs being used: 0\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Total Batches Seen\": {\"count\": 1, \"max\": 16, \"sum\": 16.0, \"min\": 16}, \"Total Records Seen\": {\"count\": 1, \"max\": 3166, \"sum\": 3166.0, \"min\": 3166}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 2, \"sum\": 2.0, \"min\": 2}}, \"EndTime\": 1605486257.249232, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"init_train_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1605486257.249195}\n", "\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:17.711] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 4, \"duration\": 461, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.637645754132952, \"sum\": 0.637645754132952, \"min\": 0.637645754132952}}, \"EndTime\": 1605486257.711742, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.711652}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.634024862561907, \"sum\": 0.634024862561907, \"min\": 0.634024862561907}}, \"EndTime\": 1605486257.711853, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.711829}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6549589838300433, \"sum\": 0.6549589838300433, \"min\": 0.6549589838300433}}, \"EndTime\": 1605486257.711918, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.711904}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6350840214320591, \"sum\": 0.6350840214320591, \"min\": 0.6350840214320591}}, \"EndTime\": 1605486257.71199, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.711971}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5590795980181013, \"sum\": 0.5590795980181013, \"min\": 0.5590795980181013}}, \"EndTime\": 1605486257.712069, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712049}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5530626787458147, \"sum\": 0.5530626787458147, \"min\": 0.5530626787458147}}, \"EndTime\": 1605486257.71213, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712112}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5623528671264648, \"sum\": 0.5623528671264648, \"min\": 0.5623528671264648}}, \"EndTime\": 1605486257.712205, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712186}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5557535144260951, \"sum\": 0.5557535144260951, \"min\": 0.5557535144260951}}, \"EndTime\": 1605486257.712266, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712248}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6491821398053851, \"sum\": 0.6491821398053851, \"min\": 0.6491821398053851}}, \"EndTime\": 1605486257.712339, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.71232}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6388907105582101, \"sum\": 0.6388907105582101, \"min\": 0.6388907105582101}}, \"EndTime\": 1605486257.712398, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712381}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.632131061553955, \"sum\": 0.632131061553955, \"min\": 0.632131061553955}}, \"EndTime\": 1605486257.712472, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712452}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.644635797228132, \"sum\": 0.644635797228132, \"min\": 0.644635797228132}}, \"EndTime\": 1605486257.712546, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712529}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5593024417332241, \"sum\": 0.5593024417332241, \"min\": 0.5593024417332241}}, \"EndTime\": 1605486257.71261, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712591}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5534083720615932, \"sum\": 0.5534083720615932, \"min\": 0.5534083720615932}}, \"EndTime\": 1605486257.712667, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.71265}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5600722231183733, \"sum\": 0.5600722231183733, \"min\": 0.5600722231183733}}, \"EndTime\": 1605486257.712737, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712718}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5548205702645438, \"sum\": 0.5548205702645438, \"min\": 0.5548205702645438}}, \"EndTime\": 1605486257.712792, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712775}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6428719002859933, \"sum\": 0.6428719002859933, \"min\": 0.6428719002859933}}, \"EndTime\": 1605486257.712854, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712837}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6333479309082031, \"sum\": 0.6333479309082031, \"min\": 0.6333479309082031}}, \"EndTime\": 1605486257.712917, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712901}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6464505686078753, \"sum\": 0.6464505686078753, \"min\": 0.6464505686078753}}, \"EndTime\": 1605486257.71299, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.712971}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6590950420924595, \"sum\": 0.6590950420924595, \"min\": 0.6590950420924595}}, \"EndTime\": 1605486257.713045, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713029}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6084901673453195, \"sum\": 0.6084901673453195, \"min\": 0.6084901673453195}}, \"EndTime\": 1605486257.713117, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713099}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6073133114406041, \"sum\": 0.6073133114406041, \"min\": 0.6073133114406041}}, \"EndTime\": 1605486257.713173, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713156}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6075217110770089, \"sum\": 0.6075217110770089, \"min\": 0.6075217110770089}}, \"EndTime\": 1605486257.713233, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713216}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6071783038548061, \"sum\": 0.6071783038548061, \"min\": 0.6071783038548061}}, \"EndTime\": 1605486257.713284, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713268}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6517755154200963, \"sum\": 0.6517755154200963, \"min\": 0.6517755154200963}}, \"EndTime\": 1605486257.713355, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713336}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6498830877031598, \"sum\": 0.6498830877031598, \"min\": 0.6498830877031598}}, \"EndTime\": 1605486257.713416, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713398}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6482956068856376, \"sum\": 0.6482956068856376, \"min\": 0.6482956068856376}}, \"EndTime\": 1605486257.713514, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713468}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6516669763837541, \"sum\": 0.6516669763837541, \"min\": 0.6516669763837541}}, \"EndTime\": 1605486257.713573, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713555}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6312719535827637, \"sum\": 0.6312719535827637, \"min\": 0.6312719535827637}}, \"EndTime\": 1605486257.71363, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713614}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6307333619253976, \"sum\": 0.6307333619253976, \"min\": 0.6307333619253976}}, \"EndTime\": 1605486257.713703, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713684}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6317810303824288, \"sum\": 0.6317810303824288, \"min\": 0.6317810303824288}}, \"EndTime\": 1605486257.713759, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713742}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6318345015389578, \"sum\": 0.6318345015389578, \"min\": 0.6318345015389578}}, \"EndTime\": 1605486257.713831, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.713812}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] #quality_metric: host=algo-1, epoch=0, train binary_classification_cross_entropy_objective =0.637645754133\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=0, criteria=binary_classification_cross_entropy_objective, value=0.553062678746\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] Epoch 0: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] Saving model for epoch: 0\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] Saved checkpoint to \"/tmp/tmp4T6MxV/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] #progress_metric: host=algo-1, completed 6 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 31, \"sum\": 31.0, \"min\": 31}, \"Total Records Seen\": {\"count\": 1, \"max\": 6132, \"sum\": 6132.0, \"min\": 6132}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 3, \"sum\": 3.0, \"min\": 3}}, \"EndTime\": 1605486257.724362, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605486257.249549}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:17 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=6244.11775595 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:18.241] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 6, \"duration\": 516, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5882803807939802, \"sum\": 0.5882803807939802, \"min\": 0.5882803807939802}}, \"EndTime\": 1605486258.241273, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.24118}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.584415454864502, \"sum\": 0.584415454864502, \"min\": 0.584415454864502}}, \"EndTime\": 1605486258.241369, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241347}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6041358239310128, \"sum\": 0.6041358239310128, \"min\": 0.6041358239310128}}, \"EndTime\": 1605486258.241441, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241422}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5853705269949777, \"sum\": 0.5853705269949777, \"min\": 0.5853705269949777}}, \"EndTime\": 1605486258.241539, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241517}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5116412108285087, \"sum\": 0.5116412108285087, \"min\": 0.5116412108285087}}, \"EndTime\": 1605486258.24161, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241591}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5117790739876883, \"sum\": 0.5117790739876883, \"min\": 0.5117790739876883}}, \"EndTime\": 1605486258.241675, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241657}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.511460702078683, \"sum\": 0.511460702078683, \"min\": 0.511460702078683}}, \"EndTime\": 1605486258.241738, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.24172}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5117059544154576, \"sum\": 0.5117059544154576, \"min\": 0.5117059544154576}}, \"EndTime\": 1605486258.241801, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241784}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5990202576773507, \"sum\": 0.5990202576773507, \"min\": 0.5990202576773507}}, \"EndTime\": 1605486258.241875, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241858}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5891933795383998, \"sum\": 0.5891933795383998, \"min\": 0.5891933795383998}}, \"EndTime\": 1605486258.24194, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241922}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5836289760044643, \"sum\": 0.5836289760044643, \"min\": 0.5836289760044643}}, \"EndTime\": 1605486258.242002, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.241984}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.594363465990339, \"sum\": 0.594363465990339, \"min\": 0.594363465990339}}, \"EndTime\": 1605486258.242062, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242046}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5116182354518345, \"sum\": 0.5116182354518345, \"min\": 0.5116182354518345}}, \"EndTime\": 1605486258.24212, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242104}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5113560349600655, \"sum\": 0.5113560349600655, \"min\": 0.5113560349600655}}, \"EndTime\": 1605486258.24218, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242165}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.511477655683245, \"sum\": 0.511477655683245, \"min\": 0.511477655683245}}, \"EndTime\": 1605486258.242238, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242223}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5115412902832032, \"sum\": 0.5115412902832032, \"min\": 0.5115412902832032}}, \"EndTime\": 1605486258.242294, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242277}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5963270487104143, \"sum\": 0.5963270487104143, \"min\": 0.5963270487104143}}, \"EndTime\": 1605486258.242353, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242336}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5887478419712612, \"sum\": 0.5887478419712612, \"min\": 0.5887478419712612}}, \"EndTime\": 1605486258.242415, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242397}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5993044744219099, \"sum\": 0.5993044744219099, \"min\": 0.5993044744219099}}, \"EndTime\": 1605486258.242475, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242458}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6077384676252092, \"sum\": 0.6077384676252092, \"min\": 0.6077384676252092}}, \"EndTime\": 1605486258.242536, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242519}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5911470331464495, \"sum\": 0.5911470331464495, \"min\": 0.5911470331464495}}, \"EndTime\": 1605486258.242595, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242577}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5908390944344657, \"sum\": 0.5908390944344657, \"min\": 0.5908390944344657}}, \"EndTime\": 1605486258.242656, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242639}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5907778930664063, \"sum\": 0.5907778930664063, \"min\": 0.5907778930664063}}, \"EndTime\": 1605486258.242717, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242699}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5910882813589914, \"sum\": 0.5910882813589914, \"min\": 0.5910882813589914}}, \"EndTime\": 1605486258.242777, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.24276}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6129420443943568, \"sum\": 0.6129420443943568, \"min\": 0.6129420443943568}}, \"EndTime\": 1605486258.242835, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.242819}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6141044044494629, \"sum\": 0.6141044044494629, \"min\": 0.6141044044494629}}, \"EndTime\": 1605486258.242896, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.24288}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6148060498918806, \"sum\": 0.6148060498918806, \"min\": 0.6148060498918806}}, \"EndTime\": 1605486258.242958, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.24294}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.613275078364781, \"sum\": 0.613275078364781, \"min\": 0.613275078364781}}, \"EndTime\": 1605486258.243019, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.243002}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6179983275277274, \"sum\": 0.6179983275277274, \"min\": 0.6179983275277274}}, \"EndTime\": 1605486258.243079, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.243062}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6182128252301897, \"sum\": 0.6182128252301897, \"min\": 0.6182128252301897}}, \"EndTime\": 1605486258.243139, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.243122}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6181222452436175, \"sum\": 0.6181222452436175, \"min\": 0.6181222452436175}}, \"EndTime\": 1605486258.243196, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.243179}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6172580310276576, \"sum\": 0.6172580310276576, \"min\": 0.6172580310276576}}, \"EndTime\": 1605486258.243256, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486258.243239}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] #quality_metric: host=algo-1, epoch=1, train binary_classification_cross_entropy_objective =0.588280380794\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=1, criteria=binary_classification_cross_entropy_objective, value=0.51135603496\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] Epoch 1: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] Saving model for epoch: 1\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] Saved checkpoint to \"/tmp/tmpQMy6fy/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] #progress_metric: host=algo-1, completed 13 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 46, \"sum\": 46.0, \"min\": 46}, \"Total Records Seen\": {\"count\": 1, \"max\": 9098, \"sum\": 9098.0, \"min\": 9098}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 4, \"sum\": 4.0, \"min\": 4}}, \"EndTime\": 1605486258.252849, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605486257.72465}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=5614.03115169 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:18.746] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 8, \"duration\": 493, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5758003153119768, \"sum\": 0.5758003153119768, \"min\": 0.5758003153119768}}, \"EndTime\": 1605486258.746974, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.746864}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5718298694065639, \"sum\": 0.5718298694065639, \"min\": 0.5718298694065639}}, \"EndTime\": 1605486258.747076, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747056}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5902846036638533, \"sum\": 0.5902846036638533, \"min\": 0.5902846036638533}}, \"EndTime\": 1605486258.747147, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747128}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5727040699550083, \"sum\": 0.5727040699550083, \"min\": 0.5727040699550083}}, \"EndTime\": 1605486258.74721, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747193}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5107579803466797, \"sum\": 0.5107579803466797, \"min\": 0.5107579803466797}}, \"EndTime\": 1605486258.747269, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747254}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5106482342311314, \"sum\": 0.5106482342311314, \"min\": 0.5106482342311314}}, \"EndTime\": 1605486258.747326, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.74731}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5107393945966449, \"sum\": 0.5107393945966449, \"min\": 0.5107393945966449}}, \"EndTime\": 1605486258.747387, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747371}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5107848957606724, \"sum\": 0.5107848957606724, \"min\": 0.5107848957606724}}, \"EndTime\": 1605486258.747443, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747428}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5857466452462332, \"sum\": 0.5857466452462332, \"min\": 0.5857466452462332}}, \"EndTime\": 1605486258.747497, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747483}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5762719236101423, \"sum\": 0.5762719236101423, \"min\": 0.5762719236101423}}, \"EndTime\": 1605486258.747551, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747535}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.57189850125994, \"sum\": 0.57189850125994, \"min\": 0.57189850125994}}, \"EndTime\": 1605486258.747617, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747601}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5811126572745187, \"sum\": 0.5811126572745187, \"min\": 0.5811126572745187}}, \"EndTime\": 1605486258.747674, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747658}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5093015725272042, \"sum\": 0.5093015725272042, \"min\": 0.5093015725272042}}, \"EndTime\": 1605486258.747732, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747715}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5091952541896275, \"sum\": 0.5091952541896275, \"min\": 0.5091952541896275}}, \"EndTime\": 1605486258.747789, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747774}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5091850471496582, \"sum\": 0.5091850471496582, \"min\": 0.5091850471496582}}, \"EndTime\": 1605486258.747847, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.74783}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.509218270438058, \"sum\": 0.509218270438058, \"min\": 0.509218270438058}}, \"EndTime\": 1605486258.747916, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747899}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5887706020900182, \"sum\": 0.5887706020900182, \"min\": 0.5887706020900182}}, \"EndTime\": 1605486258.747971, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.747957}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5850768906729562, \"sum\": 0.5850768906729562, \"min\": 0.5850768906729562}}, \"EndTime\": 1605486258.748025, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748011}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5908012253897531, \"sum\": 0.5908012253897531, \"min\": 0.5908012253897531}}, \"EndTime\": 1605486258.74808, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748065}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5960572324480329, \"sum\": 0.5960572324480329, \"min\": 0.5960572324480329}}, \"EndTime\": 1605486258.748139, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748122}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5965159225463867, \"sum\": 0.5965159225463867, \"min\": 0.5965159225463867}}, \"EndTime\": 1605486258.748195, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748181}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5970208495003837, \"sum\": 0.5970208495003837, \"min\": 0.5970208495003837}}, \"EndTime\": 1605486258.748249, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748235}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5970087323869978, \"sum\": 0.5970087323869978, \"min\": 0.5970087323869978}}, \"EndTime\": 1605486258.748307, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748292}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5965255628313337, \"sum\": 0.5965255628313337, \"min\": 0.5965255628313337}}, \"EndTime\": 1605486258.748366, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748349}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6130256843566895, \"sum\": 0.6130256843566895, \"min\": 0.6130256843566895}}, \"EndTime\": 1605486258.748422, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748407}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6129018892560686, \"sum\": 0.6129018892560686, \"min\": 0.6129018892560686}}, \"EndTime\": 1605486258.748474, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.74846}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6130914715358189, \"sum\": 0.6130914715358189, \"min\": 0.6130914715358189}}, \"EndTime\": 1605486258.748527, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748513}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6128841699872698, \"sum\": 0.6128841699872698, \"min\": 0.6128841699872698}}, \"EndTime\": 1605486258.74859, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748574}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6242315810067313, \"sum\": 0.6242315810067313, \"min\": 0.6242315810067313}}, \"EndTime\": 1605486258.748645, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748631}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6245166233607701, \"sum\": 0.6245166233607701, \"min\": 0.6245166233607701}}, \"EndTime\": 1605486258.748709, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748695}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6241741725376674, \"sum\": 0.6241741725376674, \"min\": 0.6241741725376674}}, \"EndTime\": 1605486258.748765, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.74875}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6251368413652693, \"sum\": 0.6251368413652693, \"min\": 0.6251368413652693}}, \"EndTime\": 1605486258.748822, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.748806}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] #quality_metric: host=algo-1, epoch=2, train binary_classification_cross_entropy_objective =0.575800315312\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=2, criteria=binary_classification_cross_entropy_objective, value=0.50918504715\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] Epoch 2: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] Saving model for epoch: 2\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] Saved checkpoint to \"/tmp/tmpy4hr1E/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] #progress_metric: host=algo-1, completed 20 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Total Records Seen\": {\"count\": 1, \"max\": 12064, \"sum\": 12064.0, \"min\": 12064}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 5, \"sum\": 5.0, \"min\": 5}}, \"EndTime\": 1605486258.756699, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605486258.253124}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:18 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=5888.23870447 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:19.335] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 10, \"duration\": 578, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5653207642691476, \"sum\": 0.5653207642691476, \"min\": 0.5653207642691476}}, \"EndTime\": 1605486259.335457, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.335369}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5612058912004744, \"sum\": 0.5612058912004744, \"min\": 0.5612058912004744}}, \"EndTime\": 1605486259.335568, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.335546}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5785590853009905, \"sum\": 0.5785590853009905, \"min\": 0.5785590853009905}}, \"EndTime\": 1605486259.335707, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.335684}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5620190184456961, \"sum\": 0.5620190184456961, \"min\": 0.5620190184456961}}, \"EndTime\": 1605486259.335786, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.335766}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5083217048645019, \"sum\": 0.5083217048645019, \"min\": 0.5083217048645019}}, \"EndTime\": 1605486259.33585, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.335832}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5078787394932338, \"sum\": 0.5078787394932338, \"min\": 0.5078787394932338}}, \"EndTime\": 1605486259.335917, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.335899}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5083279173714774, \"sum\": 0.5083279173714774, \"min\": 0.5083279173714774}}, \"EndTime\": 1605486259.335978, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.33596}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5080213601248605, \"sum\": 0.5080213601248605, \"min\": 0.5080213601248605}}, \"EndTime\": 1605486259.336061, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336033}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5745342935834612, \"sum\": 0.5745342935834612, \"min\": 0.5745342935834612}}, \"EndTime\": 1605486259.336145, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336123}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5652319581168038, \"sum\": 0.5652319581168038, \"min\": 0.5652319581168038}}, \"EndTime\": 1605486259.336219, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.3362}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5620532717023577, \"sum\": 0.5620532717023577, \"min\": 0.5620532717023577}}, \"EndTime\": 1605486259.336291, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336273}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5699098478044783, \"sum\": 0.5699098478044783, \"min\": 0.5699098478044783}}, \"EndTime\": 1605486259.336355, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336338}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5091331808907645, \"sum\": 0.5091331808907645, \"min\": 0.5091331808907645}}, \"EndTime\": 1605486259.33643, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336411}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5091312190464564, \"sum\": 0.5091312190464564, \"min\": 0.5091312190464564}}, \"EndTime\": 1605486259.336503, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336483}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5087951905386788, \"sum\": 0.5087951905386788, \"min\": 0.5087951905386788}}, \"EndTime\": 1605486259.336581, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.33656}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5091564968654088, \"sum\": 0.5091564968654088, \"min\": 0.5091564968654088}}, \"EndTime\": 1605486259.336658, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336638}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5849486105782645, \"sum\": 0.5849486105782645, \"min\": 0.5849486105782645}}, \"EndTime\": 1605486259.336723, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336705}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5846778079441616, \"sum\": 0.5846778079441616, \"min\": 0.5846778079441616}}, \"EndTime\": 1605486259.336799, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336779}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5859801319667272, \"sum\": 0.5859801319667272, \"min\": 0.5859801319667272}}, \"EndTime\": 1605486259.336866, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336847}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5889791107177734, \"sum\": 0.5889791107177734, \"min\": 0.5889791107177734}}, \"EndTime\": 1605486259.336928, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336911}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5986649377005441, \"sum\": 0.5986649377005441, \"min\": 0.5986649377005441}}, \"EndTime\": 1605486259.336991, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.336974}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5980611092703683, \"sum\": 0.5980611092703683, \"min\": 0.5980611092703683}}, \"EndTime\": 1605486259.337054, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337037}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5986903653826032, \"sum\": 0.5986903653826032, \"min\": 0.5986903653826032}}, \"EndTime\": 1605486259.337112, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337095}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5976421437944685, \"sum\": 0.5976421437944685, \"min\": 0.5976421437944685}}, \"EndTime\": 1605486259.337185, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337166}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119437408447266, \"sum\": 0.6119437408447266, \"min\": 0.6119437408447266}}, \"EndTime\": 1605486259.337249, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337232}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6117468697684152, \"sum\": 0.6117468697684152, \"min\": 0.6117468697684152}}, \"EndTime\": 1605486259.337319, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337301}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6115300233023507, \"sum\": 0.6115300233023507, \"min\": 0.6115300233023507}}, \"EndTime\": 1605486259.337388, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337371}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119194139753069, \"sum\": 0.6119194139753069, \"min\": 0.6119194139753069}}, \"EndTime\": 1605486259.337457, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.33744}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6271515655517578, \"sum\": 0.6271515655517578, \"min\": 0.6271515655517578}}, \"EndTime\": 1605486259.337544, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337525}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6266433416094099, \"sum\": 0.6266433416094099, \"min\": 0.6266433416094099}}, \"EndTime\": 1605486259.33761, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337592}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6271597644260951, \"sum\": 0.6271597644260951, \"min\": 0.6271597644260951}}, \"EndTime\": 1605486259.337678, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.33766}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6266192136492048, \"sum\": 0.6266192136492048, \"min\": 0.6266192136492048}}, \"EndTime\": 1605486259.337746, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486259.337728}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] #quality_metric: host=algo-1, epoch=3, train binary_classification_cross_entropy_objective =0.565320764269\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=3, criteria=binary_classification_cross_entropy_objective, value=0.507878739493\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] Epoch 3: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] Saving model for epoch: 3\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] Saved checkpoint to \"/tmp/tmpqWKpFX/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] #progress_metric: host=algo-1, completed 26 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 76, \"sum\": 76.0, \"min\": 76}, \"Total Records Seen\": {\"count\": 1, \"max\": 15030, \"sum\": 15030.0, \"min\": 15030}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 6, \"sum\": 6.0, \"min\": 6}}, \"EndTime\": 1605486259.34683, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605486258.757001}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=5027.44239759 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:19.876] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 12, \"duration\": 528, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5575827326093401, \"sum\": 0.5575827326093401, \"min\": 0.5575827326093401}}, \"EndTime\": 1605486259.87652, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.876384}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5534326771327428, \"sum\": 0.5534326771327428, \"min\": 0.5534326771327428}}, \"EndTime\": 1605486259.876641, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.876616}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5696773910522461, \"sum\": 0.5696773910522461, \"min\": 0.5696773910522461}}, \"EndTime\": 1605486259.876726, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.876706}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5541913849966866, \"sum\": 0.5541913849966866, \"min\": 0.5541913849966866}}, \"EndTime\": 1605486259.87679, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.876772}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.506830804007394, \"sum\": 0.506830804007394, \"min\": 0.506830804007394}}, \"EndTime\": 1605486259.876851, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.876834}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5066314724513462, \"sum\": 0.5066314724513462, \"min\": 0.5066314724513462}}, \"EndTime\": 1605486259.876912, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.876895}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5068106160845075, \"sum\": 0.5068106160845075, \"min\": 0.5068106160845075}}, \"EndTime\": 1605486259.876972, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.876955}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5066622080121721, \"sum\": 0.5066622080121721, \"min\": 0.5066622080121721}}, \"EndTime\": 1605486259.877033, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877016}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5661302484784807, \"sum\": 0.5661302484784807, \"min\": 0.5661302484784807}}, \"EndTime\": 1605486259.877093, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877076}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.55709045955113, \"sum\": 0.55709045955113, \"min\": 0.55709045955113}}, \"EndTime\": 1605486259.877153, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877136}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5548588562011719, \"sum\": 0.5548588562011719, \"min\": 0.5548588562011719}}, \"EndTime\": 1605486259.877211, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877196}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.56160003389631, \"sum\": 0.56160003389631, \"min\": 0.56160003389631}}, \"EndTime\": 1605486259.877266, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877252}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5085948726109096, \"sum\": 0.5085948726109096, \"min\": 0.5085948726109096}}, \"EndTime\": 1605486259.87733, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877314}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5083023888724191, \"sum\": 0.5083023888724191, \"min\": 0.5083023888724191}}, \"EndTime\": 1605486259.877384, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.87737}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5085396575927734, \"sum\": 0.5085396575927734, \"min\": 0.5085396575927734}}, \"EndTime\": 1605486259.877439, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877426}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5083411952427456, \"sum\": 0.5083411952427456, \"min\": 0.5083411952427456}}, \"EndTime\": 1605486259.877525, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877502}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5847108432224819, \"sum\": 0.5847108432224819, \"min\": 0.5847108432224819}}, \"EndTime\": 1605486259.877604, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877584}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5858351707458496, \"sum\": 0.5858351707458496, \"min\": 0.5858351707458496}}, \"EndTime\": 1605486259.877671, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877652}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5849497767857142, \"sum\": 0.5849497767857142, \"min\": 0.5849497767857142}}, \"EndTime\": 1605486259.877712, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877702}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5864124080113002, \"sum\": 0.5864124080113002, \"min\": 0.5864124080113002}}, \"EndTime\": 1605486259.87774, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877733}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5985960578918457, \"sum\": 0.5985960578918457, \"min\": 0.5985960578918457}}, \"EndTime\": 1605486259.877789, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877775}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5965376581464495, \"sum\": 0.5965376581464495, \"min\": 0.5965376581464495}}, \"EndTime\": 1605486259.87782, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877812}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5981988661629813, \"sum\": 0.5981988661629813, \"min\": 0.5981988661629813}}, \"EndTime\": 1605486259.877848, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.87784}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.59631623676845, \"sum\": 0.59631623676845, \"min\": 0.59631623676845}}, \"EndTime\": 1605486259.877908, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.87789}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6118321718488421, \"sum\": 0.6118321718488421, \"min\": 0.6118321718488421}}, \"EndTime\": 1605486259.87797, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.877958}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119981901986259, \"sum\": 0.6119981901986259, \"min\": 0.6119981901986259}}, \"EndTime\": 1605486259.878032, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.878015}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6120226342337473, \"sum\": 0.6120226342337473, \"min\": 0.6120226342337473}}, \"EndTime\": 1605486259.87807, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.878061}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6118924168178014, \"sum\": 0.6118924168178014, \"min\": 0.6118924168178014}}, \"EndTime\": 1605486259.878098, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.87809}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6279076548985073, \"sum\": 0.6279076548985073, \"min\": 0.6279076548985073}}, \"EndTime\": 1605486259.878152, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.878135}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6265868868146623, \"sum\": 0.6265868868146623, \"min\": 0.6265868868146623}}, \"EndTime\": 1605486259.878193, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.878183}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6281863403320312, \"sum\": 0.6281863403320312, \"min\": 0.6281863403320312}}, \"EndTime\": 1605486259.878221, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.878214}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6269790431431361, \"sum\": 0.6269790431431361, \"min\": 0.6269790431431361}}, \"EndTime\": 1605486259.878273, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.878257}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] #quality_metric: host=algo-1, epoch=4, train binary_classification_cross_entropy_objective =0.557582732609\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=4, criteria=binary_classification_cross_entropy_objective, value=0.506631472451\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] Epoch 4: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] Saving model for epoch: 4\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] Saved checkpoint to \"/tmp/tmp74stoh/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] #progress_metric: host=algo-1, completed 33 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 91, \"sum\": 91.0, \"min\": 91}, \"Total Records Seen\": {\"count\": 1, \"max\": 17996, \"sum\": 17996.0, \"min\": 17996}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 7, \"sum\": 7.0, \"min\": 7}}, \"EndTime\": 1605486259.885866, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605486259.347133}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:19 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=5504.09972971 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:20.287] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 14, \"duration\": 401, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5510115269252233, \"sum\": 0.5510115269252233, \"min\": 0.5510115269252233}}, \"EndTime\": 1605486260.288102, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288015}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5467751966203962, \"sum\": 0.5467751966203962, \"min\": 0.5467751966203962}}, \"EndTime\": 1605486260.288192, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288169}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5620956584385464, \"sum\": 0.5620956584385464, \"min\": 0.5620956584385464}}, \"EndTime\": 1605486260.288264, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288245}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5474806104387556, \"sum\": 0.5474806104387556, \"min\": 0.5474806104387556}}, \"EndTime\": 1605486260.288333, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288314}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5064070374625069, \"sum\": 0.5064070374625069, \"min\": 0.5064070374625069}}, \"EndTime\": 1605486260.288415, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288395}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5065335355486189, \"sum\": 0.5065335355486189, \"min\": 0.5065335355486189}}, \"EndTime\": 1605486260.28848, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288464}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5064097104753766, \"sum\": 0.5064097104753766, \"min\": 0.5064097104753766}}, \"EndTime\": 1605486260.28855, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288534}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.50651031766619, \"sum\": 0.50651031766619, \"min\": 0.50651031766619}}, \"EndTime\": 1605486260.288613, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288595}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5589762605939592, \"sum\": 0.5589762605939592, \"min\": 0.5589762605939592}}, \"EndTime\": 1605486260.288682, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288664}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5501413726806641, \"sum\": 0.5501413726806641, \"min\": 0.5501413726806641}}, \"EndTime\": 1605486260.288769, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.28874}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5487509836469378, \"sum\": 0.5487509836469378, \"min\": 0.5487509836469378}}, \"EndTime\": 1605486260.288836, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288818}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.554435670035226, \"sum\": 0.554435670035226, \"min\": 0.554435670035226}}, \"EndTime\": 1605486260.28891, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288892}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5074986784798758, \"sum\": 0.5074986784798758, \"min\": 0.5074986784798758}}, \"EndTime\": 1605486260.288974, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.288956}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5075444085257393, \"sum\": 0.5075444085257393, \"min\": 0.5075444085257393}}, \"EndTime\": 1605486260.289036, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289019}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5075264576503209, \"sum\": 0.5075264576503209, \"min\": 0.5075264576503209}}, \"EndTime\": 1605486260.289133, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289119}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5075351170131138, \"sum\": 0.5075351170131138, \"min\": 0.5075351170131138}}, \"EndTime\": 1605486260.289198, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289181}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5853144454956055, \"sum\": 0.5853144454956055, \"min\": 0.5853144454956055}}, \"EndTime\": 1605486260.289263, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289246}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.58596770967756, \"sum\": 0.58596770967756, \"min\": 0.58596770967756}}, \"EndTime\": 1605486260.289323, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289306}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5851368059430804, \"sum\": 0.5851368059430804, \"min\": 0.5851368059430804}}, \"EndTime\": 1605486260.289384, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289367}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.585536937713623, \"sum\": 0.585536937713623, \"min\": 0.585536937713623}}, \"EndTime\": 1605486260.289453, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289435}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5982324627467565, \"sum\": 0.5982324627467565, \"min\": 0.5982324627467565}}, \"EndTime\": 1605486260.289551, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289531}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5949019731794085, \"sum\": 0.5949019731794085, \"min\": 0.5949019731794085}}, \"EndTime\": 1605486260.289615, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289597}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5976986231122698, \"sum\": 0.5976986231122698, \"min\": 0.5976986231122698}}, \"EndTime\": 1605486260.289672, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289656}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.594764883858817, \"sum\": 0.594764883858817, \"min\": 0.594764883858817}}, \"EndTime\": 1605486260.289746, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289727}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6120860753740583, \"sum\": 0.6120860753740583, \"min\": 0.6120860753740583}}, \"EndTime\": 1605486260.289821, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289802}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6120792089189802, \"sum\": 0.6120792089189802, \"min\": 0.6120792089189802}}, \"EndTime\": 1605486260.289898, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289878}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6120626531328474, \"sum\": 0.6120626531328474, \"min\": 0.6120626531328474}}, \"EndTime\": 1605486260.289961, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.289945}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6121136828831264, \"sum\": 0.6121136828831264, \"min\": 0.6121136828831264}}, \"EndTime\": 1605486260.290019, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.290003}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6286861147199359, \"sum\": 0.6286861147199359, \"min\": 0.6286861147199359}}, \"EndTime\": 1605486260.290091, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.290072}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6260309192112514, \"sum\": 0.6260309192112514, \"min\": 0.6260309192112514}}, \"EndTime\": 1605486260.290155, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.290137}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6290051405770438, \"sum\": 0.6290051405770438, \"min\": 0.6290051405770438}}, \"EndTime\": 1605486260.290215, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.290199}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6262182807922363, \"sum\": 0.6262182807922363, \"min\": 0.6262182807922363}}, \"EndTime\": 1605486260.290274, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486260.290257}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] #quality_metric: host=algo-1, epoch=5, train binary_classification_cross_entropy_objective =0.551011526925\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=5, criteria=binary_classification_cross_entropy_objective, value=0.506407037463\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] Saving model for epoch: 5\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] Saved checkpoint to \"/tmp/tmpyaN9LB/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] #progress_metric: host=algo-1, completed 40 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 106, \"sum\": 106.0, \"min\": 106}, \"Total Records Seen\": {\"count\": 1, \"max\": 20962, \"sum\": 20962.0, \"min\": 20962}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}}, \"EndTime\": 1605486260.296552, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605486259.886149}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=7225.22917411 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:20.686] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 16, \"duration\": 389, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5453344944545201, \"sum\": 0.5453344944545201, \"min\": 0.5453344944545201}}, \"EndTime\": 1605486260.686481, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.686405}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5410564231872559, \"sum\": 0.5410564231872559, \"min\": 0.5410564231872559}}, \"EndTime\": 1605486260.686588, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.686565}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5554916327340262, \"sum\": 0.5554916327340262, \"min\": 0.5554916327340262}}, \"EndTime\": 1605486260.68666, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.686641}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.541707763671875, \"sum\": 0.541707763671875, \"min\": 0.541707763671875}}, \"EndTime\": 1605486260.686707, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.686693}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5065479169573103, \"sum\": 0.5065479169573103, \"min\": 0.5065479169573103}}, \"EndTime\": 1605486260.686768, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.68675}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5067241150992258, \"sum\": 0.5067241150992258, \"min\": 0.5067241150992258}}, \"EndTime\": 1605486260.686832, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.686814}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5065507452828544, \"sum\": 0.5065507452828544, \"min\": 0.5065507452828544}}, \"EndTime\": 1605486260.686906, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.686887}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5067158944266184, \"sum\": 0.5067158944266184, \"min\": 0.5067158944266184}}, \"EndTime\": 1605486260.686974, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.686956}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5527623340061733, \"sum\": 0.5527623340061733, \"min\": 0.5527623340061733}}, \"EndTime\": 1605486260.68703, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687013}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5441873577662877, \"sum\": 0.5441873577662877, \"min\": 0.5441873577662877}}, \"EndTime\": 1605486260.687095, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687077}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5434700775146485, \"sum\": 0.5434700775146485, \"min\": 0.5434700775146485}}, \"EndTime\": 1605486260.687158, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687141}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5482183183942523, \"sum\": 0.5482183183942523, \"min\": 0.5482183183942523}}, \"EndTime\": 1605486260.687232, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687215}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5075705364772252, \"sum\": 0.5075705364772252, \"min\": 0.5075705364772252}}, \"EndTime\": 1605486260.687294, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687276}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5077824565342495, \"sum\": 0.5077824565342495, \"min\": 0.5077824565342495}}, \"EndTime\": 1605486260.687353, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687335}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5075893947056361, \"sum\": 0.5075893947056361, \"min\": 0.5075893947056361}}, \"EndTime\": 1605486260.687415, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687397}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5077697454180037, \"sum\": 0.5077697454180037, \"min\": 0.5077697454180037}}, \"EndTime\": 1605486260.687475, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687457}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5856521442958287, \"sum\": 0.5856521442958287, \"min\": 0.5856521442958287}}, \"EndTime\": 1605486260.687531, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687515}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5856385612487793, \"sum\": 0.5856385612487793, \"min\": 0.5856385612487793}}, \"EndTime\": 1605486260.687585, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687571}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.585436943599156, \"sum\": 0.585436943599156, \"min\": 0.585436943599156}}, \"EndTime\": 1605486260.687638, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687623}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.585300339290074, \"sum\": 0.585300339290074, \"min\": 0.585300339290074}}, \"EndTime\": 1605486260.687699, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687681}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5980607632228306, \"sum\": 0.5980607632228306, \"min\": 0.5980607632228306}}, \"EndTime\": 1605486260.687765, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687747}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5937113680158342, \"sum\": 0.5937113680158342, \"min\": 0.5937113680158342}}, \"EndTime\": 1605486260.687828, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687811}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5975832285199847, \"sum\": 0.5975832285199847, \"min\": 0.5975832285199847}}, \"EndTime\": 1605486260.687886, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687871}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5935659790039063, \"sum\": 0.5935659790039063, \"min\": 0.5935659790039063}}, \"EndTime\": 1605486260.687944, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687926}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119719995771136, \"sum\": 0.6119719995771136, \"min\": 0.6119719995771136}}, \"EndTime\": 1605486260.688007, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.687988}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119934027535575, \"sum\": 0.6119934027535575, \"min\": 0.6119934027535575}}, \"EndTime\": 1605486260.688068, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.68805}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119577271597726, \"sum\": 0.6119577271597726, \"min\": 0.6119577271597726}}, \"EndTime\": 1605486260.688128, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.688111}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119791276114328, \"sum\": 0.6119791276114328, \"min\": 0.6119791276114328}}, \"EndTime\": 1605486260.688188, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.688171}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.629653252192906, \"sum\": 0.629653252192906, \"min\": 0.629653252192906}}, \"EndTime\": 1605486260.688249, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.688232}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6255613463265556, \"sum\": 0.6255613463265556, \"min\": 0.6255613463265556}}, \"EndTime\": 1605486260.688292, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.688282}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6299737903050014, \"sum\": 0.6299737903050014, \"min\": 0.6299737903050014}}, \"EndTime\": 1605486260.688347, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.688331}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.625832315172468, \"sum\": 0.625832315172468, \"min\": 0.625832315172468}}, \"EndTime\": 1605486260.688399, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.688388}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] #quality_metric: host=algo-1, epoch=6, train binary_classification_cross_entropy_objective =0.545334494455\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=6, criteria=binary_classification_cross_entropy_objective, value=0.506547916957\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] Saving model for epoch: 6\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] Saved checkpoint to \"/tmp/tmpjnHjCc/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] #progress_metric: host=algo-1, completed 46 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 121, \"sum\": 121.0, \"min\": 121}, \"Total Records Seen\": {\"count\": 1, \"max\": 23928, \"sum\": 23928.0, \"min\": 23928}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 9, \"sum\": 9.0, \"min\": 9}}, \"EndTime\": 1605486260.696042, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605486260.296825}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:20 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=7427.08806435 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:21.106] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 18, \"duration\": 410, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5405281530107771, \"sum\": 0.5405281530107771, \"min\": 0.5405281530107771}}, \"EndTime\": 1605486261.106918, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.106831}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5362567111424037, \"sum\": 0.5362567111424037, \"min\": 0.5362567111424037}}, \"EndTime\": 1605486261.107011, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.106989}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5498522186279297, \"sum\": 0.5498522186279297, \"min\": 0.5498522186279297}}, \"EndTime\": 1605486261.107083, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107064}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5368561254228864, \"sum\": 0.5368561254228864, \"min\": 0.5368561254228864}}, \"EndTime\": 1605486261.10714, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107128}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.506512508392334, \"sum\": 0.506512508392334, \"min\": 0.506512508392334}}, \"EndTime\": 1605486261.107196, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107181}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5066240392412458, \"sum\": 0.5066240392412458, \"min\": 0.5066240392412458}}, \"EndTime\": 1605486261.107256, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.10724}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5065147835867746, \"sum\": 0.5065147835867746, \"min\": 0.5065147835867746}}, \"EndTime\": 1605486261.107331, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107313}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5066209874834333, \"sum\": 0.5066209874834333, \"min\": 0.5066209874834333}}, \"EndTime\": 1605486261.107395, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107383}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5474716322762626, \"sum\": 0.5474716322762626, \"min\": 0.5474716322762626}}, \"EndTime\": 1605486261.107454, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107436}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.539199275970459, \"sum\": 0.539199275970459, \"min\": 0.539199275970459}}, \"EndTime\": 1605486261.10752, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107502}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.538999195098877, \"sum\": 0.538999195098877, \"min\": 0.538999195098877}}, \"EndTime\": 1605486261.107593, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107576}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5429511860438756, \"sum\": 0.5429511860438756, \"min\": 0.5429511860438756}}, \"EndTime\": 1605486261.107655, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107638}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5076573780604772, \"sum\": 0.5076573780604772, \"min\": 0.5076573780604772}}, \"EndTime\": 1605486261.107715, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107699}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5078389685494559, \"sum\": 0.5078389685494559, \"min\": 0.5078389685494559}}, \"EndTime\": 1605486261.107749, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107741}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5076405906677246, \"sum\": 0.5076405906677246, \"min\": 0.5076405906677246}}, \"EndTime\": 1605486261.107778, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.10777}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5078211157662528, \"sum\": 0.5078211157662528, \"min\": 0.5078211157662528}}, \"EndTime\": 1605486261.107822, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107806}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5857111522129603, \"sum\": 0.5857111522129603, \"min\": 0.5857111522129603}}, \"EndTime\": 1605486261.107896, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107876}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5855745070321219, \"sum\": 0.5855745070321219, \"min\": 0.5855745070321219}}, \"EndTime\": 1605486261.107963, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.107946}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5856265504019601, \"sum\": 0.5856265504019601, \"min\": 0.5856265504019601}}, \"EndTime\": 1605486261.108024, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108007}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5853996113368443, \"sum\": 0.5853996113368443, \"min\": 0.5853996113368443}}, \"EndTime\": 1605486261.108087, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.10807}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5981342233930316, \"sum\": 0.5981342233930316, \"min\": 0.5981342233930316}}, \"EndTime\": 1605486261.10815, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108133}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.592819687979562, \"sum\": 0.592819687979562, \"min\": 0.592819687979562}}, \"EndTime\": 1605486261.10821, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108194}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5977781432015555, \"sum\": 0.5977781432015555, \"min\": 0.5977781432015555}}, \"EndTime\": 1605486261.108273, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108256}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5926365879603794, \"sum\": 0.5926365879603794, \"min\": 0.5926365879603794}}, \"EndTime\": 1605486261.108334, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108317}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119528198242188, \"sum\": 0.6119528198242188, \"min\": 0.6119528198242188}}, \"EndTime\": 1605486261.108402, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108384}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119704273768833, \"sum\": 0.6119704273768833, \"min\": 0.6119704273768833}}, \"EndTime\": 1605486261.108466, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108448}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119587598528181, \"sum\": 0.6119587598528181, \"min\": 0.6119587598528181}}, \"EndTime\": 1605486261.108527, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.10851}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119660295758929, \"sum\": 0.6119660295758929, \"min\": 0.6119660295758929}}, \"EndTime\": 1605486261.108589, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108573}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6307636451721191, \"sum\": 0.6307636451721191, \"min\": 0.6307636451721191}}, \"EndTime\": 1605486261.108651, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108634}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6251236234392439, \"sum\": 0.6251236234392439, \"min\": 0.6251236234392439}}, \"EndTime\": 1605486261.108713, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108695}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6310931995936803, \"sum\": 0.6310931995936803, \"min\": 0.6310931995936803}}, \"EndTime\": 1605486261.108774, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.108756}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6254582895551409, \"sum\": 0.6254582895551409, \"min\": 0.6254582895551409}}, \"EndTime\": 1605486261.108836, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486261.10882}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #quality_metric: host=algo-1, epoch=7, train binary_classification_cross_entropy_objective =0.540528153011\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=7, criteria=binary_classification_cross_entropy_objective, value=0.506512508392\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] Saving model for epoch: 7\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] Saved checkpoint to \"/tmp/tmpG8PzmN/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #progress_metric: host=algo-1, completed 53 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 136, \"sum\": 136.0, \"min\": 136}, \"Total Records Seen\": {\"count\": 1, \"max\": 26894, \"sum\": 26894.0, \"min\": 26894}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}}, \"EndTime\": 1605486261.116293, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605486260.696332}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=7060.30834401 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:21.515] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 20, \"duration\": 398, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5364204488481794, \"sum\": 0.5364204488481794, \"min\": 0.5364204488481794}}, \"EndTime\": 1605486261.515707, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.515644}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5321960067749023, \"sum\": 0.5321960067749023, \"min\": 0.5321960067749023}}, \"EndTime\": 1605486261.515771, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.515759}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.544999316079276, \"sum\": 0.544999316079276, \"min\": 0.544999316079276}}, \"EndTime\": 1605486261.515824, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.515808}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5327464975629534, \"sum\": 0.5327464975629534, \"min\": 0.5327464975629534}}, \"EndTime\": 1605486261.51588, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.515868}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5064929389953613, \"sum\": 0.5064929389953613, \"min\": 0.5064929389953613}}, \"EndTime\": 1605486261.515913, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.515905}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5067237799508231, \"sum\": 0.5067237799508231, \"min\": 0.5067237799508231}}, \"EndTime\": 1605486261.515962, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.51595}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5064929417201451, \"sum\": 0.5064929417201451, \"min\": 0.5064929417201451}}, \"EndTime\": 1605486261.515994, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.515986}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.506703987121582, \"sum\": 0.506703987121582, \"min\": 0.506703987121582}}, \"EndTime\": 1605486261.516022, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516014}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.542930120740618, \"sum\": 0.542930120740618, \"min\": 0.542930120740618}}, \"EndTime\": 1605486261.516058, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516044}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5349874768938337, \"sum\": 0.5349874768938337, \"min\": 0.5349874768938337}}, \"EndTime\": 1605486261.516104, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516094}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5351787648882185, \"sum\": 0.5351787648882185, \"min\": 0.5351787648882185}}, \"EndTime\": 1605486261.516133, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516126}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5384606197902134, \"sum\": 0.5384606197902134, \"min\": 0.5384606197902134}}, \"EndTime\": 1605486261.51616, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516153}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5076615442548479, \"sum\": 0.5076615442548479, \"min\": 0.5076615442548479}}, \"EndTime\": 1605486261.516209, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516193}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.507916979108538, \"sum\": 0.507916979108538, \"min\": 0.507916979108538}}, \"EndTime\": 1605486261.516243, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516235}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5076749965122768, \"sum\": 0.5076749965122768, \"min\": 0.5076749965122768}}, \"EndTime\": 1605486261.516271, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516264}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5079068810599191, \"sum\": 0.5079068810599191, \"min\": 0.5079068810599191}}, \"EndTime\": 1605486261.516298, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.51629}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5856681796482631, \"sum\": 0.5856681796482631, \"min\": 0.5856681796482631}}, \"EndTime\": 1605486261.516347, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516336}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5856661469595773, \"sum\": 0.5856661469595773, \"min\": 0.5856661469595773}}, \"EndTime\": 1605486261.516388, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516381}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5856825038364956, \"sum\": 0.5856825038364956, \"min\": 0.5856825038364956}}, \"EndTime\": 1605486261.516414, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516407}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5855786378043039, \"sum\": 0.5855786378043039, \"min\": 0.5855786378043039}}, \"EndTime\": 1605486261.516455, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.51644}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5983882468087333, \"sum\": 0.5983882468087333, \"min\": 0.5983882468087333}}, \"EndTime\": 1605486261.516491, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516483}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5924447849818638, \"sum\": 0.5924447849818638, \"min\": 0.5924447849818638}}, \"EndTime\": 1605486261.516518, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516511}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5981591061183384, \"sum\": 0.5981591061183384, \"min\": 0.5981591061183384}}, \"EndTime\": 1605486261.516544, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516537}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.592227930341448, \"sum\": 0.592227930341448, \"min\": 0.592227930341448}}, \"EndTime\": 1605486261.51659, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516575}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119686317443848, \"sum\": 0.6119686317443848, \"min\": 0.6119686317443848}}, \"EndTime\": 1605486261.516622, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516614}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6120030757359096, \"sum\": 0.6120030757359096, \"min\": 0.6120030757359096}}, \"EndTime\": 1605486261.516649, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516642}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6119601767403738, \"sum\": 0.6119601767403738, \"min\": 0.6119601767403738}}, \"EndTime\": 1605486261.516675, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516668}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.611997378213065, \"sum\": 0.611997378213065, \"min\": 0.611997378213065}}, \"EndTime\": 1605486261.516722, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516709}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6317872047424317, \"sum\": 0.6317872047424317, \"min\": 0.6317872047424317}}, \"EndTime\": 1605486261.516752, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516745}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6250876944405692, \"sum\": 0.6250876944405692, \"min\": 0.6250876944405692}}, \"EndTime\": 1605486261.516778, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516771}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6321233722141811, \"sum\": 0.6321233722141811, \"min\": 0.6321233722141811}}, \"EndTime\": 1605486261.516804, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516797}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6254377937316895, \"sum\": 0.6254377937316895, \"min\": 0.6254377937316895}}, \"EndTime\": 1605486261.516858, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.516847}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #quality_metric: host=algo-1, epoch=8, train binary_classification_cross_entropy_objective =0.536420448848\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #early_stopping_criteria_metric: host=algo-1, epoch=8, criteria=binary_classification_cross_entropy_objective, value=0.506492938995\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] Saving model for epoch: 8\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] Saved checkpoint to \"/tmp/tmpnRjQWA/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] Early stop condition met. Stopping training.\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #progress_metric: host=algo-1, completed 100 % epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Total Batches Seen\": {\"count\": 1, \"max\": 151, \"sum\": 151.0, \"min\": 151}, \"Total Records Seen\": {\"count\": 1, \"max\": 29860, \"sum\": 29860.0, \"min\": 29860}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2966, \"sum\": 2966.0, \"min\": 2966}, \"Reset Count\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}}, \"EndTime\": 1605486261.523686, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1605486261.116572}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #throughput_metric: host=algo-1, train throughput=7283.74749203 records/second\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 WARNING 139985910667072] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 WARNING 139985910667072] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:21.525] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 22, \"duration\": 1, \"num_examples\": 1, \"num_bytes\": 11200}\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:21.640] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 25, \"duration\": 112, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m[2020-11-16 00:24:21.689] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 27, \"duration\": 38, \"num_examples\": 15, \"num_bytes\": 166096}\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #train_score (algo-1) : ('binary_classification_cross_entropy_objective', 0.509484489785604)\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #train_score (algo-1) : ('binary_classification_accuracy', 0.7356709372892785)\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #train_score (algo-1) : ('binary_f_1.000', 0.8239784463403682)\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #train_score (algo-1) : ('precision', 0.7700377675199328)\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #train_score (algo-1) : ('recall', 0.8860453887011106)\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #train_score (algo-1) : ('roc_auc_score', 0.7687978980817839)\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #quality_metric: host=algo-1, train binary_classification_cross_entropy_objective =0.509484489786\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #quality_metric: host=algo-1, train binary_classification_accuracy =0.735670937289\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #quality_metric: host=algo-1, train binary_f_1.000 =0.82397844634\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #quality_metric: host=algo-1, train precision =0.77003776752\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #quality_metric: host=algo-1, train recall =0.886045388701\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] #quality_metric: host=algo-1, train roc_auc_score =0.768797898082\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] Best model found for hyperparameters: {\"lr_scheduler_step\": 100, \"wd\": 0.0001, \"optimizer\": \"adam\", \"lr_scheduler_factor\": 0.99, \"l1\": 0.0, \"learning_rate\": 0.1, \"lr_scheduler_minimum_lr\": 1e-05}\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] Saved checkpoint to \"/tmp/tmpjPpA98/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:24:21 INFO 139985910667072] Test data is not provided.\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"totaltime\": {\"count\": 1, \"max\": 4916.153907775879, \"sum\": 4916.153907775879, \"min\": 4916.153907775879}, \"finalize.time\": {\"count\": 1, \"max\": 171.47183418273926, \"sum\": 171.47183418273926, \"min\": 171.47183418273926}, \"initialize.time\": {\"count\": 1, \"max\": 205.7950496673584, \"sum\": 205.7950496673584, \"min\": 205.7950496673584}, \"check_early_stopping.time\": {\"count\": 9, \"max\": 1.2080669403076172, \"sum\": 5.861043930053711, \"min\": 0.1728534698486328}, \"setuptime\": {\"count\": 1, \"max\": 34.42096710205078, \"sum\": 34.42096710205078, \"min\": 34.42096710205078}, \"update.time\": {\"count\": 9, \"max\": 586.655855178833, \"sum\": 4246.23966217041, \"min\": 396.95191383361816}, \"epochs\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}}, \"EndTime\": 1605486261.700291, \"Dimensions\": {\"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1605486257.040859}\n", "\u001b[0m\n", "\n", "2020-11-16 00:24:30 Uploading - Uploading generated training model\n", "2020-11-16 00:24:30 Completed - Training job completed\n", "Training seconds: 70\n", "Billable seconds: 70\n", "-----------------!2020-11-16 00:33:33 Starting - Starting the training job...\n", "2020-11-16 00:33:36 Starting - Launching requested ML instances.........\n", "2020-11-16 00:35:07 Starting - Preparing the instances for training......\n", "2020-11-16 00:36:25 Downloading - Downloading input data...\n", "2020-11-16 00:37:02 Training - Training image download completed. Training in progress..\u001b[34mDocker entrypoint called with argument(s): train\u001b[0m\n", "\u001b[34mRunning default environment configuration script\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Reading default configuration from /opt/amazon/lib/python2.7/site-packages/algorithm/resources/default-input.json: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'auto', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'1000', u'huber_delta': u'1.0', u'num_classes': u'1', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Merging with provided configuration from /opt/ml/input/config/hyperparameters.json: {u'feature_dim': u'4', u'mini_batch_size': u'200', u'predictor_type': u'binary_classifier'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Final configuration: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'4', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'200', u'huber_delta': u'1.0', u'num_classes': u'1', u'predictor_type': u'binary_classifier', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 WARNING 140309854623552] Loggers have already been setup.\u001b[0m\n", "\u001b[34mProcess 1 is a worker.\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Using default worker.\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Checkpoint loading and saving are disabled.\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:07.569] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 0, \"duration\": 15, \"num_examples\": 1, \"num_bytes\": 12000}\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Create Store: local\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:07.635] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 1, \"duration\": 65, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Scaler algorithm parameters\n", " \u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Scaling model computed with parameters:\n", " {'stdev_weight': \u001b[0m\n", "\u001b[34m[3.7846494 3.6940358 5.127486 2.5804543]\u001b[0m\n", "\u001b[34m, 'stdev_label': None, 'mean_label': None, 'mean_weight': \u001b[0m\n", "\u001b[34m[22.776623 22.476126 15.355499 3.1725004]\u001b[0m\n", "\u001b[34m}\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] nvidia-smi took: 0.0252130031586 secs to identify 0 gpus\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:07 INFO 140309854623552] Number of GPUs being used: 0\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Total Batches Seen\": {\"count\": 1, \"max\": 12, \"sum\": 12.0, \"min\": 12}, \"Total Records Seen\": {\"count\": 1, \"max\": 2287, \"sum\": 2287.0, \"min\": 2287}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 2, \"sum\": 2.0, \"min\": 2}}, \"EndTime\": 1605487027.734462, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"init_train_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1605487027.734425}\n", "\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:08.080] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 4, \"duration\": 345, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6556354675292969, \"sum\": 0.6556354675292969, \"min\": 0.6556354675292969}}, \"EndTime\": 1605487028.080802, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.080704}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6588098678588867, \"sum\": 0.6588098678588867, \"min\": 0.6588098678588867}}, \"EndTime\": 1605487028.080908, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.080889}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6739308242797851, \"sum\": 0.6739308242797851, \"min\": 0.6739308242797851}}, \"EndTime\": 1605487028.080974, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.08096}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6688716354370117, \"sum\": 0.6688716354370117, \"min\": 0.6688716354370117}}, \"EndTime\": 1605487028.081032, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081013}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5968777809143067, \"sum\": 0.5968777809143067, \"min\": 0.5968777809143067}}, \"EndTime\": 1605487028.081112, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081091}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5905601577758789, \"sum\": 0.5905601577758789, \"min\": 0.5905601577758789}}, \"EndTime\": 1605487028.081172, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081154}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5982935066223145, \"sum\": 0.5982935066223145, \"min\": 0.5982935066223145}}, \"EndTime\": 1605487028.081232, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081215}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5949179801940918, \"sum\": 0.5949179801940918, \"min\": 0.5949179801940918}}, \"EndTime\": 1605487028.081293, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081276}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.678425407409668, \"sum\": 0.678425407409668, \"min\": 0.678425407409668}}, \"EndTime\": 1605487028.081351, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081339}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6720165328979493, \"sum\": 0.6720165328979493, \"min\": 0.6720165328979493}}, \"EndTime\": 1605487028.081416, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081403}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6594603843688965, \"sum\": 0.6594603843688965, \"min\": 0.6594603843688965}}, \"EndTime\": 1605487028.081483, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081465}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6700916366577149, \"sum\": 0.6700916366577149, \"min\": 0.6700916366577149}}, \"EndTime\": 1605487028.081529, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081519}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5956535873413086, \"sum\": 0.5956535873413086, \"min\": 0.5956535873413086}}, \"EndTime\": 1605487028.081582, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081565}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5956762008666993, \"sum\": 0.5956762008666993, \"min\": 0.5956762008666993}}, \"EndTime\": 1605487028.081643, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081631}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5932641105651856, \"sum\": 0.5932641105651856, \"min\": 0.5932641105651856}}, \"EndTime\": 1605487028.0817, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081682}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5819111251831055, \"sum\": 0.5819111251831055, \"min\": 0.5819111251831055}}, \"EndTime\": 1605487028.081775, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081756}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6801255340576172, \"sum\": 0.6801255340576172, \"min\": 0.6801255340576172}}, \"EndTime\": 1605487028.081849, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081831}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6538012771606445, \"sum\": 0.6538012771606445, \"min\": 0.6538012771606445}}, \"EndTime\": 1605487028.081921, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081903}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6520976142883301, \"sum\": 0.6520976142883301, \"min\": 0.6520976142883301}}, \"EndTime\": 1605487028.081993, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.081975}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6542397117614747, \"sum\": 0.6542397117614747, \"min\": 0.6542397117614747}}, \"EndTime\": 1605487028.082049, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082034}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6415051612854004, \"sum\": 0.6415051612854004, \"min\": 0.6415051612854004}}, \"EndTime\": 1605487028.082108, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082091}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6435424270629883, \"sum\": 0.6435424270629883, \"min\": 0.6435424270629883}}, \"EndTime\": 1605487028.082162, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082148}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6404759407043457, \"sum\": 0.6404759407043457, \"min\": 0.6404759407043457}}, \"EndTime\": 1605487028.08221, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082197}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6425398597717286, \"sum\": 0.6425398597717286, \"min\": 0.6425398597717286}}, \"EndTime\": 1605487028.082272, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082256}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6851014633178711, \"sum\": 0.6851014633178711, \"min\": 0.6851014633178711}}, \"EndTime\": 1605487028.082331, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082314}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6867370986938477, \"sum\": 0.6867370986938477, \"min\": 0.6867370986938477}}, \"EndTime\": 1605487028.082402, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082382}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6791136703491211, \"sum\": 0.6791136703491211, \"min\": 0.6791136703491211}}, \"EndTime\": 1605487028.082478, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082459}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6756955490112305, \"sum\": 0.6756955490112305, \"min\": 0.6756955490112305}}, \"EndTime\": 1605487028.082549, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082531}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7083576431274414, \"sum\": 0.7083576431274414, \"min\": 0.7083576431274414}}, \"EndTime\": 1605487028.082623, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082603}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7077159423828125, \"sum\": 0.7077159423828125, \"min\": 0.7077159423828125}}, \"EndTime\": 1605487028.082698, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082678}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7068484497070312, \"sum\": 0.7068484497070312, \"min\": 0.7068484497070312}}, \"EndTime\": 1605487028.082768, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082751}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7083702697753906, \"sum\": 0.7083702697753906, \"min\": 0.7083702697753906}}, \"EndTime\": 1605487028.08283, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487028.082812}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #quality_metric: host=algo-1, epoch=0, train binary_classification_cross_entropy_objective =0.655635467529\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #early_stopping_criteria_metric: host=algo-1, epoch=0, criteria=binary_classification_cross_entropy_objective, value=0.581911125183\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Epoch 0: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Saving model for epoch: 0\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Saved checkpoint to \"/tmp/tmpwNSx_S/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #progress_metric: host=algo-1, completed 6 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Total Batches Seen\": {\"count\": 1, \"max\": 23, \"sum\": 23.0, \"min\": 23}, \"Total Records Seen\": {\"count\": 1, \"max\": 4374, \"sum\": 4374.0, \"min\": 4374}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 3, \"sum\": 3.0, \"min\": 3}}, \"EndTime\": 1605487028.093806, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487027.734692}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #throughput_metric: host=algo-1, train throughput=5809.59761816 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:08.383] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 6, \"duration\": 289, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6341747093200684, \"sum\": 0.6341747093200684, \"min\": 0.6341747093200684}}, \"EndTime\": 1605487028.38354, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.383444}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6369302711486816, \"sum\": 0.6369302711486816, \"min\": 0.6369302711486816}}, \"EndTime\": 1605487028.383637, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.383616}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6509008255004883, \"sum\": 0.6509008255004883, \"min\": 0.6509008255004883}}, \"EndTime\": 1605487028.383684, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.383673}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6460561180114746, \"sum\": 0.6460561180114746, \"min\": 0.6460561180114746}}, \"EndTime\": 1605487028.383724, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.383709}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5317274436950684, \"sum\": 0.5317274436950684, \"min\": 0.5317274436950684}}, \"EndTime\": 1605487028.383787, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.383768}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5302793502807617, \"sum\": 0.5302793502807617, \"min\": 0.5302793502807617}}, \"EndTime\": 1605487028.383851, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.383833}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5323576507568359, \"sum\": 0.5323576507568359, \"min\": 0.5323576507568359}}, \"EndTime\": 1605487028.383951, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.383929}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5314966812133789, \"sum\": 0.5314966812133789, \"min\": 0.5314966812133789}}, \"EndTime\": 1605487028.38402, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384001}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6549221000671387, \"sum\": 0.6549221000671387, \"min\": 0.6549221000671387}}, \"EndTime\": 1605487028.384081, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384065}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6492220115661621, \"sum\": 0.6492220115661621, \"min\": 0.6492220115661621}}, \"EndTime\": 1605487028.384139, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384123}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6376742401123047, \"sum\": 0.6376742401123047, \"min\": 0.6376742401123047}}, \"EndTime\": 1605487028.384192, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384178}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6472092132568359, \"sum\": 0.6472092132568359, \"min\": 0.6472092132568359}}, \"EndTime\": 1605487028.384274, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384255}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5334305915832519, \"sum\": 0.5334305915832519, \"min\": 0.5334305915832519}}, \"EndTime\": 1605487028.384349, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384331}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.531789005279541, \"sum\": 0.531789005279541, \"min\": 0.531789005279541}}, \"EndTime\": 1605487028.384413, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384396}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5320266609191895, \"sum\": 0.5320266609191895, \"min\": 0.5320266609191895}}, \"EndTime\": 1605487028.384491, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.38447}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5300234260559082, \"sum\": 0.5300234260559082, \"min\": 0.5300234260559082}}, \"EndTime\": 1605487028.384567, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384548}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6571204643249512, \"sum\": 0.6571204643249512, \"min\": 0.6571204643249512}}, \"EndTime\": 1605487028.384643, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384623}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6341899871826172, \"sum\": 0.6341899871826172, \"min\": 0.6341899871826172}}, \"EndTime\": 1605487028.384707, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.38469}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6334832382202148, \"sum\": 0.6334832382202148, \"min\": 0.6334832382202148}}, \"EndTime\": 1605487028.384768, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384752}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6362813568115234, \"sum\": 0.6362813568115234, \"min\": 0.6362813568115234}}, \"EndTime\": 1605487028.384832, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384815}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6741529159545898, \"sum\": 0.6741529159545898, \"min\": 0.6741529159545898}}, \"EndTime\": 1605487028.384892, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384874}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.667632095336914, \"sum\": 0.667632095336914, \"min\": 0.667632095336914}}, \"EndTime\": 1605487028.384953, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384935}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6779353981018066, \"sum\": 0.6779353981018066, \"min\": 0.6779353981018066}}, \"EndTime\": 1605487028.385014, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.384997}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6675700988769532, \"sum\": 0.6675700988769532, \"min\": 0.6675700988769532}}, \"EndTime\": 1605487028.385077, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.38506}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6734340934753418, \"sum\": 0.6734340934753418, \"min\": 0.6734340934753418}}, \"EndTime\": 1605487028.385135, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.385118}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.673029010772705, \"sum\": 0.673029010772705, \"min\": 0.673029010772705}}, \"EndTime\": 1605487028.385193, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.385178}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6743097114562988, \"sum\": 0.6743097114562988, \"min\": 0.6743097114562988}}, \"EndTime\": 1605487028.385231, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.385223}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6760931205749512, \"sum\": 0.6760931205749512, \"min\": 0.6760931205749512}}, \"EndTime\": 1605487028.385267, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.385253}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7344856491088867, \"sum\": 0.7344856491088867, \"min\": 0.7344856491088867}}, \"EndTime\": 1605487028.385326, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.38531}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.734478988647461, \"sum\": 0.734478988647461, \"min\": 0.734478988647461}}, \"EndTime\": 1605487028.385372, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.385357}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7330490264892578, \"sum\": 0.7330490264892578, \"min\": 0.7330490264892578}}, \"EndTime\": 1605487028.38543, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.385414}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7341711730957031, \"sum\": 0.7341711730957031, \"min\": 0.7341711730957031}}, \"EndTime\": 1605487028.385484, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.385468}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #quality_metric: host=algo-1, epoch=1, train binary_classification_cross_entropy_objective =0.63417470932\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #early_stopping_criteria_metric: host=algo-1, epoch=1, criteria=binary_classification_cross_entropy_objective, value=0.530023426056\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Epoch 1: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Saving model for epoch: 1\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Saved checkpoint to \"/tmp/tmpAkaZTe/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #progress_metric: host=algo-1, completed 13 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Total Batches Seen\": {\"count\": 1, \"max\": 34, \"sum\": 34.0, \"min\": 34}, \"Total Records Seen\": {\"count\": 1, \"max\": 6461, \"sum\": 6461.0, \"min\": 6461}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 4, \"sum\": 4.0, \"min\": 4}}, \"EndTime\": 1605487028.39288, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487028.094089}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #throughput_metric: host=algo-1, train throughput=6981.40218673 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:08.752] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 8, \"duration\": 358, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6197612495422363, \"sum\": 0.6197612495422363, \"min\": 0.6197612495422363}}, \"EndTime\": 1605487028.75223, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752142}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6220798377990723, \"sum\": 0.6220798377990723, \"min\": 0.6220798377990723}}, \"EndTime\": 1605487028.752315, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752295}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6356197242736816, \"sum\": 0.6356197242736816, \"min\": 0.6356197242736816}}, \"EndTime\": 1605487028.752379, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.75236}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6305569725036622, \"sum\": 0.6305569725036622, \"min\": 0.6305569725036622}}, \"EndTime\": 1605487028.752446, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752427}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5023296089172363, \"sum\": 0.5023296089172363, \"min\": 0.5023296089172363}}, \"EndTime\": 1605487028.752525, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752506}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5037544059753418, \"sum\": 0.5037544059753418, \"min\": 0.5037544059753418}}, \"EndTime\": 1605487028.752599, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.75258}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5014720535278321, \"sum\": 0.5014720535278321, \"min\": 0.5014720535278321}}, \"EndTime\": 1605487028.752672, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752654}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.502558307647705, \"sum\": 0.502558307647705, \"min\": 0.502558307647705}}, \"EndTime\": 1605487028.752731, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752716}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6388125419616699, \"sum\": 0.6388125419616699, \"min\": 0.6388125419616699}}, \"EndTime\": 1605487028.75279, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752773}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6336985664367676, \"sum\": 0.6336985664367676, \"min\": 0.6336985664367676}}, \"EndTime\": 1605487028.752846, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752833}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6230717544555664, \"sum\": 0.6230717544555664, \"min\": 0.6230717544555664}}, \"EndTime\": 1605487028.752883, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752869}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6317098617553711, \"sum\": 0.6317098617553711, \"min\": 0.6317098617553711}}, \"EndTime\": 1605487028.752952, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752934}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49902207946777344, \"sum\": 0.49902207946777344, \"min\": 0.49902207946777344}}, \"EndTime\": 1605487028.753017, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.752998}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5001858100891113, \"sum\": 0.5001858100891113, \"min\": 0.5001858100891113}}, \"EndTime\": 1605487028.753088, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753069}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5001663856506348, \"sum\": 0.5001663856506348, \"min\": 0.5001663856506348}}, \"EndTime\": 1605487028.753153, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753136}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5014154052734375, \"sum\": 0.5014154052734375, \"min\": 0.5014154052734375}}, \"EndTime\": 1605487028.753204, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753194}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6436052513122559, \"sum\": 0.6436052513122559, \"min\": 0.6436052513122559}}, \"EndTime\": 1605487028.753258, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753243}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6247568893432617, \"sum\": 0.6247568893432617, \"min\": 0.6247568893432617}}, \"EndTime\": 1605487028.753332, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753312}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6250237922668457, \"sum\": 0.6250237922668457, \"min\": 0.6250237922668457}}, \"EndTime\": 1605487028.753394, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753376}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6271777648925781, \"sum\": 0.6271777648925781, \"min\": 0.6271777648925781}}, \"EndTime\": 1605487028.753468, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753449}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.661199031829834, \"sum\": 0.661199031829834, \"min\": 0.661199031829834}}, \"EndTime\": 1605487028.753542, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753523}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6656376914978027, \"sum\": 0.6656376914978027, \"min\": 0.6656376914978027}}, \"EndTime\": 1605487028.753604, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753588}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6572118682861328, \"sum\": 0.6572118682861328, \"min\": 0.6572118682861328}}, \"EndTime\": 1605487028.753662, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753645}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6659173316955567, \"sum\": 0.6659173316955567, \"min\": 0.6659173316955567}}, \"EndTime\": 1605487028.75372, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753703}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6744616661071777, \"sum\": 0.6744616661071777, \"min\": 0.6744616661071777}}, \"EndTime\": 1605487028.753771, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753757}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6743483390808106, \"sum\": 0.6743483390808106, \"min\": 0.6743483390808106}}, \"EndTime\": 1605487028.75384, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753821}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6770909652709961, \"sum\": 0.6770909652709961, \"min\": 0.6770909652709961}}, \"EndTime\": 1605487028.753916, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753896}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6768310165405274, \"sum\": 0.6768310165405274, \"min\": 0.6768310165405274}}, \"EndTime\": 1605487028.753978, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.753961}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7000812530517578, \"sum\": 0.7000812530517578, \"min\": 0.7000812530517578}}, \"EndTime\": 1605487028.75403, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.754019}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6994568481445312, \"sum\": 0.6994568481445312, \"min\": 0.6994568481445312}}, \"EndTime\": 1605487028.754071, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.754056}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6986679611206055, \"sum\": 0.6986679611206055, \"min\": 0.6986679611206055}}, \"EndTime\": 1605487028.754142, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.754122}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7024537887573242, \"sum\": 0.7024537887573242, \"min\": 0.7024537887573242}}, \"EndTime\": 1605487028.754208, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.754191}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #quality_metric: host=algo-1, epoch=2, train binary_classification_cross_entropy_objective =0.619761249542\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #early_stopping_criteria_metric: host=algo-1, epoch=2, criteria=binary_classification_cross_entropy_objective, value=0.499022079468\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Epoch 2: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Saving model for epoch: 2\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] Saved checkpoint to \"/tmp/tmp5e3lZk/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #progress_metric: host=algo-1, completed 20 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Total Batches Seen\": {\"count\": 1, \"max\": 45, \"sum\": 45.0, \"min\": 45}, \"Total Records Seen\": {\"count\": 1, \"max\": 8548, \"sum\": 8548.0, \"min\": 8548}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 5, \"sum\": 5.0, \"min\": 5}}, \"EndTime\": 1605487028.761007, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487028.393184}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:08 INFO 140309854623552] #throughput_metric: host=algo-1, train throughput=5671.93747432 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:09.064] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 10, \"duration\": 302, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6051033363342285, \"sum\": 0.6051033363342285, \"min\": 0.6051033363342285}}, \"EndTime\": 1605487029.064438, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064338}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.606867130279541, \"sum\": 0.606867130279541, \"min\": 0.606867130279541}}, \"EndTime\": 1605487029.064522, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064503}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6200845527648926, \"sum\": 0.6200845527648926, \"min\": 0.6200845527648926}}, \"EndTime\": 1605487029.064577, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064565}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6147160797119141, \"sum\": 0.6147160797119141, \"min\": 0.6147160797119141}}, \"EndTime\": 1605487029.064611, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064603}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49391859817504885, \"sum\": 0.49391859817504885, \"min\": 0.49391859817504885}}, \"EndTime\": 1605487029.064641, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064633}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49341569137573243, \"sum\": 0.49341569137573243, \"min\": 0.49341569137573243}}, \"EndTime\": 1605487029.064669, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064662}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.494058837890625, \"sum\": 0.494058837890625, \"min\": 0.494058837890625}}, \"EndTime\": 1605487029.064716, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064701}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4937565650939941, \"sum\": 0.4937565650939941, \"min\": 0.4937565650939941}}, \"EndTime\": 1605487029.064758, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064749}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6225660514831542, \"sum\": 0.6225660514831542, \"min\": 0.6225660514831542}}, \"EndTime\": 1605487029.064787, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.06478}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6178404541015625, \"sum\": 0.6178404541015625, \"min\": 0.6178404541015625}}, \"EndTime\": 1605487029.064814, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064807}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6082935523986817, \"sum\": 0.6082935523986817, \"min\": 0.6082935523986817}}, \"EndTime\": 1605487029.064842, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064835}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6158878936767578, \"sum\": 0.6158878936767578, \"min\": 0.6158878936767578}}, \"EndTime\": 1605487029.064869, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064862}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4891533546447754, \"sum\": 0.4891533546447754, \"min\": 0.4891533546447754}}, \"EndTime\": 1605487029.0649, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064892}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4886359214782715, \"sum\": 0.4886359214782715, \"min\": 0.4886359214782715}}, \"EndTime\": 1605487029.064927, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.06492}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4887596778869629, \"sum\": 0.4887596778869629, \"min\": 0.4887596778869629}}, \"EndTime\": 1605487029.064953, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064946}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4881991195678711, \"sum\": 0.4881991195678711, \"min\": 0.4881991195678711}}, \"EndTime\": 1605487029.064979, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064972}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6326073951721192, \"sum\": 0.6326073951721192, \"min\": 0.6326073951721192}}, \"EndTime\": 1605487029.065005, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.064998}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6193580589294434, \"sum\": 0.6193580589294434, \"min\": 0.6193580589294434}}, \"EndTime\": 1605487029.065032, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065024}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6200655975341797, \"sum\": 0.6200655975341797, \"min\": 0.6200655975341797}}, \"EndTime\": 1605487029.065058, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065051}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6208483581542968, \"sum\": 0.6208483581542968, \"min\": 0.6208483581542968}}, \"EndTime\": 1605487029.065083, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065077}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6218261337280273, \"sum\": 0.6218261337280273, \"min\": 0.6218261337280273}}, \"EndTime\": 1605487029.065109, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065102}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6226631278991699, \"sum\": 0.6226631278991699, \"min\": 0.6226631278991699}}, \"EndTime\": 1605487029.065146, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065139}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6235209693908691, \"sum\": 0.6235209693908691, \"min\": 0.6235209693908691}}, \"EndTime\": 1605487029.065171, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065165}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6230196380615234, \"sum\": 0.6230196380615234, \"min\": 0.6230196380615234}}, \"EndTime\": 1605487029.065198, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065191}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6766007652282715, \"sum\": 0.6766007652282715, \"min\": 0.6766007652282715}}, \"EndTime\": 1605487029.065223, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065216}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6763639144897461, \"sum\": 0.6763639144897461, \"min\": 0.6763639144897461}}, \"EndTime\": 1605487029.065256, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065243}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6759825134277344, \"sum\": 0.6759825134277344, \"min\": 0.6759825134277344}}, \"EndTime\": 1605487029.06529, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065282}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6766696395874023, \"sum\": 0.6766696395874023, \"min\": 0.6766696395874023}}, \"EndTime\": 1605487029.065316, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065309}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6741667938232422, \"sum\": 0.6741667938232422, \"min\": 0.6741667938232422}}, \"EndTime\": 1605487029.065341, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065334}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6742713928222657, \"sum\": 0.6742713928222657, \"min\": 0.6742713928222657}}, \"EndTime\": 1605487029.065366, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065359}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6739395065307617, \"sum\": 0.6739395065307617, \"min\": 0.6739395065307617}}, \"EndTime\": 1605487029.065392, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.065385}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6747828254699707, \"sum\": 0.6747828254699707, \"min\": 0.6747828254699707}}, \"EndTime\": 1605487029.065417, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487029.06541}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #quality_metric: host=algo-1, epoch=3, train binary_classification_cross_entropy_objective =0.605103336334\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #early_stopping_criteria_metric: host=algo-1, epoch=3, criteria=binary_classification_cross_entropy_objective, value=0.488199119568\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] Epoch 3: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] Saving model for epoch: 3\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] Saved checkpoint to \"/tmp/tmpID9STn/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #progress_metric: host=algo-1, completed 26 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Total Batches Seen\": {\"count\": 1, \"max\": 56, \"sum\": 56.0, \"min\": 56}, \"Total Records Seen\": {\"count\": 1, \"max\": 10635, \"sum\": 10635.0, \"min\": 10635}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 6, \"sum\": 6.0, \"min\": 6}}, \"EndTime\": 1605487029.072289, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487028.761245}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #throughput_metric: host=algo-1, train throughput=6707.99620825 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:09.374] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 12, \"duration\": 301, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5923070182800293, \"sum\": 0.5923070182800293, \"min\": 0.5923070182800293}}, \"EndTime\": 1605487029.374247, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374162}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5934754791259765, \"sum\": 0.5934754791259765, \"min\": 0.5934754791259765}}, \"EndTime\": 1605487029.374359, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374336}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6063276786804199, \"sum\": 0.6063276786804199, \"min\": 0.6063276786804199}}, \"EndTime\": 1605487029.374424, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374411}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6007009239196778, \"sum\": 0.6007009239196778, \"min\": 0.6007009239196778}}, \"EndTime\": 1605487029.374488, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.37447}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49883197021484377, \"sum\": 0.49883197021484377, \"min\": 0.49883197021484377}}, \"EndTime\": 1605487029.374565, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374546}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49870824432373045, \"sum\": 0.49870824432373045, \"min\": 0.49870824432373045}}, \"EndTime\": 1605487029.374631, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374612}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49870326614379884, \"sum\": 0.49870326614379884, \"min\": 0.49870326614379884}}, \"EndTime\": 1605487029.374709, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374689}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4988750762939453, \"sum\": 0.4988750762939453, \"min\": 0.4988750762939453}}, \"EndTime\": 1605487029.374786, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374767}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.608296859741211, \"sum\": 0.608296859741211, \"min\": 0.608296859741211}}, \"EndTime\": 1605487029.374862, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374843}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6037945861816406, \"sum\": 0.6037945861816406, \"min\": 0.6037945861816406}}, \"EndTime\": 1605487029.374928, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.37491}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5953990669250488, \"sum\": 0.5953990669250488, \"min\": 0.5953990669250488}}, \"EndTime\": 1605487029.375013, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.374994}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6018939208984375, \"sum\": 0.6018939208984375, \"min\": 0.6018939208984375}}, \"EndTime\": 1605487029.375088, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375069}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4986510887145996, \"sum\": 0.4986510887145996, \"min\": 0.4986510887145996}}, \"EndTime\": 1605487029.375151, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375133}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4988982620239258, \"sum\": 0.4988982620239258, \"min\": 0.4988982620239258}}, \"EndTime\": 1605487029.375225, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375205}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4987527198791504, \"sum\": 0.4987527198791504, \"min\": 0.4987527198791504}}, \"EndTime\": 1605487029.375299, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.37528}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4988159446716309, \"sum\": 0.4988159446716309, \"min\": 0.4988159446716309}}, \"EndTime\": 1605487029.375359, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375343}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6259318618774414, \"sum\": 0.6259318618774414, \"min\": 0.6259318618774414}}, \"EndTime\": 1605487029.375413, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375398}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6188163528442383, \"sum\": 0.6188163528442383, \"min\": 0.6188163528442383}}, \"EndTime\": 1605487029.375479, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375462}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6193174858093262, \"sum\": 0.6193174858093262, \"min\": 0.6193174858093262}}, \"EndTime\": 1605487029.37554, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375523}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6191062660217285, \"sum\": 0.6191062660217285, \"min\": 0.6191062660217285}}, \"EndTime\": 1605487029.375597, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375583}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6299294509887695, \"sum\": 0.6299294509887695, \"min\": 0.6299294509887695}}, \"EndTime\": 1605487029.375629, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375621}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6279189720153808, \"sum\": 0.6279189720153808, \"min\": 0.6279189720153808}}, \"EndTime\": 1605487029.375655, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375648}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.629274959564209, \"sum\": 0.629274959564209, \"min\": 0.629274959564209}}, \"EndTime\": 1605487029.375713, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375699}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6282150688171386, \"sum\": 0.6282150688171386, \"min\": 0.6282150688171386}}, \"EndTime\": 1605487029.375767, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375751}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6765978164672851, \"sum\": 0.6765978164672851, \"min\": 0.6765978164672851}}, \"EndTime\": 1605487029.375826, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375809}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6767360153198242, \"sum\": 0.6767360153198242, \"min\": 0.6767360153198242}}, \"EndTime\": 1605487029.375884, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.375867}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.676021743774414, \"sum\": 0.676021743774414, \"min\": 0.676021743774414}}, \"EndTime\": 1605487029.375972, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.37596}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6754323806762695, \"sum\": 0.6754323806762695, \"min\": 0.6754323806762695}}, \"EndTime\": 1605487029.376027, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.376011}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6877170944213867, \"sum\": 0.6877170944213867, \"min\": 0.6877170944213867}}, \"EndTime\": 1605487029.37609, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.376072}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6880698394775391, \"sum\": 0.6880698394775391, \"min\": 0.6880698394775391}}, \"EndTime\": 1605487029.376164, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.376145}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6882654113769531, \"sum\": 0.6882654113769531, \"min\": 0.6882654113769531}}, \"EndTime\": 1605487029.376231, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.376212}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6875294952392578, \"sum\": 0.6875294952392578, \"min\": 0.6875294952392578}}, \"EndTime\": 1605487029.376291, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.376274}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #quality_metric: host=algo-1, epoch=4, train binary_classification_cross_entropy_objective =0.59230701828\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #early_stopping_criteria_metric: host=algo-1, epoch=4, criteria=binary_classification_cross_entropy_objective, value=0.498651088715\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] Saving model for epoch: 4\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] Saved checkpoint to \"/tmp/tmpRKRBuI/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #progress_metric: host=algo-1, completed 33 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Total Batches Seen\": {\"count\": 1, \"max\": 67, \"sum\": 67.0, \"min\": 67}, \"Total Records Seen\": {\"count\": 1, \"max\": 12722, \"sum\": 12722.0, \"min\": 12722}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 7, \"sum\": 7.0, \"min\": 7}}, \"EndTime\": 1605487029.382934, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487029.07246}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #throughput_metric: host=algo-1, train throughput=6719.38624652 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:09.759] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 14, \"duration\": 376, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.581334789276123, \"sum\": 0.581334789276123, \"min\": 0.581334789276123}}, \"EndTime\": 1605487029.759855, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.759756}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5819423942565918, \"sum\": 0.5819423942565918, \"min\": 0.5819423942565918}}, \"EndTime\": 1605487029.759981, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.759959}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5944002876281739, \"sum\": 0.5944002876281739, \"min\": 0.5944002876281739}}, \"EndTime\": 1605487029.760056, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760036}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5885651741027832, \"sum\": 0.5885651741027832, \"min\": 0.5885651741027832}}, \"EndTime\": 1605487029.76013, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.76011}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49236834716796873, \"sum\": 0.49236834716796873, \"min\": 0.49236834716796873}}, \"EndTime\": 1605487029.760193, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760178}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4945363464355469, \"sum\": 0.4945363464355469, \"min\": 0.4945363464355469}}, \"EndTime\": 1605487029.760258, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.76024}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49191286087036135, \"sum\": 0.49191286087036135, \"min\": 0.49191286087036135}}, \"EndTime\": 1605487029.760322, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760304}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4935973434448242, \"sum\": 0.4935973434448242, \"min\": 0.4935973434448242}}, \"EndTime\": 1605487029.760385, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760367}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5960123901367187, \"sum\": 0.5960123901367187, \"min\": 0.5960123901367187}}, \"EndTime\": 1605487029.760456, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760439}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5916014900207519, \"sum\": 0.5916014900207519, \"min\": 0.5916014900207519}}, \"EndTime\": 1605487029.760518, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.7605}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5843301391601563, \"sum\": 0.5843301391601563, \"min\": 0.5843301391601563}}, \"EndTime\": 1605487029.760578, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760561}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5897945365905761, \"sum\": 0.5897945365905761, \"min\": 0.5897945365905761}}, \"EndTime\": 1605487029.760638, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760621}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49661445236206053, \"sum\": 0.49661445236206053, \"min\": 0.49661445236206053}}, \"EndTime\": 1605487029.7607, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760682}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4993660545349121, \"sum\": 0.4993660545349121, \"min\": 0.4993660545349121}}, \"EndTime\": 1605487029.760761, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760744}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4978565330505371, \"sum\": 0.4978565330505371, \"min\": 0.4978565330505371}}, \"EndTime\": 1605487029.760821, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760805}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5011259269714355, \"sum\": 0.5011259269714355, \"min\": 0.5011259269714355}}, \"EndTime\": 1605487029.760883, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760865}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6227033882141113, \"sum\": 0.6227033882141113, \"min\": 0.6227033882141113}}, \"EndTime\": 1605487029.760945, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760927}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.620539493560791, \"sum\": 0.620539493560791, \"min\": 0.620539493560791}}, \"EndTime\": 1605487029.761005, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.760988}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6205267028808594, \"sum\": 0.6205267028808594, \"min\": 0.6205267028808594}}, \"EndTime\": 1605487029.761066, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.76105}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6200584182739258, \"sum\": 0.6200584182739258, \"min\": 0.6200584182739258}}, \"EndTime\": 1605487029.761129, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761111}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6356771965026855, \"sum\": 0.6356771965026855, \"min\": 0.6356771965026855}}, \"EndTime\": 1605487029.76119, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761173}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6373911399841309, \"sum\": 0.6373911399841309, \"min\": 0.6373911399841309}}, \"EndTime\": 1605487029.76125, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761233}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6349834480285644, \"sum\": 0.6349834480285644, \"min\": 0.6349834480285644}}, \"EndTime\": 1605487029.761309, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761293}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6372133178710937, \"sum\": 0.6372133178710937, \"min\": 0.6372133178710937}}, \"EndTime\": 1605487029.761368, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761351}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6754119682312012, \"sum\": 0.6754119682312012, \"min\": 0.6754119682312012}}, \"EndTime\": 1605487029.76143, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761412}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6753462677001953, \"sum\": 0.6753462677001953, \"min\": 0.6753462677001953}}, \"EndTime\": 1605487029.761489, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761472}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6754363212585449, \"sum\": 0.6754363212585449, \"min\": 0.6754363212585449}}, \"EndTime\": 1605487029.76155, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761533}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6756022033691407, \"sum\": 0.6756022033691407, \"min\": 0.6756022033691407}}, \"EndTime\": 1605487029.761613, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761595}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6879720916748047, \"sum\": 0.6879720916748047, \"min\": 0.6879720916748047}}, \"EndTime\": 1605487029.761674, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761656}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6895301055908203, \"sum\": 0.6895301055908203, \"min\": 0.6895301055908203}}, \"EndTime\": 1605487029.76174, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761717}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6881953926086426, \"sum\": 0.6881953926086426, \"min\": 0.6881953926086426}}, \"EndTime\": 1605487029.761802, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761785}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6903295135498047, \"sum\": 0.6903295135498047, \"min\": 0.6903295135498047}}, \"EndTime\": 1605487029.761864, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.761846}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #quality_metric: host=algo-1, epoch=5, train binary_classification_cross_entropy_objective =0.581334789276\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #early_stopping_criteria_metric: host=algo-1, epoch=5, criteria=binary_classification_cross_entropy_objective, value=0.49191286087\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] Saving model for epoch: 5\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] Saved checkpoint to \"/tmp/tmp6uFDnE/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #progress_metric: host=algo-1, completed 40 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Total Batches Seen\": {\"count\": 1, \"max\": 78, \"sum\": 78.0, \"min\": 78}, \"Total Records Seen\": {\"count\": 1, \"max\": 14809, \"sum\": 14809.0, \"min\": 14809}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}}, \"EndTime\": 1605487029.771289, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487029.383225}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:09 INFO 140309854623552] #throughput_metric: host=algo-1, train throughput=5376.1371022 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:10.091] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 16, \"duration\": 319, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5719401664733886, \"sum\": 0.5719401664733886, \"min\": 0.5719401664733886}}, \"EndTime\": 1605487030.091211, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091111}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5720376014709473, \"sum\": 0.5720376014709473, \"min\": 0.5720376014709473}}, \"EndTime\": 1605487030.091298, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091277}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5841081581115722, \"sum\": 0.5841081581115722, \"min\": 0.5841081581115722}}, \"EndTime\": 1605487030.091356, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091344}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5781074523925781, \"sum\": 0.5781074523925781, \"min\": 0.5781074523925781}}, \"EndTime\": 1605487030.091414, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091397}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49146702194213865, \"sum\": 0.49146702194213865, \"min\": 0.49146702194213865}}, \"EndTime\": 1605487030.09147, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091459}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4923815803527832, \"sum\": 0.4923815803527832, \"min\": 0.4923815803527832}}, \"EndTime\": 1605487030.09152, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091504}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4914534339904785, \"sum\": 0.4914534339904785, \"min\": 0.4914534339904785}}, \"EndTime\": 1605487030.091581, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091564}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.492054801940918, \"sum\": 0.492054801940918, \"min\": 0.492054801940918}}, \"EndTime\": 1605487030.091637, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091622}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5854732818603515, \"sum\": 0.5854732818603515, \"min\": 0.5854732818603515}}, \"EndTime\": 1605487030.091699, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091682}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5810643577575684, \"sum\": 0.5810643577575684, \"min\": 0.5810643577575684}}, \"EndTime\": 1605487030.091757, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091746}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5748397254943848, \"sum\": 0.5748397254943848, \"min\": 0.5748397254943848}}, \"EndTime\": 1605487030.091814, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091798}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.57938330078125, \"sum\": 0.57938330078125, \"min\": 0.57938330078125}}, \"EndTime\": 1605487030.091888, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091871}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4938884620666504, \"sum\": 0.4938884620666504, \"min\": 0.4938884620666504}}, \"EndTime\": 1605487030.091992, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.091971}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4952766151428223, \"sum\": 0.4952766151428223, \"min\": 0.4952766151428223}}, \"EndTime\": 1605487030.092066, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092047}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4941532516479492, \"sum\": 0.4941532516479492, \"min\": 0.4941532516479492}}, \"EndTime\": 1605487030.092138, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.09212}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49575234222412107, \"sum\": 0.49575234222412107, \"min\": 0.49575234222412107}}, \"EndTime\": 1605487030.092195, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.09218}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6216414985656739, \"sum\": 0.6216414985656739, \"min\": 0.6216414985656739}}, \"EndTime\": 1605487030.092253, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092236}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6221803436279297, \"sum\": 0.6221803436279297, \"min\": 0.6221803436279297}}, \"EndTime\": 1605487030.092326, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092307}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6218546981811524, \"sum\": 0.6218546981811524, \"min\": 0.6218546981811524}}, \"EndTime\": 1605487030.0924, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092382}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6215831069946289, \"sum\": 0.6215831069946289, \"min\": 0.6215831069946289}}, \"EndTime\": 1605487030.092457, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092446}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6315207595825195, \"sum\": 0.6315207595825195, \"min\": 0.6315207595825195}}, \"EndTime\": 1605487030.09251, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092494}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6339841918945313, \"sum\": 0.6339841918945313, \"min\": 0.6339841918945313}}, \"EndTime\": 1605487030.092569, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092553}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6321962242126464, \"sum\": 0.6321962242126464, \"min\": 0.6321962242126464}}, \"EndTime\": 1605487030.092603, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092595}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6340477600097656, \"sum\": 0.6340477600097656, \"min\": 0.6340477600097656}}, \"EndTime\": 1605487030.092654, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092639}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6753828659057617, \"sum\": 0.6753828659057617, \"min\": 0.6753828659057617}}, \"EndTime\": 1605487030.092724, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092707}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6754612426757812, \"sum\": 0.6754612426757812, \"min\": 0.6754612426757812}}, \"EndTime\": 1605487030.092787, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092769}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.675661319732666, \"sum\": 0.675661319732666, \"min\": 0.675661319732666}}, \"EndTime\": 1605487030.092858, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092839}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6757239837646485, \"sum\": 0.6757239837646485, \"min\": 0.6757239837646485}}, \"EndTime\": 1605487030.092927, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.092911}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6817322998046875, \"sum\": 0.6817322998046875, \"min\": 0.6817322998046875}}, \"EndTime\": 1605487030.092986, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.09297}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6830495529174805, \"sum\": 0.6830495529174805, \"min\": 0.6830495529174805}}, \"EndTime\": 1605487030.093058, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.093039}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6817018585205078, \"sum\": 0.6817018585205078, \"min\": 0.6817018585205078}}, \"EndTime\": 1605487030.093124, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.093105}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6835907745361328, \"sum\": 0.6835907745361328, \"min\": 0.6835907745361328}}, \"EndTime\": 1605487030.093183, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487030.093166}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #quality_metric: host=algo-1, epoch=6, train binary_classification_cross_entropy_objective =0.571940166473\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #early_stopping_criteria_metric: host=algo-1, epoch=6, criteria=binary_classification_cross_entropy_objective, value=0.49145343399\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] Saving model for epoch: 6\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] Saved checkpoint to \"/tmp/tmpVUDhAb/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #progress_metric: host=algo-1, completed 46 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Total Batches Seen\": {\"count\": 1, \"max\": 89, \"sum\": 89.0, \"min\": 89}, \"Total Records Seen\": {\"count\": 1, \"max\": 16896, \"sum\": 16896.0, \"min\": 16896}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 9, \"sum\": 9.0, \"min\": 9}}, \"EndTime\": 1605487030.099895, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487029.77161}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #throughput_metric: host=algo-1, train throughput=6353.94969836 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:10.423] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 18, \"duration\": 323, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5636245498657226, \"sum\": 0.5636245498657226, \"min\": 0.5636245498657226}}, \"EndTime\": 1605487030.424139, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424038}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5632367973327637, \"sum\": 0.5632367973327637, \"min\": 0.5632367973327637}}, \"EndTime\": 1605487030.424242, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424219}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5750062217712403, \"sum\": 0.5750062217712403, \"min\": 0.5750062217712403}}, \"EndTime\": 1605487030.424314, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424295}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5688189888000488, \"sum\": 0.5688189888000488, \"min\": 0.5688189888000488}}, \"EndTime\": 1605487030.424381, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424363}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49443558502197266, \"sum\": 0.49443558502197266, \"min\": 0.49443558502197266}}, \"EndTime\": 1605487030.424447, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424429}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4946299171447754, \"sum\": 0.4946299171447754, \"min\": 0.4946299171447754}}, \"EndTime\": 1605487030.424513, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424494}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4945528450012207, \"sum\": 0.4945528450012207, \"min\": 0.4945528450012207}}, \"EndTime\": 1605487030.424574, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424557}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.49488971710205076, \"sum\": 0.49488971710205076, \"min\": 0.49488971710205076}}, \"EndTime\": 1605487030.42464, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424622}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5761798248291016, \"sum\": 0.5761798248291016, \"min\": 0.5761798248291016}}, \"EndTime\": 1605487030.424703, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424685}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.571687068939209, \"sum\": 0.571687068939209, \"min\": 0.571687068939209}}, \"EndTime\": 1605487030.424764, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424747}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5664372177124023, \"sum\": 0.5664372177124023, \"min\": 0.5664372177124023}}, \"EndTime\": 1605487030.424825, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424808}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5701526908874511, \"sum\": 0.5701526908874511, \"min\": 0.5701526908874511}}, \"EndTime\": 1605487030.424889, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424871}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4956012573242187, \"sum\": 0.4956012573242187, \"min\": 0.4956012573242187}}, \"EndTime\": 1605487030.424952, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424934}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.495544490814209, \"sum\": 0.495544490814209, \"min\": 0.495544490814209}}, \"EndTime\": 1605487030.425015, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.424997}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4951608467102051, \"sum\": 0.4951608467102051, \"min\": 0.4951608467102051}}, \"EndTime\": 1605487030.425074, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425058}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.4950503730773926, \"sum\": 0.4950503730773926, \"min\": 0.4950503730773926}}, \"EndTime\": 1605487030.425133, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425117}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6214595718383789, \"sum\": 0.6214595718383789, \"min\": 0.6214595718383789}}, \"EndTime\": 1605487030.425198, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425179}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6226315155029297, \"sum\": 0.6226315155029297, \"min\": 0.6226315155029297}}, \"EndTime\": 1605487030.425259, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425242}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6223736381530762, \"sum\": 0.6223736381530762, \"min\": 0.6223736381530762}}, \"EndTime\": 1605487030.425319, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425302}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.622346019744873, \"sum\": 0.622346019744873, \"min\": 0.622346019744873}}, \"EndTime\": 1605487030.425379, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425362}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6297521667480469, \"sum\": 0.6297521667480469, \"min\": 0.6297521667480469}}, \"EndTime\": 1605487030.42544, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425423}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6314647102355957, \"sum\": 0.6314647102355957, \"min\": 0.6314647102355957}}, \"EndTime\": 1605487030.425498, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425481}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6296482887268067, \"sum\": 0.6296482887268067, \"min\": 0.6296482887268067}}, \"EndTime\": 1605487030.425563, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425544}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6315528793334961, \"sum\": 0.6315528793334961, \"min\": 0.6315528793334961}}, \"EndTime\": 1605487030.425625, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425607}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6758378067016602, \"sum\": 0.6758378067016602, \"min\": 0.6758378067016602}}, \"EndTime\": 1605487030.425684, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425667}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6758712348937989, \"sum\": 0.6758712348937989, \"min\": 0.6758712348937989}}, \"EndTime\": 1605487030.425744, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425727}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6757376556396484, \"sum\": 0.6757376556396484, \"min\": 0.6757376556396484}}, \"EndTime\": 1605487030.425803, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425786}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6757763595581054, \"sum\": 0.6757763595581054, \"min\": 0.6757763595581054}}, \"EndTime\": 1605487030.425865, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425849}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6841713790893554, \"sum\": 0.6841713790893554, \"min\": 0.6841713790893554}}, \"EndTime\": 1605487030.425922, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425905}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.684825668334961, \"sum\": 0.684825668334961, \"min\": 0.684825668334961}}, \"EndTime\": 1605487030.425976, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.425962}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.684192237854004, \"sum\": 0.684192237854004, \"min\": 0.684192237854004}}, \"EndTime\": 1605487030.426031, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.426014}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6846019973754883, \"sum\": 0.6846019973754883, \"min\": 0.6846019973754883}}, \"EndTime\": 1605487030.426092, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.426073}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #quality_metric: host=algo-1, epoch=7, train binary_classification_cross_entropy_objective =0.563624549866\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #early_stopping_criteria_metric: host=algo-1, epoch=7, criteria=binary_classification_cross_entropy_objective, value=0.494435585022\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] Saving model for epoch: 7\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] Saved checkpoint to \"/tmp/tmpg17j0J/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] Early stop condition met. Stopping training.\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #progress_metric: host=algo-1, completed 100 % epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Total Batches Seen\": {\"count\": 1, \"max\": 100, \"sum\": 100.0, \"min\": 100}, \"Total Records Seen\": {\"count\": 1, \"max\": 18983, \"sum\": 18983.0, \"min\": 18983}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 2087, \"sum\": 2087.0, \"min\": 2087}, \"Reset Count\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}}, \"EndTime\": 1605487030.437391, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487030.100192}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #throughput_metric: host=algo-1, train throughput=6186.84000396 records/second\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 WARNING 140309854623552] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 WARNING 140309854623552] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:10.439] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 20, \"duration\": 1, \"num_examples\": 1, \"num_bytes\": 12000}\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:10.583] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 23, \"duration\": 140, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m[2020-11-16 00:37:10.645] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 25, \"duration\": 44, \"num_examples\": 11, \"num_bytes\": 125220}\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #train_score (algo-1) : ('binary_classification_cross_entropy_objective', 0.49385064586133093)\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #train_score (algo-1) : ('binary_classification_accuracy', 0.7585050311451845)\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #train_score (algo-1) : ('binary_f_1.000', 0.784799316823228)\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #train_score (algo-1) : ('precision', 0.8294223826714802)\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #train_score (algo-1) : ('recall', 0.7447325769854133)\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #train_score (algo-1) : ('roc_auc_score', 0.8360695685548764)\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #quality_metric: host=algo-1, train binary_classification_cross_entropy_objective =0.493850645861\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #quality_metric: host=algo-1, train binary_classification_accuracy =0.758505031145\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #quality_metric: host=algo-1, train binary_f_1.000 =0.784799316823\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #quality_metric: host=algo-1, train precision =0.829422382671\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #quality_metric: host=algo-1, train recall =0.744732576985\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] #quality_metric: host=algo-1, train roc_auc_score =0.836069568555\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] Best model found for hyperparameters: {\"lr_scheduler_step\": 100, \"wd\": 0.01, \"optimizer\": \"adam\", \"lr_scheduler_factor\": 0.99, \"l1\": 0.0, \"learning_rate\": 0.1, \"lr_scheduler_minimum_lr\": 0.0001}\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] Saved checkpoint to \"/tmp/tmpcqFTIT/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:37:10 INFO 140309854623552] Test data is not provided.\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"totaltime\": {\"count\": 1, \"max\": 3364.319086074829, \"sum\": 3364.319086074829, \"min\": 3364.319086074829}, \"finalize.time\": {\"count\": 1, \"max\": 214.54286575317383, \"sum\": 214.54286575317383, \"min\": 214.54286575317383}, \"initialize.time\": {\"count\": 1, \"max\": 178.81202697753906, \"sum\": 178.81202697753906, \"min\": 178.81202697753906}, \"check_early_stopping.time\": {\"count\": 8, \"max\": 0.9119510650634766, \"sum\": 4.1942596435546875, \"min\": 0.18906593322753906}, \"setuptime\": {\"count\": 1, \"max\": 26.007890701293945, \"sum\": 26.007890701293945, \"min\": 26.007890701293945}, \"update.time\": {\"count\": 8, \"max\": 383.99291038513184, \"sum\": 2675.062894821167, \"min\": 296.34714126586914}, \"epochs\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}}, \"EndTime\": 1605487030.659252, \"Dimensions\": {\"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1605487027.553222}\n", "\u001b[0m\n", "\n", "2020-11-16 00:37:20 Uploading - Uploading generated training model\n", "2020-11-16 00:37:20 Completed - Training job completed\n", "Training seconds: 55\n", "Billable seconds: 55\n", "-----------------!2020-11-16 00:46:18 Starting - Starting the training job...\n", "2020-11-16 00:46:20 Starting - Launching requested ML instances.........\n", "2020-11-16 00:47:50 Starting - Preparing the instances for training......\n", "2020-11-16 00:49:19 Downloading - Downloading input data...\n", "2020-11-16 00:49:32 Training - Downloading the training image.\u001b[34mDocker entrypoint called with argument(s): train\u001b[0m\n", "\u001b[34mRunning default environment configuration script\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:56 INFO 140177577584448] Reading default configuration from /opt/amazon/lib/python2.7/site-packages/algorithm/resources/default-input.json: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'auto', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'1000', u'huber_delta': u'1.0', u'num_classes': u'1', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:56 INFO 140177577584448] Merging with provided configuration from /opt/ml/input/config/hyperparameters.json: {u'feature_dim': u'3', u'mini_batch_size': u'200', u'predictor_type': u'binary_classifier'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:56 INFO 140177577584448] Final configuration: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'3', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'200', u'huber_delta': u'1.0', u'num_classes': u'1', u'predictor_type': u'binary_classifier', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:56 WARNING 140177577584448] Loggers have already been setup.\u001b[0m\n", "\u001b[34mProcess 1 is a worker.\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:56 INFO 140177577584448] Using default worker.\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Checkpoint loading and saving are disabled.\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:57.182] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 0, \"duration\": 15, \"num_examples\": 1, \"num_bytes\": 11200}\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Create Store: local\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:57.232] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 1, \"duration\": 49, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Scaler algorithm parameters\n", " \u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Scaling model computed with parameters:\n", " {'stdev_weight': \u001b[0m\n", "\u001b[34m[3.6276784 3.708638 2.4676423]\u001b[0m\n", "\u001b[34m, 'stdev_label': None, 'mean_label': None, 'mean_weight': \u001b[0m\n", "\u001b[34m[22.726074 22.870358 6.4400005]\u001b[0m\n", "\u001b[34m}\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] nvidia-smi took: 0.0252249240875 secs to identify 0 gpus\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Number of GPUs being used: 0\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Total Batches Seen\": {\"count\": 1, \"max\": 9, \"sum\": 9.0, \"min\": 9}, \"Total Records Seen\": {\"count\": 1, \"max\": 1681, \"sum\": 1681.0, \"min\": 1681}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 2, \"sum\": 2.0, \"min\": 2}}, \"EndTime\": 1605487797.335357, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"init_train_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1605487797.335292}\n", "\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:57.566] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 4, \"duration\": 230, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6847719573974609, \"sum\": 0.6847719573974609, \"min\": 0.6847719573974609}}, \"EndTime\": 1605487797.566342, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566244}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6788633074079241, \"sum\": 0.6788633074079241, \"min\": 0.6788633074079241}}, \"EndTime\": 1605487797.566448, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566427}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7093746730259487, \"sum\": 0.7093746730259487, \"min\": 0.7093746730259487}}, \"EndTime\": 1605487797.566522, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566502}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6805233437674386, \"sum\": 0.6805233437674386, \"min\": 0.6805233437674386}}, \"EndTime\": 1605487797.566593, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566574}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6549218096051898, \"sum\": 0.6549218096051898, \"min\": 0.6549218096051898}}, \"EndTime\": 1605487797.566659, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.56664}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6426505497523717, \"sum\": 0.6426505497523717, \"min\": 0.6426505497523717}}, \"EndTime\": 1605487797.566725, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566706}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6620791898454939, \"sum\": 0.6620791898454939, \"min\": 0.6620791898454939}}, \"EndTime\": 1605487797.566775, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566764}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6478470884050642, \"sum\": 0.6478470884050642, \"min\": 0.6478470884050642}}, \"EndTime\": 1605487797.566837, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566819}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7014443751743862, \"sum\": 0.7014443751743862, \"min\": 0.7014443751743862}}, \"EndTime\": 1605487797.566885, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566875}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.686005379813058, \"sum\": 0.686005379813058, \"min\": 0.686005379813058}}, \"EndTime\": 1605487797.566941, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566924}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6773564147949219, \"sum\": 0.6773564147949219, \"min\": 0.6773564147949219}}, \"EndTime\": 1605487797.567008, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.566984}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6947728838239398, \"sum\": 0.6947728838239398, \"min\": 0.6947728838239398}}, \"EndTime\": 1605487797.567065, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567048}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6546882792881557, \"sum\": 0.6546882792881557, \"min\": 0.6546882792881557}}, \"EndTime\": 1605487797.567124, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567109}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6431979479108538, \"sum\": 0.6431979479108538, \"min\": 0.6431979479108538}}, \"EndTime\": 1605487797.567192, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567173}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6580340358189174, \"sum\": 0.6580340358189174, \"min\": 0.6580340358189174}}, \"EndTime\": 1605487797.567257, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567239}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6456500298636301, \"sum\": 0.6456500298636301, \"min\": 0.6456500298636301}}, \"EndTime\": 1605487797.567323, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567304}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6914764295305524, \"sum\": 0.6914764295305524, \"min\": 0.6914764295305524}}, \"EndTime\": 1605487797.567386, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567369}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6779577854701451, \"sum\": 0.6779577854701451, \"min\": 0.6779577854701451}}, \"EndTime\": 1605487797.567449, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567431}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6968207114083427, \"sum\": 0.6968207114083427, \"min\": 0.6968207114083427}}, \"EndTime\": 1605487797.56752, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.5675}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7140933990478515, \"sum\": 0.7140933990478515, \"min\": 0.7140933990478515}}, \"EndTime\": 1605487797.567583, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567565}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6649264526367188, \"sum\": 0.6649264526367188, \"min\": 0.6649264526367188}}, \"EndTime\": 1605487797.567642, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567625}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6661874171665737, \"sum\": 0.6661874171665737, \"min\": 0.6661874171665737}}, \"EndTime\": 1605487797.567696, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567682}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6688926696777344, \"sum\": 0.6688926696777344, \"min\": 0.6688926696777344}}, \"EndTime\": 1605487797.567763, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.567745}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6661637333461217, \"sum\": 0.6661637333461217, \"min\": 0.6661637333461217}}, \"EndTime\": 1605487797.567827, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.56781}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6988330623081752, \"sum\": 0.6988330623081752, \"min\": 0.6988330623081752}}, \"EndTime\": 1605487797.567887, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.56787}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6968572017124721, \"sum\": 0.6968572017124721, \"min\": 0.6968572017124721}}, \"EndTime\": 1605487797.56796, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.56794}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6931787218366351, \"sum\": 0.6931787218366351, \"min\": 0.6931787218366351}}, \"EndTime\": 1605487797.568032, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.568013}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6994777679443359, \"sum\": 0.6994777679443359, \"min\": 0.6994777679443359}}, \"EndTime\": 1605487797.568105, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.568085}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7117443738664899, \"sum\": 0.7117443738664899, \"min\": 0.7117443738664899}}, \"EndTime\": 1605487797.568181, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.568161}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.713960440499442, \"sum\": 0.713960440499442, \"min\": 0.713960440499442}}, \"EndTime\": 1605487797.568246, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.568227}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7147993687220983, \"sum\": 0.7147993687220983, \"min\": 0.7147993687220983}}, \"EndTime\": 1605487797.568311, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.568292}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7089545004708426, \"sum\": 0.7089545004708426, \"min\": 0.7089545004708426}}, \"EndTime\": 1605487797.568373, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.568355}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] #quality_metric: host=algo-1, epoch=0, train binary_classification_cross_entropy_objective =0.684771957397\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] #early_stopping_criteria_metric: host=algo-1, epoch=0, criteria=binary_classification_cross_entropy_objective, value=0.642650549752\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Epoch 0: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Saving model for epoch: 0\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Saved checkpoint to \"/tmp/tmpij8QwO/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] #progress_metric: host=algo-1, completed 6 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Total Batches Seen\": {\"count\": 1, \"max\": 17, \"sum\": 17.0, \"min\": 17}, \"Total Records Seen\": {\"count\": 1, \"max\": 3162, \"sum\": 3162.0, \"min\": 3162}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 3, \"sum\": 3.0, \"min\": 3}}, \"EndTime\": 1605487797.579425, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1605487797.335676}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] #throughput_metric: host=algo-1, train throughput=6072.90720085 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:57.848] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 6, \"duration\": 268, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6710912105015346, \"sum\": 0.6710912105015346, \"min\": 0.6710912105015346}}, \"EndTime\": 1605487797.848967, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.848807}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6651641409737723, \"sum\": 0.6651641409737723, \"min\": 0.6651641409737723}}, \"EndTime\": 1605487797.849105, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.84908}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6946474347795759, \"sum\": 0.6946474347795759, \"min\": 0.6946474347795759}}, \"EndTime\": 1605487797.849205, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849184}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.666675796508789, \"sum\": 0.666675796508789, \"min\": 0.666675796508789}}, \"EndTime\": 1605487797.849276, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849257}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5602065931047712, \"sum\": 0.5602065931047712, \"min\": 0.5602065931047712}}, \"EndTime\": 1605487797.84938, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.84936}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5555398123604911, \"sum\": 0.5555398123604911, \"min\": 0.5555398123604911}}, \"EndTime\": 1605487797.849461, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849428}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.563250607081822, \"sum\": 0.563250607081822, \"min\": 0.563250607081822}}, \"EndTime\": 1605487797.849559, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849537}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5575196184430804, \"sum\": 0.5575196184430804, \"min\": 0.5575196184430804}}, \"EndTime\": 1605487797.849656, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849605}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6871641649518694, \"sum\": 0.6871641649518694, \"min\": 0.6871641649518694}}, \"EndTime\": 1605487797.849731, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849711}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6721328408377512, \"sum\": 0.6721328408377512, \"min\": 0.6721328408377512}}, \"EndTime\": 1605487797.84983, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849777}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.664357659476144, \"sum\": 0.664357659476144, \"min\": 0.664357659476144}}, \"EndTime\": 1605487797.849903, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849883}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6804521942138672, \"sum\": 0.6804521942138672, \"min\": 0.6804521942138672}}, \"EndTime\": 1605487797.850026, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.849947}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5602396665300642, \"sum\": 0.5602396665300642, \"min\": 0.5602396665300642}}, \"EndTime\": 1605487797.850099, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850079}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5558867427280971, \"sum\": 0.5558867427280971, \"min\": 0.5558867427280971}}, \"EndTime\": 1605487797.850216, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850194}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5620862906319755, \"sum\": 0.5620862906319755, \"min\": 0.5620862906319755}}, \"EndTime\": 1605487797.85029, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.85027}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5566635131835938, \"sum\": 0.5566635131835938, \"min\": 0.5566635131835938}}, \"EndTime\": 1605487797.850389, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850368}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6786903817313058, \"sum\": 0.6786903817313058, \"min\": 0.6786903817313058}}, \"EndTime\": 1605487797.850456, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850437}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6651828983851842, \"sum\": 0.6651828983851842, \"min\": 0.6651828983851842}}, \"EndTime\": 1605487797.850556, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850535}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.683646992274693, \"sum\": 0.683646992274693, \"min\": 0.683646992274693}}, \"EndTime\": 1605487797.850629, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.85061}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6985908399309431, \"sum\": 0.6985908399309431, \"min\": 0.6985908399309431}}, \"EndTime\": 1605487797.850726, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850704}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6724324689592633, \"sum\": 0.6724324689592633, \"min\": 0.6724324689592633}}, \"EndTime\": 1605487797.850791, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850773}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6680428641183036, \"sum\": 0.6680428641183036, \"min\": 0.6680428641183036}}, \"EndTime\": 1605487797.85091, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850864}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.664455441066197, \"sum\": 0.664455441066197, \"min\": 0.664455441066197}}, \"EndTime\": 1605487797.850988, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.850968}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6694985307965959, \"sum\": 0.6694985307965959, \"min\": 0.6694985307965959}}, \"EndTime\": 1605487797.851104, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.85104}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6946494511195591, \"sum\": 0.6946494511195591, \"min\": 0.6946494511195591}}, \"EndTime\": 1605487797.851182, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.851162}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6927333395821708, \"sum\": 0.6927333395821708, \"min\": 0.6927333395821708}}, \"EndTime\": 1605487797.851279, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.851227}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6921010480608258, \"sum\": 0.6921010480608258, \"min\": 0.6921010480608258}}, \"EndTime\": 1605487797.851349, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.851331}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6937323107038226, \"sum\": 0.6937323107038226, \"min\": 0.6937323107038226}}, \"EndTime\": 1605487797.851447, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.851394}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7131102098737444, \"sum\": 0.7131102098737444, \"min\": 0.7131102098737444}}, \"EndTime\": 1605487797.851516, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.851497}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7110306113106863, \"sum\": 0.7110306113106863, \"min\": 0.7110306113106863}}, \"EndTime\": 1605487797.851618, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.851594}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7102523149762835, \"sum\": 0.7102523149762835, \"min\": 0.7102523149762835}}, \"EndTime\": 1605487797.851691, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.851672}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7170843723842076, \"sum\": 0.7170843723842076, \"min\": 0.7170843723842076}}, \"EndTime\": 1605487797.851788, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.85176}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] #quality_metric: host=algo-1, epoch=1, train binary_classification_cross_entropy_objective =0.671091210502\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] #early_stopping_criteria_metric: host=algo-1, epoch=1, criteria=binary_classification_cross_entropy_objective, value=0.55553981236\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Epoch 1: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Saving model for epoch: 1\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] Saved checkpoint to \"/tmp/tmpu9h0Qv/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] #progress_metric: host=algo-1, completed 13 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Total Batches Seen\": {\"count\": 1, \"max\": 25, \"sum\": 25.0, \"min\": 25}, \"Total Records Seen\": {\"count\": 1, \"max\": 4643, \"sum\": 4643.0, \"min\": 4643}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 4, \"sum\": 4.0, \"min\": 4}}, \"EndTime\": 1605487797.861895, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1605487797.579701}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:57 INFO 140177577584448] #throughput_metric: host=algo-1, train throughput=5246.26740257 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:58.153] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 8, \"duration\": 291, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6612460218157087, \"sum\": 0.6612460218157087, \"min\": 0.6612460218157087}}, \"EndTime\": 1605487798.154052, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.153938}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6553184291294643, \"sum\": 0.6553184291294643, \"min\": 0.6553184291294643}}, \"EndTime\": 1605487798.154149, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.15413}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.683748310634068, \"sum\": 0.683748310634068, \"min\": 0.683748310634068}}, \"EndTime\": 1605487798.154209, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154194}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6567361886160714, \"sum\": 0.6567361886160714, \"min\": 0.6567361886160714}}, \"EndTime\": 1605487798.154263, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154249}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5420398766653879, \"sum\": 0.5420398766653879, \"min\": 0.5420398766653879}}, \"EndTime\": 1605487798.154316, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154302}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5413344900948661, \"sum\": 0.5413344900948661, \"min\": 0.5413344900948661}}, \"EndTime\": 1605487798.154367, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154354}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5426819556100028, \"sum\": 0.5426819556100028, \"min\": 0.5426819556100028}}, \"EndTime\": 1605487798.154417, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154404}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5416428538731166, \"sum\": 0.5416428538731166, \"min\": 0.5416428538731166}}, \"EndTime\": 1605487798.154479, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154453}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6766714913504465, \"sum\": 0.6766714913504465, \"min\": 0.6766714913504465}}, \"EndTime\": 1605487798.154537, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.15452}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6619686562674386, \"sum\": 0.6619686562674386, \"min\": 0.6619686562674386}}, \"EndTime\": 1605487798.154597, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154579}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6550835309709822, \"sum\": 0.6550835309709822, \"min\": 0.6550835309709822}}, \"EndTime\": 1605487798.15465, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154637}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6700645664760044, \"sum\": 0.6700645664760044, \"min\": 0.6700645664760044}}, \"EndTime\": 1605487798.154699, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154686}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5425730623517717, \"sum\": 0.5425730623517717, \"min\": 0.5425730623517717}}, \"EndTime\": 1605487798.154745, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154733}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5417962973458427, \"sum\": 0.5417962973458427, \"min\": 0.5417962973458427}}, \"EndTime\": 1605487798.154792, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.15478}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5428745923723494, \"sum\": 0.5428745923723494, \"min\": 0.5428745923723494}}, \"EndTime\": 1605487798.154839, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154826}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5419011688232422, \"sum\": 0.5419011688232422, \"min\": 0.5419011688232422}}, \"EndTime\": 1605487798.154885, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154873}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.670060054234096, \"sum\": 0.670060054234096, \"min\": 0.670060054234096}}, \"EndTime\": 1605487798.154931, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154919}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6569789777483259, \"sum\": 0.6569789777483259, \"min\": 0.6569789777483259}}, \"EndTime\": 1605487798.154978, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.154965}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6746194022042411, \"sum\": 0.6746194022042411, \"min\": 0.6746194022042411}}, \"EndTime\": 1605487798.155023, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155011}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6874453299386161, \"sum\": 0.6874453299386161, \"min\": 0.6874453299386161}}, \"EndTime\": 1605487798.15507, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155057}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6576537758963449, \"sum\": 0.6576537758963449, \"min\": 0.6576537758963449}}, \"EndTime\": 1605487798.155116, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155104}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6662977600097656, \"sum\": 0.6662977600097656, \"min\": 0.6662977600097656}}, \"EndTime\": 1605487798.155162, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.15515}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6684203556605748, \"sum\": 0.6684203556605748, \"min\": 0.6684203556605748}}, \"EndTime\": 1605487798.155209, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155196}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6623662458147321, \"sum\": 0.6623662458147321, \"min\": 0.6623662458147321}}, \"EndTime\": 1605487798.155255, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155243}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6902782113211495, \"sum\": 0.6902782113211495, \"min\": 0.6902782113211495}}, \"EndTime\": 1605487798.155301, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155289}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6917108808244977, \"sum\": 0.6917108808244977, \"min\": 0.6917108808244977}}, \"EndTime\": 1605487798.155346, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155334}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6924308558872768, \"sum\": 0.6924308558872768, \"min\": 0.6924308558872768}}, \"EndTime\": 1605487798.155398, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155384}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6908509608677456, \"sum\": 0.6908509608677456, \"min\": 0.6908509608677456}}, \"EndTime\": 1605487798.155446, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155433}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7144862474714007, \"sum\": 0.7144862474714007, \"min\": 0.7144862474714007}}, \"EndTime\": 1605487798.155493, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.15548}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7131506129673549, \"sum\": 0.7131506129673549, \"min\": 0.7131506129673549}}, \"EndTime\": 1605487798.155597, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155579}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7118759482247489, \"sum\": 0.7118759482247489, \"min\": 0.7118759482247489}}, \"EndTime\": 1605487798.155649, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155635}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7141510009765625, \"sum\": 0.7141510009765625, \"min\": 0.7141510009765625}}, \"EndTime\": 1605487798.1557, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487798.155685}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #quality_metric: host=algo-1, epoch=2, train binary_classification_cross_entropy_objective =0.661246021816\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #early_stopping_criteria_metric: host=algo-1, epoch=2, criteria=binary_classification_cross_entropy_objective, value=0.541334490095\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] Epoch 2: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] Saving model for epoch: 2\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] Saved checkpoint to \"/tmp/tmpnQ6QSn/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #progress_metric: host=algo-1, completed 20 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Total Batches Seen\": {\"count\": 1, \"max\": 33, \"sum\": 33.0, \"min\": 33}, \"Total Records Seen\": {\"count\": 1, \"max\": 6124, \"sum\": 6124.0, \"min\": 6124}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 5, \"sum\": 5.0, \"min\": 5}}, \"EndTime\": 1605487798.16599, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1605487797.862168}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #throughput_metric: host=algo-1, train throughput=4872.39994729 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:58.499] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 10, \"duration\": 332, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6534639957972935, \"sum\": 0.6534639957972935, \"min\": 0.6534639957972935}}, \"EndTime\": 1605487798.499693, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.499596}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6475389862060547, \"sum\": 0.6475389862060547, \"min\": 0.6475389862060547}}, \"EndTime\": 1605487798.499786, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.499765}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6750165448869978, \"sum\": 0.6750165448869978, \"min\": 0.6750165448869978}}, \"EndTime\": 1605487798.499859, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.499839}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6489004843575614, \"sum\": 0.6489004843575614, \"min\": 0.6489004843575614}}, \"EndTime\": 1605487798.499929, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.499909}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5380184718540737, \"sum\": 0.5380184718540737, \"min\": 0.5380184718540737}}, \"EndTime\": 1605487798.499996, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.499977}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5387565394810268, \"sum\": 0.5387565394810268, \"min\": 0.5387565394810268}}, \"EndTime\": 1605487798.500062, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500044}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5379090826851981, \"sum\": 0.5379090826851981, \"min\": 0.5379090826851981}}, \"EndTime\": 1605487798.500127, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500109}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5385331780569894, \"sum\": 0.5385331780569894, \"min\": 0.5385331780569894}}, \"EndTime\": 1605487798.500197, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500177}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6682981218610491, \"sum\": 0.6682981218610491, \"min\": 0.6682981218610491}}, \"EndTime\": 1605487798.50026, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500242}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6538470023018973, \"sum\": 0.6538470023018973, \"min\": 0.6538470023018973}}, \"EndTime\": 1605487798.500332, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500315}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6477692249843052, \"sum\": 0.6477692249843052, \"min\": 0.6477692249843052}}, \"EndTime\": 1605487798.500392, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500375}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6618269566127232, \"sum\": 0.6618269566127232, \"min\": 0.6618269566127232}}, \"EndTime\": 1605487798.500453, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500436}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.537300921848842, \"sum\": 0.537300921848842, \"min\": 0.537300921848842}}, \"EndTime\": 1605487798.500541, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.50052}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5378306252615792, \"sum\": 0.5378306252615792, \"min\": 0.5378306252615792}}, \"EndTime\": 1605487798.500611, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500592}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5379569462367466, \"sum\": 0.5379569462367466, \"min\": 0.5379569462367466}}, \"EndTime\": 1605487798.500672, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500655}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5376369094848633, \"sum\": 0.5376369094848633, \"min\": 0.5376369094848633}}, \"EndTime\": 1605487798.500737, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500718}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6640874590192523, \"sum\": 0.6640874590192523, \"min\": 0.6640874590192523}}, \"EndTime\": 1605487798.500801, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500782}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6519358934674944, \"sum\": 0.6519358934674944, \"min\": 0.6519358934674944}}, \"EndTime\": 1605487798.500863, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500845}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6682586996895926, \"sum\": 0.6682586996895926, \"min\": 0.6682586996895926}}, \"EndTime\": 1605487798.500924, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500906}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6791258566720145, \"sum\": 0.6791258566720145, \"min\": 0.6791258566720145}}, \"EndTime\": 1605487798.500986, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.500968}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6526445225306919, \"sum\": 0.6526445225306919, \"min\": 0.6526445225306919}}, \"EndTime\": 1605487798.501048, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501031}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6434929384504046, \"sum\": 0.6434929384504046, \"min\": 0.6434929384504046}}, \"EndTime\": 1605487798.501108, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.50109}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6414096723284041, \"sum\": 0.6414096723284041, \"min\": 0.6414096723284041}}, \"EndTime\": 1605487798.501171, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501153}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6472881753104074, \"sum\": 0.6472881753104074, \"min\": 0.6472881753104074}}, \"EndTime\": 1605487798.501235, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501216}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.691613017490932, \"sum\": 0.691613017490932, \"min\": 0.691613017490932}}, \"EndTime\": 1605487798.501298, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501279}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6926882934570312, \"sum\": 0.6926882934570312, \"min\": 0.6926882934570312}}, \"EndTime\": 1605487798.50136, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501342}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.69361449105399, \"sum\": 0.69361449105399, \"min\": 0.69361449105399}}, \"EndTime\": 1605487798.501423, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501405}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6918858773367745, \"sum\": 0.6918858773367745, \"min\": 0.6918858773367745}}, \"EndTime\": 1605487798.501484, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501465}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6919941166469029, \"sum\": 0.6919941166469029, \"min\": 0.6919941166469029}}, \"EndTime\": 1605487798.501545, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501527}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6936340659005301, \"sum\": 0.6936340659005301, \"min\": 0.6936340659005301}}, \"EndTime\": 1605487798.501606, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501589}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6952329799107143, \"sum\": 0.6952329799107143, \"min\": 0.6952329799107143}}, \"EndTime\": 1605487798.501667, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501649}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6904227447509765, \"sum\": 0.6904227447509765, \"min\": 0.6904227447509765}}, \"EndTime\": 1605487798.501729, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.501711}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #quality_metric: host=algo-1, epoch=3, train binary_classification_cross_entropy_objective =0.653463995797\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #early_stopping_criteria_metric: host=algo-1, epoch=3, criteria=binary_classification_cross_entropy_objective, value=0.537300921849\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] Epoch 3: Loss improved. Updating best model\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] Saving model for epoch: 3\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] Saved checkpoint to \"/tmp/tmp68NKfd/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #progress_metric: host=algo-1, completed 26 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Total Batches Seen\": {\"count\": 1, \"max\": 41, \"sum\": 41.0, \"min\": 41}, \"Total Records Seen\": {\"count\": 1, \"max\": 7605, \"sum\": 7605.0, \"min\": 7605}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 6, \"sum\": 6.0, \"min\": 6}}, \"EndTime\": 1605487798.51064, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1605487798.166458}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #throughput_metric: host=algo-1, train throughput=4301.29620174 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:58.778] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 12, \"duration\": 267, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.645896235874721, \"sum\": 0.645896235874721, \"min\": 0.645896235874721}}, \"EndTime\": 1605487798.77903, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.778921}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6399400929042272, \"sum\": 0.6399400929042272, \"min\": 0.6399400929042272}}, \"EndTime\": 1605487798.779119, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779098}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6665884726388114, \"sum\": 0.6665884726388114, \"min\": 0.6665884726388114}}, \"EndTime\": 1605487798.779191, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779171}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6412640162876674, \"sum\": 0.6412640162876674, \"min\": 0.6412640162876674}}, \"EndTime\": 1605487798.779259, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.77924}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5396884645734514, \"sum\": 0.5396884645734514, \"min\": 0.5396884645734514}}, \"EndTime\": 1605487798.779328, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779308}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5400857925415039, \"sum\": 0.5400857925415039, \"min\": 0.5400857925415039}}, \"EndTime\": 1605487798.779394, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779376}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5395366559709821, \"sum\": 0.5395366559709821, \"min\": 0.5395366559709821}}, \"EndTime\": 1605487798.779459, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779441}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5400541687011718, \"sum\": 0.5400541687011718, \"min\": 0.5400541687011718}}, \"EndTime\": 1605487798.77952, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779502}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6601988111223493, \"sum\": 0.6601988111223493, \"min\": 0.6601988111223493}}, \"EndTime\": 1605487798.779577, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779559}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6459204374040877, \"sum\": 0.6459204374040877, \"min\": 0.6459204374040877}}, \"EndTime\": 1605487798.779642, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779625}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6406216376168388, \"sum\": 0.6406216376168388, \"min\": 0.6406216376168388}}, \"EndTime\": 1605487798.779707, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779689}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6538217163085938, \"sum\": 0.6538217163085938, \"min\": 0.6538217163085938}}, \"EndTime\": 1605487798.77976, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779744}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5381442533220564, \"sum\": 0.5381442533220564, \"min\": 0.5381442533220564}}, \"EndTime\": 1605487798.779818, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779802}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5382865033830915, \"sum\": 0.5382865033830915, \"min\": 0.5382865033830915}}, \"EndTime\": 1605487798.77988, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779863}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.538368274143764, \"sum\": 0.538368274143764, \"min\": 0.538368274143764}}, \"EndTime\": 1605487798.779933, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779916}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5382451248168946, \"sum\": 0.5382451248168946, \"min\": 0.5382451248168946}}, \"EndTime\": 1605487798.77998, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.779964}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6589230019705636, \"sum\": 0.6589230019705636, \"min\": 0.6589230019705636}}, \"EndTime\": 1605487798.780045, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780027}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6482746832711356, \"sum\": 0.6482746832711356, \"min\": 0.6482746832711356}}, \"EndTime\": 1605487798.780104, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780086}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6626716722760881, \"sum\": 0.6626716722760881, \"min\": 0.6626716722760881}}, \"EndTime\": 1605487798.780157, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780142}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6718192400251116, \"sum\": 0.6718192400251116, \"min\": 0.6718192400251116}}, \"EndTime\": 1605487798.780222, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780204}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.655323622567313, \"sum\": 0.655323622567313, \"min\": 0.655323622567313}}, \"EndTime\": 1605487798.780278, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780262}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.663215811593192, \"sum\": 0.663215811593192, \"min\": 0.663215811593192}}, \"EndTime\": 1605487798.780335, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780319}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6645681435721261, \"sum\": 0.6645681435721261, \"min\": 0.6645681435721261}}, \"EndTime\": 1605487798.7804, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780382}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6606042044503349, \"sum\": 0.6606042044503349, \"min\": 0.6606042044503349}}, \"EndTime\": 1605487798.780455, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780439}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6932119532993861, \"sum\": 0.6932119532993861, \"min\": 0.6932119532993861}}, \"EndTime\": 1605487798.780534, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780492}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6929877362932477, \"sum\": 0.6929877362932477, \"min\": 0.6929877362932477}}, \"EndTime\": 1605487798.780591, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780574}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6935525730678014, \"sum\": 0.6935525730678014, \"min\": 0.6935525730678014}}, \"EndTime\": 1605487798.780647, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.78063}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6926588330950055, \"sum\": 0.6926588330950055, \"min\": 0.6926588330950055}}, \"EndTime\": 1605487798.780707, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780689}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.706327144077846, \"sum\": 0.706327144077846, \"min\": 0.706327144077846}}, \"EndTime\": 1605487798.780774, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780756}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7074032374790736, \"sum\": 0.7074032374790736, \"min\": 0.7074032374790736}}, \"EndTime\": 1605487798.780834, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780816}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.70671021597726, \"sum\": 0.70671021597726, \"min\": 0.70671021597726}}, \"EndTime\": 1605487798.780884, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780869}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7090703582763672, \"sum\": 0.7090703582763672, \"min\": 0.7090703582763672}}, \"EndTime\": 1605487798.780952, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.780934}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #quality_metric: host=algo-1, epoch=4, train binary_classification_cross_entropy_objective =0.645896235875\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #early_stopping_criteria_metric: host=algo-1, epoch=4, criteria=binary_classification_cross_entropy_objective, value=0.538144253322\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] Saving model for epoch: 4\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] Saved checkpoint to \"/tmp/tmpvVI0M4/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #progress_metric: host=algo-1, completed 33 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Total Batches Seen\": {\"count\": 1, \"max\": 49, \"sum\": 49.0, \"min\": 49}, \"Total Records Seen\": {\"count\": 1, \"max\": 9086, \"sum\": 9086.0, \"min\": 9086}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 7, \"sum\": 7.0, \"min\": 7}}, \"EndTime\": 1605487798.789013, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1605487798.510928}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:58 INFO 140177577584448] #throughput_metric: host=algo-1, train throughput=5323.24080781 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:59.057] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 14, \"duration\": 268, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6386141204833984, \"sum\": 0.6386141204833984, \"min\": 0.6386141204833984}}, \"EndTime\": 1605487799.058029, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.057931}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6326006916591099, \"sum\": 0.6326006916591099, \"min\": 0.6326006916591099}}, \"EndTime\": 1605487799.058117, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058096}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6584548732212612, \"sum\": 0.6584548732212612, \"min\": 0.6584548732212612}}, \"EndTime\": 1605487799.058192, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058172}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6338865389142717, \"sum\": 0.6338865389142717, \"min\": 0.6338865389142717}}, \"EndTime\": 1605487799.058284, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058236}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5389403697422572, \"sum\": 0.5389403697422572, \"min\": 0.5389403697422572}}, \"EndTime\": 1605487799.058353, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058335}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5389178030831473, \"sum\": 0.5389178030831473, \"min\": 0.5389178030831473}}, \"EndTime\": 1605487799.058456, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058401}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5390537915910993, \"sum\": 0.5390537915910993, \"min\": 0.5390537915910993}}, \"EndTime\": 1605487799.058524, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058504}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.539027214050293, \"sum\": 0.539027214050293, \"min\": 0.539027214050293}}, \"EndTime\": 1605487799.058626, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058573}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6523990849086216, \"sum\": 0.6523990849086216, \"min\": 0.6523990849086216}}, \"EndTime\": 1605487799.058707, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058689}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6382802963256836, \"sum\": 0.6382802963256836, \"min\": 0.6382802963256836}}, \"EndTime\": 1605487799.058808, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058756}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6337549809047154, \"sum\": 0.6337549809047154, \"min\": 0.6337549809047154}}, \"EndTime\": 1605487799.05888, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.05886}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6460568400791713, \"sum\": 0.6460568400791713, \"min\": 0.6460568400791713}}, \"EndTime\": 1605487799.05898, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.058954}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5369071469988141, \"sum\": 0.5369071469988141, \"min\": 0.5369071469988141}}, \"EndTime\": 1605487799.059055, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059036}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5370500728062221, \"sum\": 0.5370500728062221, \"min\": 0.5370500728062221}}, \"EndTime\": 1605487799.059144, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059092}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.536905266898019, \"sum\": 0.536905266898019, \"min\": 0.536905266898019}}, \"EndTime\": 1605487799.059211, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059193}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5370151846749442, \"sum\": 0.5370151846749442, \"min\": 0.5370151846749442}}, \"EndTime\": 1605487799.059299, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059253}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6545300510951451, \"sum\": 0.6545300510951451, \"min\": 0.6545300510951451}}, \"EndTime\": 1605487799.059397, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059375}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6459315327235631, \"sum\": 0.6459315327235631, \"min\": 0.6459315327235631}}, \"EndTime\": 1605487799.059462, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059445}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6578190503801619, \"sum\": 0.6578190503801619, \"min\": 0.6578190503801619}}, \"EndTime\": 1605487799.059557, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059538}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6654592786516462, \"sum\": 0.6654592786516462, \"min\": 0.6654592786516462}}, \"EndTime\": 1605487799.059634, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059614}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6494650323050363, \"sum\": 0.6494650323050363, \"min\": 0.6494650323050363}}, \"EndTime\": 1605487799.059698, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059684}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6445305524553572, \"sum\": 0.6445305524553572, \"min\": 0.6445305524553572}}, \"EndTime\": 1605487799.059739, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.05973}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6434775652204241, \"sum\": 0.6434775652204241, \"min\": 0.6434775652204241}}, \"EndTime\": 1605487799.059774, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059765}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6461407252720424, \"sum\": 0.6461407252720424, \"min\": 0.6461407252720424}}, \"EndTime\": 1605487799.059803, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059795}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.692893818446568, \"sum\": 0.692893818446568, \"min\": 0.692893818446568}}, \"EndTime\": 1605487799.059835, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059827}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6929843793596541, \"sum\": 0.6929843793596541, \"min\": 0.6929843793596541}}, \"EndTime\": 1605487799.059863, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059856}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6927476065499442, \"sum\": 0.6927476065499442, \"min\": 0.6927476065499442}}, \"EndTime\": 1605487799.059895, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059887}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6931512887137277, \"sum\": 0.6931512887137277, \"min\": 0.6931512887137277}}, \"EndTime\": 1605487799.059928, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.05992}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6944676208496093, \"sum\": 0.6944676208496093, \"min\": 0.6944676208496093}}, \"EndTime\": 1605487799.059961, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059953}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6943283299037388, \"sum\": 0.6943283299037388, \"min\": 0.6943283299037388}}, \"EndTime\": 1605487799.059993, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.059985}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6934650639125279, \"sum\": 0.6934650639125279, \"min\": 0.6934650639125279}}, \"EndTime\": 1605487799.060026, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.060018}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6943901715959822, \"sum\": 0.6943901715959822, \"min\": 0.6943901715959822}}, \"EndTime\": 1605487799.060053, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487799.060046}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, epoch=5, train binary_classification_cross_entropy_objective =0.638614120483\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #early_stopping_criteria_metric: host=algo-1, epoch=5, criteria=binary_classification_cross_entropy_objective, value=0.536905266898\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Saving model for epoch: 5\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Saved checkpoint to \"/tmp/tmpoGMYdM/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #progress_metric: host=algo-1, completed 40 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Total Batches Seen\": {\"count\": 1, \"max\": 57, \"sum\": 57.0, \"min\": 57}, \"Total Records Seen\": {\"count\": 1, \"max\": 10567, \"sum\": 10567.0, \"min\": 10567}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}}, \"EndTime\": 1605487799.067858, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1605487798.789301}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #throughput_metric: host=algo-1, train throughput=5314.51470616 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:59.269] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 16, \"duration\": 200, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6318944713047573, \"sum\": 0.6318944713047573, \"min\": 0.6318944713047573}}, \"EndTime\": 1605487799.269265, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269159}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6258229119437081, \"sum\": 0.6258229119437081, \"min\": 0.6258229119437081}}, \"EndTime\": 1605487799.269362, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269342}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6508998053414481, \"sum\": 0.6508998053414481, \"min\": 0.6508998053414481}}, \"EndTime\": 1605487799.269431, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269414}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6270694024222238, \"sum\": 0.6270694024222238, \"min\": 0.6270694024222238}}, \"EndTime\": 1605487799.26949, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269473}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5378296879359654, \"sum\": 0.5378296879359654, \"min\": 0.5378296879359654}}, \"EndTime\": 1605487799.269556, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269537}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5376292419433594, \"sum\": 0.5376292419433594, \"min\": 0.5376292419433594}}, \"EndTime\": 1605487799.26962, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269603}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5380103737967354, \"sum\": 0.5380103737967354, \"min\": 0.5380103737967354}}, \"EndTime\": 1605487799.269684, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269665}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5377172524588448, \"sum\": 0.5377172524588448, \"min\": 0.5377172524588448}}, \"EndTime\": 1605487799.269745, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269728}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6451754651750837, \"sum\": 0.6451754651750837, \"min\": 0.6451754651750837}}, \"EndTime\": 1605487799.269807, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269789}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6312283488682339, \"sum\": 0.6312283488682339, \"min\": 0.6312283488682339}}, \"EndTime\": 1605487799.269876, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269857}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6274302237374442, \"sum\": 0.6274302237374442, \"min\": 0.6274302237374442}}, \"EndTime\": 1605487799.269952, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269934}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6388434818812779, \"sum\": 0.6388434818812779, \"min\": 0.6388434818812779}}, \"EndTime\": 1605487799.270016, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.269998}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5375400761195591, \"sum\": 0.5375400761195591, \"min\": 0.5375400761195591}}, \"EndTime\": 1605487799.27007, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270053}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5380268805367606, \"sum\": 0.5380268805367606, \"min\": 0.5380268805367606}}, \"EndTime\": 1605487799.270135, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270117}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5376483208792551, \"sum\": 0.5376483208792551, \"min\": 0.5376483208792551}}, \"EndTime\": 1605487799.270203, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270184}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5379004669189453, \"sum\": 0.5379004669189453, \"min\": 0.5379004669189453}}, \"EndTime\": 1605487799.270256, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.27024}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.651105466570173, \"sum\": 0.651105466570173, \"min\": 0.651105466570173}}, \"EndTime\": 1605487799.270308, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270292}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6448629651750837, \"sum\": 0.6448629651750837, \"min\": 0.6448629651750837}}, \"EndTime\": 1605487799.270368, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270351}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6539120047433036, \"sum\": 0.6539120047433036, \"min\": 0.6539120047433036}}, \"EndTime\": 1605487799.270419, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270402}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6602929033551898, \"sum\": 0.6602929033551898, \"min\": 0.6602929033551898}}, \"EndTime\": 1605487799.270471, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270455}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6522635051182338, \"sum\": 0.6522635051182338, \"min\": 0.6522635051182338}}, \"EndTime\": 1605487799.270523, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270507}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6556434086390904, \"sum\": 0.6556434086390904, \"min\": 0.6556434086390904}}, \"EndTime\": 1605487799.270574, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270559}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6560571943010602, \"sum\": 0.6560571943010602, \"min\": 0.6560571943010602}}, \"EndTime\": 1605487799.270625, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.27061}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6551728548322405, \"sum\": 0.6551728548322405, \"min\": 0.6551728548322405}}, \"EndTime\": 1605487799.270677, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270661}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6930604335239955, \"sum\": 0.6930604335239955, \"min\": 0.6930604335239955}}, \"EndTime\": 1605487799.270728, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270712}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6926846095493862, \"sum\": 0.6926846095493862, \"min\": 0.6926846095493862}}, \"EndTime\": 1605487799.270779, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270763}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6924631609235491, \"sum\": 0.6924631609235491, \"min\": 0.6924631609235491}}, \"EndTime\": 1605487799.270831, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270815}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6931122153145927, \"sum\": 0.6931122153145927, \"min\": 0.6931122153145927}}, \"EndTime\": 1605487799.270883, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270867}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.701703371320452, \"sum\": 0.701703371320452, \"min\": 0.701703371320452}}, \"EndTime\": 1605487799.270934, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.270919}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7016393389020648, \"sum\": 0.7016393389020648, \"min\": 0.7016393389020648}}, \"EndTime\": 1605487799.270988, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.27097}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7014774976457868, \"sum\": 0.7014774976457868, \"min\": 0.7014774976457868}}, \"EndTime\": 1605487799.271039, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.271023}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.7006882585797991, \"sum\": 0.7006882585797991, \"min\": 0.7006882585797991}}, \"EndTime\": 1605487799.27109, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.271074}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, epoch=6, train binary_classification_cross_entropy_objective =0.631894471305\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #early_stopping_criteria_metric: host=algo-1, epoch=6, criteria=binary_classification_cross_entropy_objective, value=0.53754007612\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Saving model for epoch: 6\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Saved checkpoint to \"/tmp/tmplGeW5J/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #progress_metric: host=algo-1, completed 46 % of epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Total Batches Seen\": {\"count\": 1, \"max\": 65, \"sum\": 65.0, \"min\": 65}, \"Total Records Seen\": {\"count\": 1, \"max\": 12048, \"sum\": 12048.0, \"min\": 12048}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 9, \"sum\": 9.0, \"min\": 9}}, \"EndTime\": 1605487799.27771, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1605487799.068121}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #throughput_metric: host=algo-1, train throughput=7062.37021777 records/second\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:59.497] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 18, \"duration\": 219, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6257479531424386, \"sum\": 0.6257479531424386, \"min\": 0.6257479531424386}}, \"EndTime\": 1605487799.497544, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.497435}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6196199853079659, \"sum\": 0.6196199853079659, \"min\": 0.6196199853079659}}, \"EndTime\": 1605487799.497626, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.497611}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6439534487043108, \"sum\": 0.6439534487043108, \"min\": 0.6439534487043108}}, \"EndTime\": 1605487799.497688, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.497673}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6208263560703823, \"sum\": 0.6208263560703823, \"min\": 0.6208263560703823}}, \"EndTime\": 1605487799.497724, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.497715}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5369816698346819, \"sum\": 0.5369816698346819, \"min\": 0.5369816698346819}}, \"EndTime\": 1605487799.497793, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.497774}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5370708247593471, \"sum\": 0.5370708247593471, \"min\": 0.5370708247593471}}, \"EndTime\": 1605487799.497858, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.497842}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5370299312046596, \"sum\": 0.5370299312046596, \"min\": 0.5370299312046596}}, \"EndTime\": 1605487799.497907, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.497897}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.537077010018485, \"sum\": 0.537077010018485, \"min\": 0.537077010018485}}, \"EndTime\": 1605487799.497964, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.497946}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6385475376674107, \"sum\": 0.6385475376674107, \"min\": 0.6385475376674107}}, \"EndTime\": 1605487799.498022, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498006}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6247796358380999, \"sum\": 0.6247796358380999, \"min\": 0.6247796358380999}}, \"EndTime\": 1605487799.498086, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498067}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6216473770141602, \"sum\": 0.6216473770141602, \"min\": 0.6216473770141602}}, \"EndTime\": 1605487799.498168, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.49814}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6322094345092774, \"sum\": 0.6322094345092774, \"min\": 0.6322094345092774}}, \"EndTime\": 1605487799.498229, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498212}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5382921927315848, \"sum\": 0.5382921927315848, \"min\": 0.5382921927315848}}, \"EndTime\": 1605487799.4983, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498281}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5390296718052455, \"sum\": 0.5390296718052455, \"min\": 0.5390296718052455}}, \"EndTime\": 1605487799.498362, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498344}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5381832722255162, \"sum\": 0.5381832722255162, \"min\": 0.5381832722255162}}, \"EndTime\": 1605487799.498429, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498412}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.5389727238246372, \"sum\": 0.5389727238246372, \"min\": 0.5389727238246372}}, \"EndTime\": 1605487799.498491, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498473}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.648616463797433, \"sum\": 0.648616463797433, \"min\": 0.648616463797433}}, \"EndTime\": 1605487799.498564, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498545}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6446932329450334, \"sum\": 0.6446932329450334, \"min\": 0.6446932329450334}}, \"EndTime\": 1605487799.498628, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.49861}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6509214564732143, \"sum\": 0.6509214564732143, \"min\": 0.6509214564732143}}, \"EndTime\": 1605487799.498687, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.49867}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6562464032854353, \"sum\": 0.6562464032854353, \"min\": 0.6562464032854353}}, \"EndTime\": 1605487799.498739, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498723}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6507694353376116, \"sum\": 0.6507694353376116, \"min\": 0.6507694353376116}}, \"EndTime\": 1605487799.498799, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498783}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6495787974766323, \"sum\": 0.6495787974766323, \"min\": 0.6495787974766323}}, \"EndTime\": 1605487799.498857, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498841}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6488709803989955, \"sum\": 0.6488709803989955, \"min\": 0.6488709803989955}}, \"EndTime\": 1605487799.498916, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.4989}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6496602303641184, \"sum\": 0.6496602303641184, \"min\": 0.6496602303641184}}, \"EndTime\": 1605487799.498971, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.498958}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6927409362792969, \"sum\": 0.6927409362792969, \"min\": 0.6927409362792969}}, \"EndTime\": 1605487799.49902, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.499004}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6926378740583148, \"sum\": 0.6926378740583148, \"min\": 0.6926378740583148}}, \"EndTime\": 1605487799.49908, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.499063}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6924281311035156, \"sum\": 0.6924281311035156, \"min\": 0.6924281311035156}}, \"EndTime\": 1605487799.499148, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.499131}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.692623040335519, \"sum\": 0.692623040335519, \"min\": 0.692623040335519}}, \"EndTime\": 1605487799.499212, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.499194}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6968828255789621, \"sum\": 0.6968828255789621, \"min\": 0.6968828255789621}}, \"EndTime\": 1605487799.499251, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.499242}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6975470188685826, \"sum\": 0.6975470188685826, \"min\": 0.6975470188685826}}, \"EndTime\": 1605487799.499301, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.499284}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6964437757219587, \"sum\": 0.6964437757219587, \"min\": 0.6964437757219587}}, \"EndTime\": 1605487799.499358, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.499347}\n", "\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"train_binary_classification_cross_entropy_objective\": {\"count\": 1, \"max\": 0.6981465693882534, \"sum\": 0.6981465693882534, \"min\": 0.6981465693882534}}, \"EndTime\": 1605487799.499393, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.499384}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, epoch=7, train binary_classification_cross_entropy_objective =0.625747953142\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #early_stopping_criteria_metric: host=algo-1, epoch=7, criteria=binary_classification_cross_entropy_objective, value=0.536981669835\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Saving model for epoch: 7\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Saved checkpoint to \"/tmp/tmpP1E6W3/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Early stop condition met. Stopping training.\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #progress_metric: host=algo-1, completed 100 % epochs\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Total Batches Seen\": {\"count\": 1, \"max\": 73, \"sum\": 73.0, \"min\": 73}, \"Total Records Seen\": {\"count\": 1, \"max\": 13529, \"sum\": 13529.0, \"min\": 13529}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 1481, \"sum\": 1481.0, \"min\": 1481}, \"Reset Count\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}}, \"EndTime\": 1605487799.506449, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1605487799.277983}\n", "\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #throughput_metric: host=algo-1, train throughput=6478.67257682 records/second\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 WARNING 140177577584448] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 WARNING 140177577584448] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:59.508] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 20, \"duration\": 0, \"num_examples\": 1, \"num_bytes\": 11200}\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:59.609] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 23, \"duration\": 99, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m[2020-11-16 00:49:59.642] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 25, \"duration\": 23, \"num_examples\": 8, \"num_bytes\": 82936}\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #train_score (algo-1) : ('binary_classification_cross_entropy_objective', 0.5411434675856751)\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #train_score (algo-1) : ('binary_classification_accuracy', 0.7332883187035787)\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #train_score (algo-1) : ('binary_f_1.000', 0.7228070175438597)\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #train_score (algo-1) : ('precision', 0.7584683357879234)\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #train_score (algo-1) : ('recall', 0.6903485254691689)\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #train_score (algo-1) : ('roc_auc_score', 0.7916616512556766)\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, train binary_classification_cross_entropy_objective =0.541143467586\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, train binary_classification_accuracy =0.733288318704\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, train binary_f_1.000 =0.722807017544\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, train precision =0.758468335788\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, train recall =0.690348525469\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] #quality_metric: host=algo-1, train roc_auc_score =0.791661651256\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Best model found for hyperparameters: {\"lr_scheduler_step\": 10, \"wd\": 0.01, \"optimizer\": \"adam\", \"lr_scheduler_factor\": 0.99, \"l1\": 0.0, \"learning_rate\": 0.1, \"lr_scheduler_minimum_lr\": 1e-05}\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Saved checkpoint to \"/tmp/tmphbudmw/mx-mod-0000.params\"\u001b[0m\n", "\u001b[34m[11/16/2020 00:49:59 INFO 140177577584448] Test data is not provided.\u001b[0m\n", "\u001b[34m#metrics {\"Metrics\": {\"totaltime\": {\"count\": 1, \"max\": 2733.5920333862305, \"sum\": 2733.5920333862305, \"min\": 2733.5920333862305}, \"finalize.time\": {\"count\": 1, \"max\": 141.19410514831543, \"sum\": 141.19410514831543, \"min\": 141.19410514831543}, \"initialize.time\": {\"count\": 1, \"max\": 167.39487648010254, \"sum\": 167.39487648010254, \"min\": 167.39487648010254}, \"check_early_stopping.time\": {\"count\": 8, \"max\": 1.0981559753417969, \"sum\": 4.754304885864258, \"min\": 0.19216537475585938}, \"setuptime\": {\"count\": 1, \"max\": 25.877952575683594, \"sum\": 25.877952575683594, \"min\": 25.877952575683594}, \"update.time\": {\"count\": 8, \"max\": 341.6900634765625, \"sum\": 2144.4549560546875, \"min\": 207.3230743408203}, \"epochs\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}}, \"EndTime\": 1605487799.652723, \"Dimensions\": {\"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1605487797.164191}\n", "\u001b[0m\n", "\n", "2020-11-16 00:50:10 Uploading - Uploading generated training model\n", "2020-11-16 00:50:10 Completed - Training job completed\n", "Training seconds: 51\n", "Billable seconds: 51\n", "-------------!" ] } ], "source": [ "import io\n", "import os\n", "import boto3\n", "import sagemaker\n", "import sagemaker.amazon.common as smac\n", "\n", "role = sagemaker.get_execution_role()\n", "\n", "prefix = 'sagemaker'\n", "\n", "for position in predictions:\n", " buffer = io.BytesIO()\n", " smac.write_numpy_to_dense_tensor(\n", " buffer, \n", " predictions[position]['training_vectors'],\n", " predictions[position]['training_labels']\n", " )\n", " buffer.seek(0)\n", "\n", " key = '{}-training-data'.format(position)\n", "\n", " boto3.resource('s3').Bucket(bucket).Object(os.path.join(prefix, 'train', key)).upload_fileobj(buffer)\n", "\n", " s3_train_data = 's3://{}/{}/train/{}'.format(bucket, prefix, key)\n", " output_location = 's3://{}/{}/{}/output'.format(bucket, prefix, position)\n", "\n", " container = sagemaker.image_uris.retrieve('linear-learner', boto3.Session().region_name)\n", "\n", " session = sagemaker.Session()\n", "\n", " linear = sagemaker.estimator.Estimator(\n", " container,\n", " role, \n", " instance_count = 1, \n", " instance_type = 'ml.c4.xlarge',\n", " output_path = output_location,\n", " sagemaker_session = session\n", " )\n", "\n", " num_features = predictions[position]['training_vectors'][0].shape[0]\n", "\n", " linear.set_hyperparameters(\n", " feature_dim = num_features,\n", " predictor_type = 'binary_classifier',\n", " mini_batch_size = 200\n", " )\n", "\n", " linear.fit({ 'train': s3_train_data })\n", "\n", " predictions[position]['predictor'] = linear.deploy(\n", " initial_instance_count = 1,\n", " instance_type = 'ml.m4.xlarge'\n", " )\n", "\n", " predictions[position]['predictor'].serializer = sagemaker.serializers.CSVSerializer()\n", " predictions[position]['predictor'].deserializer = sagemaker.deserializers.JSONDeserializer()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Evaluate the models\n", "\n", "Now that we've trained our models, it's time to see if they're any good. To do this, we first sum the true positives (TP), true negatives (TN), false negatives (FN), and false positives (FP) for each position using the 20% verification data we saved above.\n", "\n", "A true positive means the model said to play the player and it was the right call because they scored in the above the start quantile.\n", "\n", "A true negative means the model said don't play the player and it was the right call because they scored below the start quantile.\n", "\n", "A false positive means the model said to play the player but, was wrong because the player scored below the start quantile.\n", "\n", "A false negative means the model said don't play the player but, was wrong because the player scored above the start quantile.\n", "\n", "With these counts, we can calculate 3 very important measures for our model:\n", "1. **Precision:** This is the ratio of correct positive predictions to all positive predictions and is calculated as TP / (TP + FP). In plain English, it's the number of players we correctly predicted (TP) divided by all of the players the model thought we should play (TP + FP). Put differently, this tells you how precise the model positive predictions are.\n", "\n", "2. **Recall:** This is the ratio of correct positive predictions to positive examples and is calcualted as TP / (TP + FN). In plain English, it's the number of players we correctly predicted (TP) divided by all of the players we should've played (TP + FN). Put differently, this is a measure of how often the model recalls the a correct prediction.\n", "\n", "3. **F1 Score:** Also known as the Harmonic Mean, this measures the overall quality of the model by combining Precision and Recall. It's calculated as 2 * (Precision * Recall) / (Precision + Recall). There's a tension between Precision and Recall because Precision is a function of FP and Recall is a function of FN. The F1 Score is a way to look at these values together.\n", "\n", "Without significant more research than is done in this notebook, it's hard to say what a \"good\" score is for Precision, Recall, and the F1 Score. For Fantasy Football, it's important for both Predicion and Recall to be as high as possible because you don't want to play players that you shouldn't (Precision) and you dont want to miss players that will score a lot of points (Recall). \n", "\n", "You'll see below that we have intuitively good measures Precision, Recall, and F1 scores based on player position. But, there's a lot of room for improvement here." ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [], "source": [ "for position in predictions:\n", " tp = 0\n", " tn = 0\n", " fp = 0\n", " fn = 0\n", "\n", " verification_vectors = predictions[position]['verification_vectors']\n", " verification_labels = predictions[position]['verification_labels']\n", " \n", " predictor = predictions[position]['predictor']\n", " \n", " for i in range(0, verification_vectors.shape[0] - 1):\n", " result = predictor.predict(verification_vectors[i])['predictions'][0]\n", " \n", " if (result['predicted_label'] == 1) and (verification_labels[i] == 1):\n", " tp += 1\n", "\n", " if (result['predicted_label'] == 0) and (verification_labels[i] == 0):\n", " tn += 1\n", " \n", " if (result['predicted_label'] == 1) and (verification_labels[i] == 0):\n", " fp += 1\n", "\n", " if (result['predicted_label'] == 0) and (verification_labels[i] == 1):\n", " fn += 1\n", " \n", " precision = tp / (tp + fp)\n", " recall = tp / (tp + fn)\n", " f1score = 2 * (precision * recall) / (precision + recall)\n", " \n", " predictions[position]['verification_tp'] = tp\n", " predictions[position]['verification_tn'] = tn\n", " predictions[position]['verification_fp'] = fp\n", " predictions[position]['verification_fn'] = fn\n", " \n", " predictions[position]['precision'] = precision\n", " predictions[position]['recall'] = recall\n", " predictions[position]['f1score'] = f1score" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 4 Complete: Evaluate your trained models" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "***\n", "QB: Precision: 72.08%, Recall: 72.86%, F1 Score: 72.47%\n", "Target points: 17.32\n", "***\n", "\n", "***\n", "WR: Precision: 78.4%, Recall: 85.81%, F1 Score: 81.94%\n", "Target points: 9.8\n", "***\n", "\n", "***\n", "RB: Precision: 81.79%, Recall: 73.66%, F1 Score: 77.51%\n", "Target points: 10.84\n", "***\n", "\n", "***\n", "TE: Precision: 77.05%, Recall: 68.12%, F1 Score: 72.31%\n", "Target points: 9.6\n", "***\n", "\n" ] } ], "source": [ "for position in predictions:\n", " print('***\\n{}: Precision: {}%, Recall: {}%, F1 Score: {}%\\nTarget points: {}\\n***\\n'.format(\n", " position, \n", " round(predictions[position]['precision'] * 100.0, 2), \n", " round(predictions[position]['recall'] * 100.0, 2), \n", " round(predictions[position]['f1score'] * 100.0, 2),\n", " round(predictions[position]['minimum_points_wanted'], 2)\n", " ))\n", " " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 5 Start: Collect 2019 data to test lineup optimization\n", "\n", "Now that we've trained a model to tell us which players have upside, it's time to generate lineups.\n", "\n", "The first step is to create a data frame with player data from 2019 along with their daily fantasy sports salary, whether or not the model thinks we should consider playing them, their actual total points scored that week, and the points that Stats Perform predicted they'd score that week." ] }, { "cell_type": "code", "execution_count": 39, "metadata": {}, "outputs": [], "source": [ "scoring_2019_df = scoring_df[(scoring_df['season'] == 2019)][['player_id', 'actual_points', 'week', 'fixture_id', 'implied_game_score', 'opponent_implied_game_score']]" ] }, { "cell_type": "code", "execution_count": 40, "metadata": {}, "outputs": [], "source": [ "salary_df = pd.read_json(salary_location)" ] }, { "cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [], "source": [ "predictions_df = pd.read_json(fantasy_predictions_location)" ] }, { "cell_type": "code", "execution_count": 42, "metadata": {}, "outputs": [], "source": [ "scoring_2019_df = pd.merge(scoring_2019_df, salary_df, how = 'left', left_on = [ 'week', 'fixture_id', 'player_id' ], right_on = [ 'week', 'fixture_id', 'player_id' ])\n", "scoring_2019_df = pd.merge(scoring_2019_df, predictions_df, how = 'left', left_on = [ 'week', 'fixture_id', 'player_id' ], right_on = ['week', 'fixture_id', 'player_id' ])" ] }, { "cell_type": "code", "execution_count": 43, "metadata": {}, "outputs": [], "source": [ "scoring_2019_df = scoring_2019_df.rename(columns = {\n", " 'position_x': 'position', \n", " 'draftkings_salary': 'salary', \n", " 'attempts': 'pass_atts',\n", " 'pass_yards': 'pass_yds',\n", " 'pass_touchdowns': 'pass_tds',\n", " 'rush_yards': 'rush_yds',\n", " 'rush_touchdowns': 'rush_tds',\n", " 'receptions': 'recs',\n", " 'rush_attempts': 'rush_atts',\n", " 'reception_yards': 'rec_yds',\n", " 'reception_touchdowns': 'rec_tds'\n", "})" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We don't have predictions for `rec_targets` which we use as a feature in our model so, we use receptions as a proxy." ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [], "source": [ "scoring_2019_df['rec_targets'] = scoring_2019_df['recs']" ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [], "source": [ "scoring_2019_df['predicted_points'] = scoring_2019_df.apply(points_scored, axis = 1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Grab the columns we need here and drop duplicates from merging with the salary data due to players being available at multiple positions in some of the Stats Perform data." ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [], "source": [ "scoring_2019_df = scoring_2019_df[[\n", " 'player_id',\n", " 'position',\n", " 'week',\n", " 'fixture_id',\n", " 'predicted_points',\n", " 'actual_points',\n", " 'salary',\n", " 'pass_atts',\n", " 'rush_atts',\n", " 'rec_targets',\n", " 'implied_game_score', \n", " 'opponent_implied_game_score'\n", "]].drop_duplicates()" ] }, { "cell_type": "code", "execution_count": 47, "metadata": {}, "outputs": [], "source": [ "scoring_2019_df = scoring_2019_df.dropna(subset = [ 'position', 'predicted_points', 'actual_points', 'salary' ]).fillna(0)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This function will call our predictor for each position to infer if the player has upside based using the features defined earlier in the notebook for each position." ] }, { "cell_type": "code", "execution_count": 48, "metadata": {}, "outputs": [], "source": [ "def infer_upside(player):\n", " position = player['position']\n", " \n", " features = predictions[position]['features']\n", " predictor = predictions[position]['predictor']\n", " \n", " prediction_vector = player[features].values\n", " result = predictor.predict(prediction_vector)['predictions'][0]\n", " \n", " return result['predicted_label']" ] }, { "cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [], "source": [ "scoring_2019_df['has_upside'] = scoring_2019_df.apply(infer_upside, axis = 1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 5 Complete: You're ready to optimize\n", "\n", "We see a sample of the data here that's ready to be used to find optimal lineups." ] }, { "cell_type": "code", "execution_count": 50, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
player_idpositionweekfixture_idpredicted_pointsactual_pointssalarypass_attsrush_attsrec_targetsimplied_game_scoreopponent_implied_game_scorehas_upside
7539749185TE4214208419.1427799.405700.00.000000.0000009.24974725.5022.501
10119880033RB16214210615.33209819.904400.00.0000011.7026554.03043923.5026.501
9381837820WR1021420433.5827990.003400.00.000000.1502701.24590626.7520.250
11013945633WR9214213912.46061625.005700.00.000000.0000003.62324927.5023.500
9622840760RB10214207918.6526259.207000.00.0000021.5759221.64731627.7516.751
223216263QB16214211021.22432419.966200.035.923591.0780180.00000027.7520.251
4530606055TE1521420494.5018185.202600.00.000000.0000001.95885326.2515.750
4296602241WR4214207810.17188017.004100.00.000000.0000003.27083714.5029.000
10794922026WR221421902.49548216.104800.00.000000.0000000.81169030.0023.000
111121049915WR1321423073.36030114.503500.00.000000.0000000.96037025.7522.750
\n", "
" ], "text/plain": [ " player_id position week fixture_id predicted_points actual_points \\\n", "7539 749185 TE 4 2142084 19.142779 9.40 \n", "10119 880033 RB 16 2142106 15.332098 19.90 \n", "9381 837820 WR 10 2142043 3.582799 0.00 \n", "11013 945633 WR 9 2142139 12.460616 25.00 \n", "9622 840760 RB 10 2142079 18.652625 9.20 \n", "223 216263 QB 16 2142110 21.224324 19.96 \n", "4530 606055 TE 15 2142049 4.501818 5.20 \n", "4296 602241 WR 4 2142078 10.171880 17.00 \n", "10794 922026 WR 2 2142190 2.495482 16.10 \n", "11112 1049915 WR 13 2142307 3.360301 14.50 \n", "\n", " salary pass_atts rush_atts rec_targets implied_game_score \\\n", "7539 5700.0 0.00000 0.000000 9.249747 25.50 \n", "10119 4400.0 0.00000 11.702655 4.030439 23.50 \n", "9381 3400.0 0.00000 0.150270 1.245906 26.75 \n", "11013 5700.0 0.00000 0.000000 3.623249 27.50 \n", "9622 7000.0 0.00000 21.575922 1.647316 27.75 \n", "223 6200.0 35.92359 1.078018 0.000000 27.75 \n", "4530 2600.0 0.00000 0.000000 1.958853 26.25 \n", "4296 4100.0 0.00000 0.000000 3.270837 14.50 \n", "10794 4800.0 0.00000 0.000000 0.811690 30.00 \n", "11112 3500.0 0.00000 0.000000 0.960370 25.75 \n", "\n", " opponent_implied_game_score has_upside \n", "7539 22.50 1 \n", "10119 26.50 1 \n", "9381 20.25 0 \n", "11013 23.50 0 \n", "9622 16.75 1 \n", "223 20.25 1 \n", "4530 15.75 0 \n", "4296 29.00 0 \n", "10794 23.00 0 \n", "11112 22.75 0 " ] }, "execution_count": 50, "metadata": {}, "output_type": "execute_result" } ], "source": [ "scoring_2019_df.sample(n = 10)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 6 Start: Compare full player universe to the upside player universe based on odds\n", "\n", "In this step, we compare the size of the full universe of players to the size of the universe of players with upside. The upside universe is significantly (100s of trillions of combinations) smaller so, it's much easier to generate the top N lineups you want to play using a brute force algorithm. " ] }, { "cell_type": "code", "execution_count": 62, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 62, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "from math import factorial\n", "import matplotlib.pyplot as pyplot\n", "\n", "def nCr(n, r):\n", " return factorial(n) / (factorial(r) * factorial(n - r))\n", "\n", "combinations_full = []\n", "combinations_upside = []\n", "\n", "weeks = [*range(1, 17)]\n", "\n", "for week in weeks:\n", " num_qbs_full_universe = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['position'] == 'QB')].shape[0]\n", " num_rbs_full_universe = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['position'] == 'RB')].shape[0]\n", " num_wrs_full_universe = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['position'] == 'WR')].shape[0]\n", " num_tes_full_universe = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['position'] == 'TE')].shape[0]\n", "\n", " combinations_full_universe = num_qbs_full_universe * nCr(num_rbs_full_universe, 2) * nCr(num_wrs_full_universe, 3) * num_tes_full_universe * (num_rbs_full_universe + num_wrs_full_universe + num_tes_full_universe)\n", " combinations_full.append(combinations_full_universe)\n", " \n", " num_qbs_upside_universe = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['position'] == 'QB') & (scoring_2019_df['has_upside'] == 1)].shape[0]\n", " num_rbs_upside_universe = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['position'] == 'RB') & (scoring_2019_df['has_upside'] == 1)].shape[0]\n", " num_wrs_upside_universe = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['position'] == 'WR') & (scoring_2019_df['has_upside'] == 1)].shape[0]\n", " num_tes_upside_universe = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['position'] == 'TE') & (scoring_2019_df['has_upside'] == 1)].shape[0]\n", " \n", " combinations_upside_universe = num_qbs_upside_universe * nCr(num_rbs_upside_universe, 2) * nCr(num_wrs_upside_universe, 3) * num_tes_upside_universe * (num_rbs_upside_universe + num_wrs_upside_universe + num_tes_upside_universe)\n", " combinations_upside.append(combinations_upside_universe)\n", " \n", "pyplot.figure(figsize=(18, 6))\n", "pyplot.yscale('log')\n", "pyplot.plot(weeks, combinations_full, color = 'blue', linewidth = 3, label = \"Full Universe\")\n", "pyplot.plot(weeks, combinations_upside, color = 'green', linewidth = 3, label = \"Upside Universe\")\n", "pyplot.text(1, 1000000000000, 'Week 1 combinations from full universe = 420,264,928,923,840\\nWeek 1 combinations from upside universe = 786,240')\n", "pyplot.xlabel('Week')\n", "pyplot.ylabel('Lineup Combinations (log scale)')\n", "pyplot.title('Lineup Combinations of Full Universe vs Upside Universe (log scale)')\n", "pyplot.xticks(weeks)\n", "\n", "pyplot.legend()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 6 Complete: Using odds to filter the player universe makes life easier" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 7 Start: Test the smaller player universe against the full universe" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Optimize our lineups\n", "\n", "In this section of the notebook, we're finally going to optimize our lineups. We're going to do this using the linear optimizer library [PuLP](https://github.com/coin-or/pulp). A linear optimization is an easy way to get the overall best lineup by maximizing points, putting a constraint on the salary of $50,000, and putting a constraint on the number of players at each position. It'll give us an overall view of how the models perform.\n", "\n", "When playing daily fantasy sports you likely want more than just a single lineup so that you get exposure to more players. This expands your upside because, as we've said, it's hard to predict who will actually perform well. It also limits the downside of selecting a player that doesn't end up scoring any points. \n", "\n", "As you saw above, when you look at the full universe of players there are trillions of lineups making it much hard to find the top 10, 50, or 100 lineups to play. However, by limiting the universe to players with upside from the odds, it's much easier to find the top 10, 50, or however many lineups you need. So, all that remains is to see if the lineups generated by the upside player universe are able to compete with the lineups generated by the full player universe." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Install PuLP" ] }, { "cell_type": "code", "execution_count": 52, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Collecting package metadata (current_repodata.json): done\n", "Solving environment: / \n", "The environment is inconsistent, please check the package plan carefully\n", "The following packages are causing the inconsistency:\n", "\n", " - defaults/linux-64::pandas==1.0.1=py36h0573a6f_0\n", " - defaults/linux-64::scikit-learn==0.22.1=py36hd81dba3_0\n", " - defaults/linux-64::bkcharts==0.2=py36_0\n", " - defaults/linux-64::pytest-arraydiff==0.3=py36h39e3cac_0\n", " - defaults/linux-64::bottleneck==1.3.2=py36heb32a55_0\n", " - defaults/linux-64::pywavelets==1.1.1=py36h7b6447c_0\n", " - defaults/noarch::pytest-astropy==0.8.0=py_0\n", " - defaults/linux-64::numexpr==2.7.1=py36h423224d_0\n", " - defaults/linux-64::h5py==2.10.0=py36h7918eee_0\n", " - defaults/linux-64::bokeh==1.4.0=py36_0\n", " - defaults/linux-64::numpy-base==1.18.1=py36hde5b4d6_1\n", " - defaults/linux-64::astropy==4.0=py36h7b6447c_0\n", " - defaults/linux-64::patsy==0.5.1=py36_0\n", " - defaults/linux-64::scikit-image==0.16.2=py36h0573a6f_0\n", " - defaults/linux-64::matplotlib-base==3.1.3=py36hef1b27d_0\n", " - defaults/linux-64::imageio==2.6.1=py36_0\n", " - defaults/linux-64::pytables==3.6.1=py36h71ec239_0\n", " - defaults/linux-64::mkl_fft==1.0.15=py36ha843d7b_0\n", " - defaults/linux-64::statsmodels==0.11.0=py36h7b6447c_0\n", " - defaults/noarch::seaborn==0.10.0=py_0\n", " - defaults/linux-64::numba==0.48.0=py36h0573a6f_0\n", " - defaults/linux-64::scipy==1.4.1=py36h0b6359f_0\n", " - defaults/noarch::pytest-doctestplus==0.5.0=py_0\n", " - defaults/linux-64::mkl_random==1.1.0=py36hd6b4f25_0\n", " - defaults/noarch::dask==2.11.0=py_0\n", " - defaults/linux-64::matplotlib==3.1.3=py36_0\n", " - defaults/linux-64::numpy==1.18.1=py36h4f9e942_0\n", "failed with initial frozen solve. Retrying with flexible solve.\n", "Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.\n", "Collecting package metadata (repodata.json): done\n", "Solving environment: done\n", "\n", "\n", "==> WARNING: A newer version of conda exists. <==\n", " current version: 4.8.3\n", " latest version: 4.9.2\n", "\n", "Please update conda by running\n", "\n", " $ conda update -n base conda\n", "\n", "\n", "\n", "# All requested packages already installed.\n", "\n", "\n", "Note: you may need to restart the kernel to use updated packages.\n" ] } ], "source": [ "%conda install -c conda-forge pulp" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is a helper function to get the data ready for our linear optimizer. We put the salaries and points into a dict so that for each position, we have each player along with their salary and points to pass to the linear optimizer. " ] }, { "cell_type": "code", "execution_count": 53, "metadata": {}, "outputs": [], "source": [ "def salaries_and_points_from_df(df, points_field):\n", " salaries = {}\n", " points = {}\n", "\n", " for position in ['QB', 'WR', 'RB', 'TE']:\n", " players_at_position = df[df['position'] == position]\n", " \n", " player_ids_at_position = list(players_at_position['player_id'])\n", " \n", " salaries_at_position = dict(zip(player_ids_at_position, players_at_position['salary']))\n", " points_at_position = dict(zip(player_ids_at_position, players_at_position[points_field]))\n", " \n", " salaries[position] = salaries_at_position\n", " points[position] = points_at_position\n", "\n", " return { 'salaries': salaries, 'points': points }" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is a helper function the uses PuLP to find the optimal lineup given players, their points, and their salary. It stays within the salary cap, limits of the number of players by position so the roster is valid, and produces the lineup with the players that maximize points." ] }, { "cell_type": "code", "execution_count": 54, "metadata": {}, "outputs": [], "source": [ "import pulp\n", "\n", "def run_linear_optimization(salaries, points, num_position_available):\n", " # Create a binary variable grouped by position that tells us if the player is or is not included in the lineup.\n", " players_by_position_var = { \n", " position: pulp.LpVariable.dict(position, players, cat = \"Binary\") for position, players in salaries.items() \n", " }\n", " \n", " # Create a problem using LpMaximize since we want to maximize points.\n", " optimization_problem = pulp.LpProblem(\"best-lineup\", pulp.LpMaximize)\n", " \n", " point_sums = []\n", " salary_sums = []\n", "\n", " for position, players in players_by_position_var.items():\n", " # Create a sum of all players at their position by salary. These get flipped on and off to keep salary below 50000.\n", " salary_sums += pulp.lpSum([salaries[position][i] * players_by_position_var[position][i] for i in players])\n", " \n", " # Create a sum of all players at their position by points scored. These get flipped on and off to maximize points while staying within our constraints.\n", " point_sums += pulp.lpSum([points[position][i] * players_by_position_var[position][i] for i in players])\n", " \n", " # Add a constraint that have fewer than N players at their position where N is the number of players allowed at that position.\n", " optimization_problem += pulp.lpSum([players_by_position_var[position][i] for i in players]) <= num_position_available[position]\n", " \n", " # We want to maximize points so add that sum to the problem.\n", " optimization_problem += pulp.lpSum(point_sums)\n", " \n", " # We need salary to stay below 50000\n", " optimization_problem += pulp.lpSum(salary_sums) <= 50000\n", " \n", " optimization_problem.solve()\n", " \n", " player_ids_to_play = []\n", "\n", " for variable in optimization_problem.variables():\n", " if variable.varValue != 0:\n", " player_ids_to_play.append(variable.name.split('_')[1])\n", " \n", " return player_ids_to_play" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is another helper function that runs the linear optimization based on the specified scoring field. We run three optimizations because a FLEX player can be a WR, RB, or TE so we change the constraints in each optimization to reflect that and then pick the best point total overall." ] }, { "cell_type": "code", "execution_count": 55, "metadata": {}, "outputs": [], "source": [ "def actual_points_sum_for_week(df, scoring_field):\n", " salaries_and_points = salaries_and_points_from_df(df, scoring_field)\n", "\n", " player_ids_4_wrs = run_linear_optimization(\n", " salaries_and_points['salaries'], \n", " salaries_and_points['points'], \n", " {\n", " \"QB\": 1,\n", " \"RB\": 2,\n", " \"WR\": 4,\n", " \"TE\": 1,\n", " }\n", " )\n", "\n", " player_ids_3_rbs = run_linear_optimization(\n", " salaries_and_points['salaries'], \n", " salaries_and_points['points'], \n", " {\n", " \"QB\": 1,\n", " \"RB\": 3,\n", " \"WR\": 3,\n", " \"TE\": 1,\n", " }\n", " )\n", " \n", " player_ids_2_tes = run_linear_optimization(\n", " salaries_and_points['salaries'], \n", " salaries_and_points['points'], \n", " {\n", " \"QB\": 1,\n", " \"RB\": 2,\n", " \"WR\": 3,\n", " \"TE\": 2,\n", " }\n", " )\n", " \n", " sum_4_wrs = df[(df['player_id'].isin(player_ids_4_wrs))]['actual_points'].sum()\n", " sum_3_rbs = df[(df['player_id'].isin(player_ids_3_rbs))]['actual_points'].sum()\n", " sum_2_tes = df[(df['player_id'].isin(player_ids_2_tes))]['actual_points'].sum()\n", "\n", " return max([ sum_4_wrs, sum_3_rbs, sum_2_tes ])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, we're optimizing our lineups. For each week of the season, we're going to get the optimal points by looking at:\n", "\n", "1. The actual points. The data is from 2019 so, we know what actually happened. We'll calculate that to see what the best lineup was overall.\n", "2. The points predicted by Stats Perform, using the entire player universe. \n", "3. The points predicted by Stats Perform, limited to players with upside from the odds. This allows us to optimize from a smaller universe which makes it easier to generate multiple optimal lineups given the smaller sample size." ] }, { "cell_type": "code", "execution_count": 56, "metadata": {}, "outputs": [], "source": [ "actual_points_by_week = []\n", "predicted_points_by_week = []\n", "predicted_points_with_upside_by_week = []\n", "\n", "weeks = [*range(1, 17)]\n", "\n", "for week in weeks:\n", " week_df = scoring_2019_df[(scoring_2019_df['week'] == week)]\n", " week_with_upside_df = scoring_2019_df[(scoring_2019_df['week'] == week) & (scoring_2019_df['has_upside'] == 1)]\n", "\n", " actual_points_by_week.append(actual_points_sum_for_week(week_df, 'actual_points'))\n", " predicted_points_by_week.append(actual_points_sum_for_week(week_df, 'predicted_points'))\n", " predicted_points_with_upside_by_week.append(actual_points_sum_for_week(week_with_upside_df, 'predicted_points'))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Step 7 Complete: Success, the smaller player universe is viable\n", "\n", "The green line below is the actual best lineup. The orange is the full player universe. And the blue is the upside player universe. In many weeks, the upside player universe performs just as well, if not better, than the full player universe. \n", "\n", "The dotted grey line is the score total we’re estimating you need to rank in a daily fantasy sports contest. On average that’s about 157 and you may have noticed that we didn’t build a model for defenses so, we’re subtracting 15 from that average and setting the rank line at 142.\n", "\n", "Still, overall the upside player universe produces an average score of 160.0 which is 4.4 points lower than the full universe. So there's room for improvement here.\n", "\n", "We also call out a few examples that might be worth a closer look to see where you can improve the model, circled in red.\n", "\n", "In week 10, we have an example of the upside universe performing better than the full universe. \n", "\n", "It’s worth looking at the poor performing players the upside model filtered out that week to see if there’s a general pattern worth applying to all weeks.\n", "\n", "And as another example in week 15, the upside universe performed worse than the full universe. \n", "\n", "It’s worth looking at the high performing players the upside model missed that week to see if there’s a way to improve the model for all weeks.\n", "\n", "And with this smaller universe of players, you can generate more lineups to get exposure to more players in your daily fantasy sports contests." ] }, { "cell_type": "code", "execution_count": 64, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 64, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as pyplot\n", "\n", "scatter_size = 1000\n", "\n", "pyplot.figure(figsize=(18, 6))\n", "pyplot.plot(weeks, actual_points_by_week, color = 'green', linewidth = 3, label = \"Actual best\")\n", "pyplot.plot(weeks, predicted_points_by_week, color = 'orange', linewidth = 3, label = \"Optimized using full universe\")\n", "pyplot.plot(weeks, predicted_points_with_upside_by_week, color = 'blue', linewidth = 3, label = \"Optimized using the upside from odds universe\")\n", "pyplot.axhline(y = 142, color='grey', linewidth = 2, linestyle = 'dashed')\n", "pyplot.text(9.5, 124, 'Average points needed to rank\\n(157 with DST, assuming 142 without)')\n", "pyplot.text(.5, 200, 'Avg. score for full universe = 164.4\\nAvg. score for upside universe = 160.0')\n", "pyplot.xlabel('Week')\n", "pyplot.xticks(weeks)\n", "pyplot.ylabel('Points Scored')\n", "pyplot.scatter(\n", " [ 10, 15 ], \n", " [ predicted_points_with_upside_by_week[9], predicted_points_with_upside_by_week[14] ], \n", " scatter_size, \n", " color = 'none', \n", " edgecolor = 'red', \n", " linewidth = 3\n", ")\n", "pyplot.title('Optimized Lineup Points')\n", "\n", "pyplot.legend()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# What's next?\n", "\n", "**Ways you can improve this model**\n", "\n", "* Add scoring for DST by looking at boxscore from IDP.\n", "* Find the optimal percentile for fantasy points to make sit/start decisions. There's a tradeoff here when if you set too high of a percentile, players are aggressively labeled as not having upside from the odds. With too few players, you can't produce a valid lineup.\n", "* Find the optimal qualifiers (eg pass attempts) for inclusion in the model. Too few attempts introduces a lot of noise into the model. For example, it'll look at data points from players that don't get playing time which isn't that useful. Too many attempts and players are aggressively labeled as not having upside from the odds.\n", "* Shuffle the training and verification data. We kept the order static in this demo so that the model would have predictable results but, ordinarily you wan't to randomize the order of your data before training. \n", "* Carve off test data for the learner.\n", "* Tweak model [parameters](https://sagemaker.readthedocs.io/en/stable/api/training/estimators.html#sagemaker.estimator.Estimator.set_hyperparameters) (eg batch size, learner)." ] } ], "metadata": { "kernelspec": { "display_name": "conda_python3", "language": "python", "name": "conda_python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.10" } }, "nbformat": 4, "nbformat_minor": 4 }