{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "51e0cdcd",
   "metadata": {},
   "source": [
    "# Getting Started With Hyperparameter Optimization Using Amazon SageMaker Automatic Model Tuning"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdbfc6f0",
   "metadata": {},
   "source": [
    "## Background - What are we doing?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8f89001",
   "metadata": {},
   "source": [
    "Machine Learning (ML) models are getting better and better. Subsequently they are taking the world by storm. Their performance relies on the right training data and choosing the right model/algorithm. But it does not end here. Typically, algorithms defer some design decisions to the ML practitioner to adopt for their specific data and task. These deferred design decisions manifest themselves as so called *hyperparameters*.\n",
    "\n",
    "What does that name mean? The result of a machine learning training, the model, can be largely seen as a collection of parameters that were learned during training. Hence the parameters that are used to configure the machine learning training are then called hyperparameters, i.e. parameters describing the creation of parameters. At any rate, they are of very practical use, such as the number of epochs to train, the learning rate, the max depth of a decision tree and so forth. And they have a major impact on the ultimate performance of your model.\n",
    "\n",
    "Selecting the right parameters and their value ranges starts with the practitioner's theoretical knowledge of the model to train. This defines the boundaries of the search space, but then it is also necessary to explore this search space empirically to find a good combination of parameters that will result in a high performing model.\n",
    "\n",
    "Luckily, we do not have to manually search that space, because SageMaker Automatic Model Tuning (AMT) is here to help. We point AMT in the right direction by defining the overall search space and AMT then takes over and coordinates the exploration of that space and fully automatically runs training jobs to efficiently evaluate the space. This then results in a good combination of hyper parameter values to train a high performing model.\n",
    "\n",
    "*In an upcoming article we will extend the notion of __just__ finding the best parameters to then __also__ include learning about the search space and to what parameter ranges a model is sensitive, as well as turning a one-shot tuning activity into a multi-step conversation with the ML practitioner, to learn together. Exciting!*\n",
    "\n",
    "In this tutorial we will walk through Amazon SageMaker Automatic Model Tuning (AMT), using a built-in XGBoost algorithm provided by Amazon SageMaker. Additional information can be found in the documentation pages below:\n",
    "* For more information on running a simple hyperparameter tuning job: https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning-ex.html\n",
    "* For documentation on using the HyperParameterTuner API with the SageMaker Python SDK: https://sagemaker.readthedocs.io/en/stable/api/training/tuner.html"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55d3c147",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Overview"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dda7534b",
   "metadata": {},
   "source": [
    "This notebook is split into the following sections:\n",
    "* Setup and Imports\n",
    "* Load and Prepare dataset\n",
    "* Train a SageMaker Built-In XGBoost Algorithm\n",
    "* Train and Tune a SageMaker Built-In XGBoost Algorithm\n",
    "* View the AMT job statistics \n",
    "* Visualize AMT job results and tuned Hyperparameters\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "238d2c62",
   "metadata": {},
   "source": [
    "### Setup and Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e5026ca2",
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "820bc604",
   "metadata": {},
   "source": [
    "We ran this notebook using Amazon SageMaker with the version you see in the output of the next cell below. If your version is lower and you encounter issues, we recommend uncommenting the code below to upgrade your pip and SageMaker versions. Make sure to restart your kernel after upgrading for the changes to take effect."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ae8980f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.172.0'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import sagemaker\n",
    "\n",
    "sagemaker.__version__ "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e6e59ad6",
   "metadata": {},
   "outputs": [],
   "source": [
    "#!pip install --upgrade --quiet pip \"sagemaker==2.121.0\" # upgrade SageMaker to the recommended version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e419bc3f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import io\n",
    "import os\n",
    "import argparse\n",
    "import traceback\n",
    "import boto3\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from pathlib import Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cd7a5efc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# SDK setup\n",
    "role = sagemaker.get_execution_role()\n",
    "region = boto3.Session().region_name\n",
    "sm = boto3.client('sagemaker')\n",
    "boto_sess = boto3.Session(region_name=region)\n",
    "sm_sess = sagemaker.session.Session(boto_session=boto_sess, sagemaker_client=sm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6ad8b879",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'s3://sagemaker-eu-west-1-811243659808/amt-visualize-demo/output'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Data layout and locations. \n",
    "# To store our data we are using a prefix in the Amazon SageMaker default bucket. Feel free to adjust to your preferences.\n",
    "\n",
    "BUCKET = sm_sess.default_bucket()\n",
    "PREFIX = 'amt-visualize-demo'\n",
    "\n",
    "# Eventual output destination for our XGBoost model\n",
    "output_path = f's3://{BUCKET}/{PREFIX}/output'\n",
    "output_path"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f3075fb",
   "metadata": {},
   "source": [
    "## Load and Prepare dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6609672d",
   "metadata": {},
   "outputs": [],
   "source": [
    "!mkdir -p data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "65a9ac0a",
   "metadata": {},
   "source": [
    "The focus of this notebook is on Hyperparameter Optimization. Hence the actual task and data only play a supporting role. But to give some brief context, we are optimizing the hyperparameters of an XGBoost model that should classify handwritten digits. \n",
    "\n",
    "We use the [Optical Recognition of Handwritten Digits Data Set](https://archive.ics.uci.edu/ml/datasets/Optical+Recognition+of+Handwritten+Digits) via scikit-learn:\n",
    "\n",
    "_Dua, D. and Graff, C. (2019). UCI Machine Learning Repository [http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Science._"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9e1ced87",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "\n",
    "digits         = datasets.load_digits()\n",
    "digits_df      = pd.DataFrame(digits.data)\n",
    "digits_df['y'] = digits.target\n",
    "digits_df.insert(0, 'y', digits_df.pop('y')) # XGBoost expects the target to be the first column "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d7511dea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "digit: 4\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "print('digit:', int(digits_df.iloc[100].y))\n",
    "plt.imshow(digits_df.iloc[100, 1:].values.reshape(8, -1));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "145da7f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# randomly sort the data then split out into train 70% and validation 30% sets\n",
    "train_data, valid_data= np.split(\n",
    "    digits_df, [int(0.7 * len(digits_df))]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "de42d336",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data.to_csv('data/train.csv', index=False, header=False)\n",
    "valid_data.to_csv('data/valid.csv', index=False, header=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "448723bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "boto_sess.resource('s3').Bucket(BUCKET).Object(os.path.join(PREFIX, 'data/train/train.csv')).upload_file('data/train.csv')\n",
    "boto_sess.resource('s3').Bucket(BUCKET).Object(os.path.join(PREFIX, 'data/valid/valid.csv')).upload_file('data/valid.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64a55efe",
   "metadata": {},
   "source": [
    "We upload our train and validation datasets into [Amazon S3](https://aws.amazon.com/s3/). Amazon SageMaker will launch training jobs on our behalf and those will read the data directly from S3."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f96665c5",
   "metadata": {},
   "source": [
    "## Train an Amazon SageMaker Built-In XGBoost Algorithm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c0852f1",
   "metadata": {},
   "source": [
    "SageMaker provides many [built-in algorithms](https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html). For our example we use the popular XGBoost algorithm. To select it for our training it is identified by an image URI as described in the docs here:\n",
    "https://docs.aws.amazon.com/sagemaker/latest/dg/xgboost.html\n",
    "\n",
    "In below code you can see how we lookup the respective image and version of the algorithm. We then define the hyperparameters and instantiate an Estimator that describes our training. For a list of hyperparameters and their meaning for this algorithm check out the [documentation](https://docs.aws.amazon.com/sagemaker/latest/dg/xgboost_hyperparameters.html). \n",
    "\n",
    "Please recognize that for our small dataset we also specify the use of a small CPU based instance type for our training, specifically `ml.m5.large`.\n",
    "\n",
    "Eventually we call `fit()` to start the training and to pass in references to where our training and validation data can be found. \n",
    "\n",
    "__Side note__: Strictly speaking it would not be necessary to call the `fit()` method here for an individual training run. We will let SageMaker Automatic Model Tuning (AMT) instantiate and control the individual training runs later. We just need to pass in the Estimator we defined here as a template. We could comment out the call to `fit()`.\n",
    "\n",
    "However, calling `fit()` shows us how one training run works and helps us to get acquainted with the algorithm and the log outputs, which are also available in Amazon CloudWatch. This works even though the actual execution is happening on a different EC2 machine. SageMaker makes the log outputs directly available in this notebook here. Neat, right?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f2db8939",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:sagemaker:Creating training-job with name: sagemaker-xgboost-2023-07-14-12-00-16-020\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Algorithm container image: 141502667606.dkr.ecr.eu-west-1.amazonaws.com/sagemaker-xgboost:1.5-1\n",
      "2023-07-14 12:00:16 Starting - Starting the training job...\n",
      "2023-07-14 12:00:32 Starting - Preparing the instances for training............\n",
      "2023-07-14 12:02:49 Downloading - Downloading input data\n",
      "2023-07-14 12:02:49 Training - Training image download completed. Training in progress..\u001b[34m[2023-07-14 12:02:54.546 ip-10-0-101-47.eu-west-1.compute.internal:7 INFO utils.py:28] RULE_JOB_STOP_SIGNAL_FILENAME: None\u001b[0m\n",
      "\u001b[34m[2023-07-14 12:02:54.631 ip-10-0-101-47.eu-west-1.compute.internal:7 INFO profiler_config_parser.py:111] User has disabled profiler.\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:54:INFO] Imported framework sagemaker_xgboost_container.training\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:54:INFO] Failed to parse hyperparameter eval_metric value accuracy to Json.\u001b[0m\n",
      "\u001b[34mReturning the value itself\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:54:INFO] Failed to parse hyperparameter objective value multi:softmax to Json.\u001b[0m\n",
      "\u001b[34mReturning the value itself\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:54:INFO] No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Running XGBoost Sagemaker in algorithm mode\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Determined 0 GPU(s) available on the instance.\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] files path: /opt/ml/input/data/train\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] files path: /opt/ml/input/data/validation\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Single node training.\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Train matrix has 1257 rows and 64 columns\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Validation matrix has 540 rows\u001b[0m\n",
      "\u001b[34m[2023-07-14 12:02:55.052 ip-10-0-101-47.eu-west-1.compute.internal:7 INFO json_config.py:92] Creating hook from json_config at /opt/ml/input/config/debughookconfig.json.\u001b[0m\n",
      "\u001b[34m[2023-07-14 12:02:55.053 ip-10-0-101-47.eu-west-1.compute.internal:7 INFO hook.py:206] tensorboard_dir has not been set for the hook. SMDebug will not be exporting tensorboard summaries.\u001b[0m\n",
      "\u001b[34m[2023-07-14 12:02:55.054 ip-10-0-101-47.eu-west-1.compute.internal:7 INFO hook.py:259] Saving to /opt/ml/output/tensors\u001b[0m\n",
      "\u001b[34m[2023-07-14 12:02:55.054 ip-10-0-101-47.eu-west-1.compute.internal:7 INFO state_store.py:77] The checkpoint config file /opt/ml/input/config/checkpointconfig.json does not exist.\u001b[0m\n",
      "\u001b[34m[2023-07-14:12:02:55:INFO] Debug hook created from config\u001b[0m\n",
      "\u001b[34m[12:02:55] WARNING: ../src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softmax' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\u001b[0m\n",
      "\u001b[34m[2023-07-14 12:02:55.110 ip-10-0-101-47.eu-west-1.compute.internal:7 INFO hook.py:427] Monitoring the collections: metrics\u001b[0m\n",
      "\u001b[34m[2023-07-14 12:02:55.113 ip-10-0-101-47.eu-west-1.compute.internal:7 INFO hook.py:491] Hook is writing from the hook with pid: 7\u001b[0m\n",
      "\u001b[34m[0]#011train-mlogloss:1.55795#011train-accuracy:0.95943#011validation-mlogloss:1.70729#011validation-accuracy:0.78889\u001b[0m\n",
      "\u001b[34m[1]#011train-mlogloss:1.20572#011train-accuracy:0.97295#011validation-mlogloss:1.42358#011validation-accuracy:0.81111\u001b[0m\n",
      "\u001b[34m[2]#011train-mlogloss:0.97027#011train-accuracy:0.97932#011validation-mlogloss:1.23074#011validation-accuracy:0.81111\u001b[0m\n",
      "\u001b[34m[3]#011train-mlogloss:0.79341#011train-accuracy:0.98568#011validation-mlogloss:1.09127#011validation-accuracy:0.82222\u001b[0m\n",
      "\u001b[34m[4]#011train-mlogloss:0.65584#011train-accuracy:0.99045#011validation-mlogloss:0.97840#011validation-accuracy:0.82963\u001b[0m\n",
      "\u001b[34m[5]#011train-mlogloss:0.54703#011train-accuracy:0.99364#011validation-mlogloss:0.89070#011validation-accuracy:0.83704\u001b[0m\n",
      "\u001b[34m[6]#011train-mlogloss:0.45843#011train-accuracy:0.99602#011validation-mlogloss:0.81601#011validation-accuracy:0.84074\u001b[0m\n",
      "\u001b[34m[7]#011train-mlogloss:0.38576#011train-accuracy:0.99602#011validation-mlogloss:0.75819#011validation-accuracy:0.84444\u001b[0m\n",
      "\u001b[34m[8]#011train-mlogloss:0.32707#011train-accuracy:0.99761#011validation-mlogloss:0.70924#011validation-accuracy:0.84815\u001b[0m\n",
      "\u001b[34m[9]#011train-mlogloss:0.27772#011train-accuracy:0.99841#011validation-mlogloss:0.66924#011validation-accuracy:0.84630\u001b[0m\n",
      "\u001b[34m[10]#011train-mlogloss:0.23722#011train-accuracy:0.99920#011validation-mlogloss:0.63535#011validation-accuracy:0.85370\u001b[0m\n",
      "\u001b[34m[11]#011train-mlogloss:0.20342#011train-accuracy:0.99920#011validation-mlogloss:0.61011#011validation-accuracy:0.85185\u001b[0m\n",
      "\u001b[34m[12]#011train-mlogloss:0.17500#011train-accuracy:0.99920#011validation-mlogloss:0.58224#011validation-accuracy:0.85000\u001b[0m\n",
      "\u001b[34m[13]#011train-mlogloss:0.15114#011train-accuracy:1.00000#011validation-mlogloss:0.56029#011validation-accuracy:0.85000\u001b[0m\n",
      "\u001b[34m[14]#011train-mlogloss:0.13116#011train-accuracy:1.00000#011validation-mlogloss:0.54051#011validation-accuracy:0.85000\u001b[0m\n",
      "\n",
      "2023-07-14 12:03:16 Uploading - Uploading generated training model\n",
      "2023-07-14 12:03:16 Completed - Training job completed\n",
      "Training seconds: 52\n",
      "Billable seconds: 52\n",
      "CPU times: user 464 ms, sys: 93.2 ms, total: 557 ms\n",
      "Wall time: 3min 18s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "from sagemaker import image_uris\n",
    "from sagemaker.session import Session\n",
    "from sagemaker.inputs import TrainingInput\n",
    "\n",
    "\n",
    "# lookup the XGBoost image URI and build an XGBoost container\n",
    "xgboost_container = sagemaker.image_uris.retrieve('xgboost', region, '1.5-1')\n",
    "print('Algorithm container image:', xgboost_container)\n",
    "\n",
    "hyperparameters = {\n",
    "        'num_class': 10,\n",
    "        'max_depth': 5,\n",
    "        'eta':0.2,\n",
    "        'alpha': 0.2, \n",
    "        'objective':'multi:softmax',\n",
    "        'eval_metric':'accuracy',\n",
    "        'num_round':200,\n",
    "        'early_stopping_rounds': 5}\n",
    "\n",
    "# construct a SageMaker estimator that calls the XGBoost container\n",
    "estimator = sagemaker.estimator.Estimator(\n",
    "    image_uri=xgboost_container, \n",
    "    hyperparameters=hyperparameters,\n",
    "    role=role,\n",
    "    instance_count=1, \n",
    "    instance_type='ml.m5.large', \n",
    "    volume_size=5, # 5 GB \n",
    "    output_path=output_path\n",
    ")\n",
    "\n",
    "# define the data type and paths to the training and validation datasets\n",
    "s3_input_train = TrainingInput(s3_data=f's3://{BUCKET}/{PREFIX}/data/train', content_type='csv')\n",
    "s3_input_valid = TrainingInput(s3_data=f's3://{BUCKET}/{PREFIX}/data/valid', content_type='csv')\n",
    "\n",
    "# execute the XGBoost training job\n",
    "estimator.fit({'train': s3_input_train, 'validation': s3_input_valid}) # Optional. See comment above this cell."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9fa0a091",
   "metadata": {},
   "source": [
    "## Train and Tune an Amazon SageMaker Built-In XGBoost Algorithm\n",
    "\n",
    "Analogous to the Estimator above we now setup the HyperparameterTuner. We pass in the estimator object as a template for the trials to be run and a range of hyperparameters to define what search space to explore. Our documentation on [tuning XGBoost](https://docs.aws.amazon.com/sagemaker/latest/dg/xgboost-tuning.html) gives you some tips on choosing parameter ranges and which objective to use.\n",
    "\n",
    "You also see that we specify the number of training jobs to be run in total, as well as how many should run in parallel. Further you can select among many [strategies](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_HyperParameterTuningJobConfig.html#sagemaker-Type-HyperParameterTuningJobConfig-Strategy). In this notebook we use `Bayesian` as strategy. It uses a systematic approach to evaluate the search space. And it incorporates knowledge gained from earlier training runs (trials) when picking better hyperparameters to evaluate in the later trials, during the same __tuning__ job run. \n",
    "\n",
    "*Bayesian Search as a strategy is an excellent default. In an upcoming notebook and article we will additionally contrast and compare different strategies.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "884541f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sagemaker.tuner import IntegerParameter, ContinuousParameter, HyperparameterTuner\n",
    "\n",
    "n_jobs = 50\n",
    "n_parallel_jobs = 3\n",
    "\n",
    "hpt_ranges = {\n",
    "    'alpha': ContinuousParameter(0.01, .5),\n",
    "    'eta': ContinuousParameter(0.1, .5),\n",
    "    'min_child_weight': ContinuousParameter(0., 2.),\n",
    "    'max_depth': IntegerParameter(1, 10)\n",
    "}\n",
    "\n",
    "tuner_parameters = {\n",
    "    'estimator': estimator,\n",
    "    'base_tuning_job_name': 'bayesian',                   \n",
    "    'objective_metric_name': 'validation:accuracy',\n",
    "    'objective_type': 'Maximize',\n",
    "    'hyperparameter_ranges': hpt_ranges,\n",
    "    'strategy': 'Bayesian',\n",
    "    'max_jobs': n_jobs,\n",
    "    'max_parallel_jobs': n_parallel_jobs\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "054c5c0f",
   "metadata": {},
   "source": [
    "Analogous to the actual training with the estimator above, we also kick things off by calling `fit()`. This time on the HyperparameterTuner."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "b4d8fd71",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:sagemaker:Creating hyperparameter tuning job with name: bayesian-230714-1403\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tuning job submitted: bayesian-230714-1403.\n"
     ]
    }
   ],
   "source": [
    "tuner = HyperparameterTuner(**tuner_parameters)\n",
    "tuner.fit({'train': s3_input_train, 'validation': s3_input_valid}, wait=False)\n",
    "tuner_name = tuner.describe()['HyperParameterTuningJobName']\n",
    "print(f'tuning job submitted: {tuner_name}.')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "45cad4b4",
   "metadata": {},
   "source": [
    "Amazon SageMaker AMT, here the HyperparameterTuner, now orchestrates different training jobs (trials) over a period of time. \n",
    "\n",
    "We use `tuner.wait()` to pause this notebook's execution until the AMT job is completed and we can work with the results. Depending on the number of jobs and the level of parallelization this may take some time. For the example below it may take up to 30 minutes for 50 jobs. During this time you can view the status of your jobs in the console by navigating to Amazon SageMaker > Training > Hyperparameter tuning jobs.\n",
    "\n",
    "For more information on AMT job monitoring, see: https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning-monitor.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "030f8569",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "..............................................................................................................................................................................................................................!\n"
     ]
    }
   ],
   "source": [
    "tuner.wait()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dad8e635",
   "metadata": {},
   "source": [
    "## View the AMT job statistics and results \n",
    "\n",
    "Your tuning jobs can be accessed from the Amazon SageMaker console at https://console.aws.amazon.com/sagemaker/. Select Hyperparameter tuning jobs from the Training menu to see the list. More information here: https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning-monitor.html\n",
    "\n",
    "You can also check the results of the jobs programmatically and investigate the hyperparameters used, the final value achieved in the objective function and the total training time per job.\n",
    "\n",
    "Furthermore you can evaluate the results programmatically, which we will look into now."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e88bbfe",
   "metadata": {},
   "source": [
    "#### 1. Via the Amazon SageMaker Python SDK"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "344a09f4",
   "metadata": {},
   "source": [
    "The SDK conveniently provides access to the tuning results as a Pandas dataframe. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "04b2c585",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alpha</th>\n",
       "      <th>eta</th>\n",
       "      <th>max_depth</th>\n",
       "      <th>min_child_weight</th>\n",
       "      <th>TrainingJobName</th>\n",
       "      <th>TrainingJobStatus</th>\n",
       "      <th>FinalObjectiveValue</th>\n",
       "      <th>TrainingStartTime</th>\n",
       "      <th>TrainingEndTime</th>\n",
       "      <th>TrainingElapsedTimeSeconds</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.396073</td>\n",
       "      <td>0.125704</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.739521</td>\n",
       "      <td>bayesian-230714-1403-050-fa565d23</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.85741</td>\n",
       "      <td>2023-07-14 14:22:00+02:00</td>\n",
       "      <td>2023-07-14 14:22:42+02:00</td>\n",
       "      <td>42.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.161270</td>\n",
       "      <td>0.223446</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.293295</td>\n",
       "      <td>bayesian-230714-1403-049-11aebc98</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.88333</td>\n",
       "      <td>2023-07-14 14:21:58+02:00</td>\n",
       "      <td>2023-07-14 14:22:35+02:00</td>\n",
       "      <td>37.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.090324</td>\n",
       "      <td>0.244989</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.314184</td>\n",
       "      <td>bayesian-230714-1403-048-e4d29918</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.88333</td>\n",
       "      <td>2023-07-14 14:21:09+02:00</td>\n",
       "      <td>2023-07-14 14:21:46+02:00</td>\n",
       "      <td>37.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.173002</td>\n",
       "      <td>0.437216</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.865708</td>\n",
       "      <td>bayesian-230714-1403-047-8bd98db7</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.87593</td>\n",
       "      <td>2023-07-14 14:21:01+02:00</td>\n",
       "      <td>2023-07-14 14:21:38+02:00</td>\n",
       "      <td>37.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.115351</td>\n",
       "      <td>0.256223</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.467100</td>\n",
       "      <td>bayesian-230714-1403-046-2b5d469b</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.87778</td>\n",
       "      <td>2023-07-14 14:20:59+02:00</td>\n",
       "      <td>2023-07-14 14:21:41+02:00</td>\n",
       "      <td>42.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.251916</td>\n",
       "      <td>0.211049</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.341080</td>\n",
       "      <td>bayesian-230714-1403-045-d2d7de50</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.87222</td>\n",
       "      <td>2023-07-14 14:20:06+02:00</td>\n",
       "      <td>2023-07-14 14:20:44+02:00</td>\n",
       "      <td>38.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.207472</td>\n",
       "      <td>0.330586</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.653027</td>\n",
       "      <td>bayesian-230714-1403-044-ad96c28c</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.87037</td>\n",
       "      <td>2023-07-14 14:20:05+02:00</td>\n",
       "      <td>2023-07-14 14:20:42+02:00</td>\n",
       "      <td>37.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.123822</td>\n",
       "      <td>0.208177</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.510360</td>\n",
       "      <td>bayesian-230714-1403-043-1cc21939</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.87407</td>\n",
       "      <td>2023-07-14 14:20:03+02:00</td>\n",
       "      <td>2023-07-14 14:20:40+02:00</td>\n",
       "      <td>37.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.050407</td>\n",
       "      <td>0.247309</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.415015</td>\n",
       "      <td>bayesian-230714-1403-042-fc180574</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.88333</td>\n",
       "      <td>2023-07-14 14:19:09+02:00</td>\n",
       "      <td>2023-07-14 14:19:46+02:00</td>\n",
       "      <td>37.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.169022</td>\n",
       "      <td>0.256904</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1.075611</td>\n",
       "      <td>bayesian-230714-1403-041-6d8d8911</td>\n",
       "      <td>Completed</td>\n",
       "      <td>0.87222</td>\n",
       "      <td>2023-07-14 14:19:07+02:00</td>\n",
       "      <td>2023-07-14 14:19:44+02:00</td>\n",
       "      <td>37.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      alpha       eta  max_depth  min_child_weight  \\\n",
       "0  0.396073  0.125704        5.0          1.739521   \n",
       "1  0.161270  0.223446        3.0          0.293295   \n",
       "2  0.090324  0.244989        3.0          0.314184   \n",
       "3  0.173002  0.437216        3.0          1.865708   \n",
       "4  0.115351  0.256223        3.0          0.467100   \n",
       "5  0.251916  0.211049        3.0          1.341080   \n",
       "6  0.207472  0.330586        6.0          1.653027   \n",
       "7  0.123822  0.208177        3.0          1.510360   \n",
       "8  0.050407  0.247309        3.0          0.415015   \n",
       "9  0.169022  0.256904        7.0          1.075611   \n",
       "\n",
       "                     TrainingJobName TrainingJobStatus  FinalObjectiveValue  \\\n",
       "0  bayesian-230714-1403-050-fa565d23         Completed              0.85741   \n",
       "1  bayesian-230714-1403-049-11aebc98         Completed              0.88333   \n",
       "2  bayesian-230714-1403-048-e4d29918         Completed              0.88333   \n",
       "3  bayesian-230714-1403-047-8bd98db7         Completed              0.87593   \n",
       "4  bayesian-230714-1403-046-2b5d469b         Completed              0.87778   \n",
       "5  bayesian-230714-1403-045-d2d7de50         Completed              0.87222   \n",
       "6  bayesian-230714-1403-044-ad96c28c         Completed              0.87037   \n",
       "7  bayesian-230714-1403-043-1cc21939         Completed              0.87407   \n",
       "8  bayesian-230714-1403-042-fc180574         Completed              0.88333   \n",
       "9  bayesian-230714-1403-041-6d8d8911         Completed              0.87222   \n",
       "\n",
       "          TrainingStartTime           TrainingEndTime  \\\n",
       "0 2023-07-14 14:22:00+02:00 2023-07-14 14:22:42+02:00   \n",
       "1 2023-07-14 14:21:58+02:00 2023-07-14 14:22:35+02:00   \n",
       "2 2023-07-14 14:21:09+02:00 2023-07-14 14:21:46+02:00   \n",
       "3 2023-07-14 14:21:01+02:00 2023-07-14 14:21:38+02:00   \n",
       "4 2023-07-14 14:20:59+02:00 2023-07-14 14:21:41+02:00   \n",
       "5 2023-07-14 14:20:06+02:00 2023-07-14 14:20:44+02:00   \n",
       "6 2023-07-14 14:20:05+02:00 2023-07-14 14:20:42+02:00   \n",
       "7 2023-07-14 14:20:03+02:00 2023-07-14 14:20:40+02:00   \n",
       "8 2023-07-14 14:19:09+02:00 2023-07-14 14:19:46+02:00   \n",
       "9 2023-07-14 14:19:07+02:00 2023-07-14 14:19:44+02:00   \n",
       "\n",
       "   TrainingElapsedTimeSeconds  \n",
       "0                        42.0  \n",
       "1                        37.0  \n",
       "2                        37.0  \n",
       "3                        37.0  \n",
       "4                        42.0  \n",
       "5                        38.0  \n",
       "6                        37.0  \n",
       "7                        37.0  \n",
       "8                        37.0  \n",
       "9                        37.0  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sagemaker.HyperparameterTuningJobAnalytics(tuner_name).dataframe()[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f87a965",
   "metadata": {},
   "source": [
    "#### 2. Via the AWS SDK for Python (Boto3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8dfd2864",
   "metadata": {},
   "source": [
    "With the boto3 client we can review the results of a HPO job using [`describe_hyper_parameter_tuning_job()`](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeHyperParameterTuningJob.html) function.\n",
    "\n",
    "It returns a Python dictionary with comprehensive information. Below you see the output scoped to the `BestTrainingJob`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "91af7fb2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'TrainingJobName': 'bayesian-230714-1403-018-bdee7ec3',\n",
       " 'TrainingJobArn': 'arn:aws:sagemaker:eu-west-1:811243659808:training-job/bayesian-230714-1403-018-bdee7ec3',\n",
       " 'CreationTime': datetime.datetime(2023, 7, 14, 14, 11, 24, tzinfo=tzlocal()),\n",
       " 'TrainingStartTime': datetime.datetime(2023, 7, 14, 14, 11, 27, tzinfo=tzlocal()),\n",
       " 'TrainingEndTime': datetime.datetime(2023, 7, 14, 14, 12, 4, tzinfo=tzlocal()),\n",
       " 'TrainingJobStatus': 'Completed',\n",
       " 'TunedHyperParameters': {'alpha': '0.1282429417878869',\n",
       "  'eta': '0.22119469325684027',\n",
       "  'max_depth': '3',\n",
       "  'min_child_weight': '1.7640512969826174'},\n",
       " 'FinalHyperParameterTuningJobObjectiveMetric': {'MetricName': 'validation:accuracy',\n",
       "  'Value': 0.8851799964904785},\n",
       " 'ObjectiveStatus': 'Succeeded'}"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#sm.describe_hyper_parameter_tuning_job(HyperParameterTuningJobName=tuner_name)   # to review all data\n",
    "sm.describe_hyper_parameter_tuning_job(HyperParameterTuningJobName=tuner_name)['BestTrainingJob']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "099881e5",
   "metadata": {},
   "source": [
    "We can also utilize Boto3's [`list_training_jobs_for_hyper_parameter_tuning_job()`](https://docs.aws.amazon.com/cli/latest/reference/sagemaker/list-training-jobs-for-hyper-parameter-tuning-job.html) function to review the results. This can be sorted by the value of the objective function or by the metric definitions. More functions available for Amazon SageMaker with Boto3 are described on this page: https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/sagemaker.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "0d1fac3e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "bayesian-230714-1403-018-bdee7ec3 Metrics: {'train:mlogloss': 0.025679999962449074, 'train:accuracy': 1.0, 'validation:accuracy': 0.8851799964904785, 'validation:mlogloss': 0.36316999793052673, 'ObjectiveMetric': 0.8851799964904785}\n",
      "bayesian-230714-1403-026-5f143626 Metrics: {'train:mlogloss': 0.010710000060498714, 'train:accuracy': 1.0, 'validation:accuracy': 0.8851799964904785, 'validation:mlogloss': 0.34033000469207764, 'ObjectiveMetric': 0.8851799964904785}\n",
      "bayesian-230714-1403-027-96a57f9d Metrics: {'train:mlogloss': 0.019269999116659164, 'train:accuracy': 1.0, 'validation:accuracy': 0.8851799964904785, 'validation:mlogloss': 0.3558500111103058, 'ObjectiveMetric': 0.8851799964904785}\n",
      "bayesian-230714-1403-036-d31a2a62 Metrics: {'train:mlogloss': 0.018449999392032623, 'train:accuracy': 1.0, 'validation:accuracy': 0.8833299875259399, 'validation:mlogloss': 0.3513999879360199, 'ObjectiveMetric': 0.8833299875259399}\n",
      "bayesian-230714-1403-048-e4d29918 Metrics: {'train:mlogloss': 0.020419999957084656, 'train:accuracy': 1.0, 'validation:accuracy': 0.8833299875259399, 'validation:mlogloss': 0.35569998621940613, 'ObjectiveMetric': 0.8833299875259399}\n",
      "bayesian-230714-1403-042-fc180574 Metrics: {'train:mlogloss': 0.02061999961733818, 'train:accuracy': 1.0, 'validation:accuracy': 0.8833299875259399, 'validation:mlogloss': 0.3608799874782562, 'ObjectiveMetric': 0.8833299875259399}\n",
      "bayesian-230714-1403-031-034f7bf5 Metrics: {'train:mlogloss': 0.012620000168681145, 'train:accuracy': 1.0, 'validation:accuracy': 0.8833299875259399, 'validation:mlogloss': 0.3465900123119354, 'ObjectiveMetric': 0.8833299875259399}\n",
      "bayesian-230714-1403-007-2b9573f3 Metrics: {'train:mlogloss': 0.01623000018298626, 'train:accuracy': 1.0, 'validation:accuracy': 0.8833299875259399, 'validation:mlogloss': 0.37307998538017273, 'ObjectiveMetric': 0.8833299875259399}\n",
      "bayesian-230714-1403-004-14c67170 Metrics: {'train:mlogloss': 0.033569999039173126, 'train:accuracy': 1.0, 'validation:accuracy': 0.8833299875259399, 'validation:mlogloss': 0.3709999918937683, 'ObjectiveMetric': 0.8833299875259399}\n",
      "bayesian-230714-1403-049-11aebc98 Metrics: {'train:mlogloss': 0.018319999799132347, 'train:accuracy': 1.0, 'validation:accuracy': 0.8833299875259399, 'validation:mlogloss': 0.3549099862575531, 'ObjectiveMetric': 0.8833299875259399}\n"
     ]
    }
   ],
   "source": [
    "hpo_jobs = sm.list_training_jobs_for_hyper_parameter_tuning_job(\n",
    "    HyperParameterTuningJobName=tuner_name,\n",
    "    MaxResults=100,\n",
    "    SortBy='FinalObjectiveMetricValue',\n",
    "    SortOrder='Descending')\n",
    "\n",
    "for job in hpo_jobs['TrainingJobSummaries'][:10]:\n",
    "    job_descr = sm.describe_training_job(TrainingJobName=job['TrainingJobName'])\n",
    "    metrics = {m['MetricName']:  m['Value'] for m in job_descr['FinalMetricDataList']}\n",
    "    print(f'{job[\"TrainingJobName\"]} Metrics: {metrics}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3b96e50",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Visualize AMT job results and tuned Hyperparameters\n",
    "\n",
    "Finally, we want to interactively visualize the impact of our hyperparameters and their values on the optimization objective.\n",
    "\n",
    "To do this, we utilize the Altair statistical visualization library for Python, and have written two custom analysis scripts `job_analytics.py` and `reporting_util.py` that we make available with this notebook.\n",
    "\n",
    "In an upcoming notebook and article we will dive deeper into the analysis of this data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "2cdb24e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install -Uq pip altair"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df221864",
   "metadata": {},
   "source": [
    "Please ensure that the role used by SageMaker allows the `cloudwatch:ListMetrics` action on [IAM](https://console.aws.amazon.com/iam)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "d4396d94",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tuning job bayesian-230714-1403      status: Completed\n",
      "\n",
      "Number of training jobs with valid objective: 50\n",
      "Lowest: 0.8481500148773193 Highest 0.8851799964904785\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alpha</th>\n",
       "      <th>eta</th>\n",
       "      <th>max_depth</th>\n",
       "      <th>min_child_weight</th>\n",
       "      <th>TrainingJobName</th>\n",
       "      <th>TrainingJobStatus</th>\n",
       "      <th>TrainingStartTime</th>\n",
       "      <th>TrainingEndTime</th>\n",
       "      <th>TrainingElapsedTimeSeconds</th>\n",
       "      <th>TuningJobName</th>\n",
       "      <th>validation_accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>0.128243</td>\n",
       "      <td>0.221195</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.764051</td>\n",
       "      <td>bayesian-230714-1403-018-bdee7ec3</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:11:27+02:00</td>\n",
       "      <td>2023-07-14 14:12:04+02:00</td>\n",
       "      <td>37.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88518</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0.137948</td>\n",
       "      <td>0.245029</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.744948</td>\n",
       "      <td>bayesian-230714-1403-027-96a57f9d</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:14:09+02:00</td>\n",
       "      <td>2023-07-14 14:14:51+02:00</td>\n",
       "      <td>42.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88518</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>0.125875</td>\n",
       "      <td>0.222413</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.242353</td>\n",
       "      <td>bayesian-230714-1403-026-5f143626</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:14:00+02:00</td>\n",
       "      <td>2023-07-14 14:14:37+02:00</td>\n",
       "      <td>37.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88518</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.160020</td>\n",
       "      <td>0.218877</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>bayesian-230714-1403-032-05fea18d</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:15:48+02:00</td>\n",
       "      <td>2023-07-14 14:16:25+02:00</td>\n",
       "      <td>37.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>0.172845</td>\n",
       "      <td>0.437167</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.982770</td>\n",
       "      <td>bayesian-230714-1403-007-2b9573f3</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:07:57+02:00</td>\n",
       "      <td>2023-07-14 14:08:39+02:00</td>\n",
       "      <td>42.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>0.135411</td>\n",
       "      <td>0.217013</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.902921</td>\n",
       "      <td>bayesian-230714-1403-004-14c67170</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:07:04+02:00</td>\n",
       "      <td>2023-07-14 14:07:41+02:00</td>\n",
       "      <td>37.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.050407</td>\n",
       "      <td>0.247309</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.415015</td>\n",
       "      <td>bayesian-230714-1403-042-fc180574</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:19:09+02:00</td>\n",
       "      <td>2023-07-14 14:19:46+02:00</td>\n",
       "      <td>37.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.091448</td>\n",
       "      <td>0.251732</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.079450</td>\n",
       "      <td>bayesian-230714-1403-031-034f7bf5</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:15:16+02:00</td>\n",
       "      <td>2023-07-14 14:15:53+02:00</td>\n",
       "      <td>37.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.090324</td>\n",
       "      <td>0.244989</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.314184</td>\n",
       "      <td>bayesian-230714-1403-048-e4d29918</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:21:09+02:00</td>\n",
       "      <td>2023-07-14 14:21:46+02:00</td>\n",
       "      <td>37.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.192958</td>\n",
       "      <td>0.219448</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.085700</td>\n",
       "      <td>bayesian-230714-1403-036-d31a2a62</td>\n",
       "      <td>Completed</td>\n",
       "      <td>2023-07-14 14:16:53+02:00</td>\n",
       "      <td>2023-07-14 14:17:31+02:00</td>\n",
       "      <td>38.0</td>\n",
       "      <td>bayesian-230714-1403</td>\n",
       "      <td>0.88333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       alpha       eta  max_depth  min_child_weight                    TrainingJobName TrainingJobStatus         TrainingStartTime           TrainingEndTime  TrainingElapsedTimeSeconds         TuningJobName  validation_accuracy\n",
       "32  0.128243  0.221195        3.0          1.764051  bayesian-230714-1403-018-bdee7ec3         Completed 2023-07-14 14:11:27+02:00 2023-07-14 14:12:04+02:00                        37.0  bayesian-230714-1403              0.88518\n",
       "23  0.137948  0.245029        3.0          1.744948  bayesian-230714-1403-027-96a57f9d         Completed 2023-07-14 14:14:09+02:00 2023-07-14 14:14:51+02:00                        42.0  bayesian-230714-1403              0.88518\n",
       "24  0.125875  0.222413        3.0          0.242353  bayesian-230714-1403-026-5f143626         Completed 2023-07-14 14:14:00+02:00 2023-07-14 14:14:37+02:00                        37.0  bayesian-230714-1403              0.88518\n",
       "18  0.160020  0.218877        3.0          0.000000  bayesian-230714-1403-032-05fea18d         Completed 2023-07-14 14:15:48+02:00 2023-07-14 14:16:25+02:00                        37.0  bayesian-230714-1403              0.88333\n",
       "43  0.172845  0.437167        3.0          1.982770  bayesian-230714-1403-007-2b9573f3         Completed 2023-07-14 14:07:57+02:00 2023-07-14 14:08:39+02:00                        42.0  bayesian-230714-1403              0.88333\n",
       "46  0.135411  0.217013        3.0          1.902921  bayesian-230714-1403-004-14c67170         Completed 2023-07-14 14:07:04+02:00 2023-07-14 14:07:41+02:00                        37.0  bayesian-230714-1403              0.88333\n",
       "8   0.050407  0.247309        3.0          0.415015  bayesian-230714-1403-042-fc180574         Completed 2023-07-14 14:19:09+02:00 2023-07-14 14:19:46+02:00                        37.0  bayesian-230714-1403              0.88333\n",
       "19  0.091448  0.251732        3.0          0.079450  bayesian-230714-1403-031-034f7bf5         Completed 2023-07-14 14:15:16+02:00 2023-07-14 14:15:53+02:00                        37.0  bayesian-230714-1403              0.88333\n",
       "2   0.090324  0.244989        3.0          0.314184  bayesian-230714-1403-048-e4d29918         Completed 2023-07-14 14:21:09+02:00 2023-07-14 14:21:46+02:00                        37.0  bayesian-230714-1403              0.88333\n",
       "14  0.192958  0.219448        3.0          0.085700  bayesian-230714-1403-036-d31a2a62         Completed 2023-07-14 14:16:53+02:00 2023-07-14 14:17:31+02:00                        38.0  bayesian-230714-1403              0.88333"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARN shape dropped as it is incompatible with \"bar\".\n",
      "WARN x-scale's \"base\" is dropped as it does not work with linear scale.\n",
      "WARN x-scale's \"base\" is dropped as it does not work with linear scale.\n",
      "WARN x-scale's \"base\" is dropped as it does not work with linear scale.\n",
      "WARN x-scale's \"base\" is dropped as it does not work with linear scale.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "alt.VConcatChart(...)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from amtviz import visualize_tuning_job\n",
    "visualize_tuning_job(tuner, trials_only=True) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2492c2e-8828-4e72-a765-b93295130958",
   "metadata": {},
   "source": [
    "## Cleanup \n",
    "\n",
    "To avoid incurring unwanted costs when you’re done experimenting with HPO, you must remove all files in your S3 bucket with the prefix `amt-visualize-demo` and also [shut down Studio resources](https://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-run-and-manage-shut-down.html).\n",
    "\n",
    "Uncomment and run the following code to remove all S3 files created by this notebook. Note that`{BUCKET}` is the variable with our bucket name that we defined earlier, you may also write your bucket name as plaintext here.\n",
    "\n",
    "If you wish to keep the datasets or the model artifacts, you may modify the prefix in the code to `amt-visualize-demo/data` to only delete the data or `amt-visualize-demo/output` to only delete the model artifacts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "fb06325f-1eaf-4faf-aacc-c2eab84203b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "#!aws s3 rm s3://{BUCKET}/amt-visualize-demo --recursive"
   ]
  }
 ],
 "metadata": {
  "instance_type": "ml.t3.medium",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}