{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Forecast with Cold Start Items\n",
    "\n",
    "Consider the situation where a set of related new items are introduced into your catalog.  This could be due to bringing a brand new product to market, offering a product for the first time that was in the market and sold by others or taking a product into a new region.\n",
    "\n",
    "Despite having no historical data for the new product available, one needs to forecast the future values for those items.\n",
    "\n",
    "\n",
    "In such a situation, when no demand history is available for items, this scenario is coined as a \"cold-start problem\".  Amazon Forecast is able to handle this situation easily, when you use the [Auto Predictor](https://github.com/aws-samples/amazon-forecast-samples/blob/main/library/content/AutoPredictor.md).\n",
    "\n",
    "<img src=\"../../common/images/amazon_forecast.png\">\n",
    "\n",
    "\n",
    "# Introduction\n",
    "\n",
    "In this notebook, we walk through the process of generating forecasts for cold start items.  An important note --  generating predictions for cold-start items does require an [Item Metadata](https://github.com/aws-samples/amazon-forecast-samples/blob/main/library/content/ItemMetadata.md) dataset.  The Item Metadata contains categorical data about existing established items as well as your new cold-start items.  Amazon Forecast considers the categorical features of your cold-start items and finds similar items with established historical demand to help estimate demand for the new items.\n",
    "\n",
    "In this example, we are using a feature called \"type\" in the item metadata file where there is a near even distribution of items across types A, B, C, and D.  A new cold-start item with type \"D\" will tend to have it's future demand created by true historical values from other \"D\" typed items.\n",
    "\n",
    "Take care to engineer your item metadata file, relative to the target time series file.  Any items in the item metadata, not in the target time series will have cold-start forecasts generated.  We recommend you take care to ensure no products that are end-of-life are placed in the item metadata file, especially if they have had no activity during the target time series timespan, to ensure forecasts are not generated on dormant products.\n",
    "\n",
    "To correctly identify your cold start product, ensure that the item ID of your cold start product is entered as a row in your item metadata file and that it is not contained in the target time series file. For multiple cold start products, enter each product item ID as a separate row in the item metadata file. If you don’t yet have an item ID for your cold start product, you can use any alphanumeric combination less than 64 characters that isn’t already representative of another product in your dataset.\n",
    "\n",
    "\n",
    "# Table of Contents\n",
    "\n",
    "* Step 0: [Setting up](#setup)\n",
    "* Step 1: [Preparing the Datasets prior to Amazon Forecast](#prepare)\n",
    "* Step 2: [Preparing the Datasets inside Amazon Forecast](#import)\n",
    " * Step 2a: [Creating a Dataset Group](#create)\n",
    " * Step 2b: [Creating a Target Dataset](#target)\n",
    " * Step 2c: [Creating a Item Meta Information Dataset](#related)\n",
    " * Step 2d: [Update the Dataset Group](#update)\n",
    " * Step 2e: [Creating a Target Time Series Dataset Import Job](#targetImport)\n",
    " * Step 2f: [Creating a Item Meta Information Dataset Import Job](#relatedImport)\n",
    " * Step 2g: [Wait on Dataset Import Jobs](#ImportWait)\n",
    "* Step 3: [Create the Predictor](#algo)\n",
    "* Step 4: [Create a Forecast](#forecast)\n",
    "* Step 5: [Querying the Forecasts](#query)\n",
    "* Step 6: [Exporting the Forecasts](#export)\n",
    "* Step 7: [Clearning up your Resources](#cleanup)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 0:  First let us setup Amazon Forecast<a class=\"anchor\" id=\"setup\">\n",
    "\n",
    "This section sets up the permissions and relevant endpoints."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: boto3 in /home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages (1.23.10)\n",
      "Requirement already satisfied: botocore<1.27.0,>=1.26.10 in /home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages (from boto3) (1.26.10)\n",
      "Requirement already satisfied: s3transfer<0.6.0,>=0.5.0 in /home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages (from boto3) (0.5.2)\n",
      "Requirement already satisfied: jmespath<2.0.0,>=0.7.1 in /home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages (from boto3) (0.10.0)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.25.4 in /home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages (from botocore<1.27.0,>=1.26.10->boto3) (1.26.9)\n",
      "Requirement already satisfied: python-dateutil<3.0.0,>=2.1 in /home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages (from botocore<1.27.0,>=1.26.10->boto3) (2.8.2)\n",
      "Requirement already satisfied: six>=1.5 in /home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages (from python-dateutil<3.0.0,>=2.1->botocore<1.27.0,>=1.26.10->boto3) (1.16.0)\n"
     ]
    }
   ],
   "source": [
    "!pip install boto3 --upgrade"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "import os\n",
    "\n",
    "import boto3\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# importing forecast notebook utility from notebooks/common directory\n",
    "sys.path.insert( 0, os.path.abspath(\"../../common\") )\n",
    "import util\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (15.0, 5.0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Configure the S3 bucket name and region name for this lesson.\n",
    "\n",
    "- If you don't have an S3 bucket, create it first on S3.\n",
    "- Although we have set the region to us-west-2 as a default value below, you can choose any of the regions that the service is available in."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9d8c463d738349109f0f21941a921b10",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Text(value='', description='bucketName', placeholder='input your S3 bucket name')"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eed74a5485a74c34b6e7d23bb0207935",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Text(value='us-west-2', description='region', placeholder='input region name.')"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "text_widget_bucket = util.create_text_widget( \"bucketName\", \"input your S3 bucket name\" )\n",
    "text_widget_region = util.create_text_widget( \"region\", \"input region name.\", default_value=\"us-west-2\" )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "bucketName = text_widget_bucket.value\n",
    "assert bucketName, \"bucketName not set.\"\n",
    "\n",
    "region = text_widget_region.value\n",
    "assert region, \"region not set.\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "session = boto3.Session(region_name=region) \n",
    "forecast = session.client(service_name='forecast') \n",
    "forecast_query = session.client(service_name='forecastquery')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sagemaker import get_execution_role\n",
    "role_arn = get_execution_role()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 1: Preparing the Datasets<a class=\"anchor\" id=\"prepare\">\n",
    "    \n",
    "    \n",
    "Here we use a synthetic dataset based on [electricity]() dataset, which consists of the hourly time series for 370 households (with item id 0 to 369). \n",
    "\n",
    "In this hypothetical senario, our goal is to generate forecasts for 4 new customers with item id 370 to 373. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data/test.csv.gz to data/test.csv\n",
      "Done.\n"
     ]
    }
   ],
   "source": [
    "zipLocalFilePath = \"data/test.csv.gz\"\n",
    "localFilePath = \"data/test.csv\"\n",
    "\n",
    "util.extract_gz( zipLocalFilePath, localFilePath )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>timestamp</th>\n",
       "      <th>target_value</th>\n",
       "      <th>item_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2014-01-01 01:00:00</td>\n",
       "      <td>2.53807106598985</td>\n",
       "      <td>client_0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2014-01-01 01:00:00</td>\n",
       "      <td>23.648648648648624</td>\n",
       "      <td>client_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2014-01-01 01:00:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>client_2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2014-01-01 01:00:00</td>\n",
       "      <td>144.81707317073176</td>\n",
       "      <td>client_3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2014-01-01 01:00:00</td>\n",
       "      <td>75.0</td>\n",
       "      <td>client_4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             timestamp        target_value   item_id\n",
       "0  2014-01-01 01:00:00    2.53807106598985  client_0\n",
       "1  2014-01-01 01:00:00  23.648648648648624  client_1\n",
       "2  2014-01-01 01:00:00                 0.0  client_2\n",
       "3  2014-01-01 01:00:00  144.81707317073176  client_3\n",
       "4  2014-01-01 01:00:00                75.0  client_4"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tdf = pd.read_csv(zipLocalFilePath, dtype = object)\n",
    "tdf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "tdf['target_value'] = tdf['target_value'].astype('float')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us plot one time series first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tdf[tdf['item_id'] == 'client_1'][-24*7*2:]\\\n",
    "    .plot(x='timestamp', y='target_value', figsize=(15, 8)); "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we use an item metadata dataset that contains the information for both the non-cold start items (client 0 to 369) and cold start items (client 370 to 373). We call this meta information \"type\" in this specific case. Only one categorical feature is used in this demo, but in practice one normally has multiple categorical features.\n",
    "\n",
    "Note that for cold start items where little to none demand history exists, the algorithm can only \"transfer\" information from the existing items to the new ones through the meta information. Therefore, having informative and high quality meta data is the key for a good cold-start forecast. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>item_id</th>\n",
       "      <th>type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>369</th>\n",
       "      <td>client_369</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>370</th>\n",
       "      <td>client_370</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>371</th>\n",
       "      <td>client_371</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>372</th>\n",
       "      <td>client_372</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>373</th>\n",
       "      <td>client_373</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        item_id type\n",
       "369  client_369    D\n",
       "370  client_370    A\n",
       "371  client_371    B\n",
       "372  client_372    C\n",
       "373  client_373    D"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# this metadata contains the cold start items' metadata as well.\n",
    "localItemMetaDataFilePath = \"data/itemMetaData.csv\"\n",
    "imdf = pd.read_csv(localItemMetaDataFilePath, dtype = object)\n",
    "\n",
    "imdf.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And the following figure shows the histogram of the category \"type.\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "imdf['type'].value_counts().plot(kind='bar');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "s3 = session.client('s3')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "targetTimeseriesDatakey = \"cold-start/test.csv\"\n",
    "\n",
    "s3.upload_file(Filename=localFilePath, Bucket = bucketName, Key = f\"{targetTimeseriesDatakey}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "itemMetaDatakey = \"cold-start/itemMetaData.csv\"\n",
    "\n",
    "s3.upload_file(Filename=localItemMetaDataFilePath, Bucket = bucketName, Key = f\"{itemMetaDatakey}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "project = \"coldstart_demo\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below, we specify key input data and forecast parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "freq = \"H\"\n",
    "forecast_horizon = 48\n",
    "timestamp_format = \"yyyy-MM-dd HH:mm:ss\"\n",
    "delimiter = ','"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2a. Creating a Dataset Group<a class=\"anchor\" id=\"create\">\n",
    "First let's create a dataset group and then update it later to add our datasets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_group = f\"{project}_grp\"\n",
    "dataset_arns = []\n",
    "create_dataset_group_response = forecast.create_dataset_group(Domain=\"CUSTOM\",\n",
    "                                                          DatasetGroupName=dataset_group,\n",
    "                                                          DatasetArns=dataset_arns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_group_arn = create_dataset_group_response['DatasetGroupArn']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2b. Creating a Target Dataset<a class=\"anchor\" id=\"target\">\n",
    "In this example, we will define a target time series. This is a required dataset to use the service."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below we specify the target time series name af_demo_ts_4."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "coldstart_demo_ts\n"
     ]
    }
   ],
   "source": [
    "ts_dataset_name = f\"{project}_ts\"\n",
    "print(ts_dataset_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we specify the schema of our dataset below. Make sure the order of the attributes (columns) matches the raw \n",
    "data in the files. We follow the same three attribute format as the above example."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "ts_schema_val = [{\"AttributeName\": \"timestamp\", \"AttributeType\": \"timestamp\"},\n",
    "                {\"AttributeName\": \"target_value\", \"AttributeType\": \"float\"},\n",
    "                {\"AttributeName\": \"item_id\", \"AttributeType\": \"string\"}]\n",
    "ts_schema = {\"Attributes\": ts_schema_val}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "response = forecast.create_dataset(Domain=\"CUSTOM\",\n",
    "                               DatasetType='TARGET_TIME_SERIES',\n",
    "                               DatasetName=ts_dataset_name,\n",
    "                               DataFrequency=freq,\n",
    "                               Schema=ts_schema\n",
    "                              )\n",
    "\n",
    "ts_dataset_arn = response['DatasetArn']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2c. Creating ItemMetaData Dataset<a class=\"anchor\" id=\"related\">\n",
    "In this example, we will define a Item Metadata Dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "coldstart_demo_meta\n"
     ]
    }
   ],
   "source": [
    "item_metadata_dataset_name = f\"{project}_meta\"\n",
    "print(item_metadata_dataset_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Specify the schema of your dataset here. Make sure the order of columns matches the raw data files. We follow the same three column format as the above example."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "meta_schema_val = [{\"AttributeName\": \"item_id\", \"AttributeType\": \"string\"},\n",
    "              {\"AttributeName\": \"category\", \"AttributeType\": \"string\"}]\n",
    "meta_schema = {\"Attributes\": meta_schema_val}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "response = forecast.create_dataset(Domain=\"CUSTOM\",\n",
    "                               DatasetType='ITEM_METADATA',\n",
    "                               DatasetName=item_metadata_dataset_name,\n",
    "                               Schema=meta_schema\n",
    "                              )\n",
    "\n",
    "meta_dataset_arn = response['DatasetArn']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2d. Updating the dataset group with the datasets we created<a class=\"anchor\" id=\"update\">\n",
    "You can have multiple datasets under the same dataset group. Update it with the datasets we created before."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_arns = []\n",
    "dataset_arns.append(ts_dataset_arn)\n",
    "dataset_arns.append(meta_dataset_arn)\n",
    "update_dataset_group_response = forecast.update_dataset_group(DatasetGroupArn=dataset_group_arn, DatasetArns=dataset_arns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2e. Creating a Target Time Series Dataset Import Job<a class=\"anchor\" id=\"targetImport\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "ts_dataset_import_job_response = forecast.create_dataset_import_job(DatasetImportJobName=dataset_group+\"_1\",\n",
    "                                                             DatasetArn=ts_dataset_arn,\n",
    "                                                             DataSource= {\n",
    "                                                                 \"S3Config\" : {\n",
    "                                                                     \"Path\": f\"s3://{bucketName}/{targetTimeseriesDatakey}\",\n",
    "                                                                     \"RoleArn\": role_arn\n",
    "                                                                 } \n",
    "                                                             },\n",
    "                                                             TimestampFormat=timestamp_format)\n",
    "\n",
    "ts_dataset_import_job_arn=ts_dataset_import_job_response['DatasetImportJobArn']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2f. Creating a Item Meta Data Dataset Import Job<a class=\"anchor\" id=\"relatedImport\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "meta_dataset_import_job_response = forecast.create_dataset_import_job(DatasetImportJobName=dataset_group,\n",
    "                                                             DatasetArn=meta_dataset_arn,\n",
    "                                                             DataSource= {\n",
    "                                                                 \"S3Config\" : {\n",
    "                                                                     \"Path\": f\"s3://{bucketName}/{itemMetaDatakey}\",\n",
    "                                                                     \"RoleArn\": role_arn\n",
    "                                                                 } \n",
    "                                                             })\n",
    "\n",
    "meta_dataset_import_job_arn=meta_dataset_import_job_response['DatasetImportJobArn']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2g. Wait on the two dataset import jobs to complete<a class=\"anchor\" id=\"ImportWait\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CREATE_PENDING \n",
      "CREATE_IN_PROGRESS .........\n",
      "ACTIVE \n",
      "CREATE_IN_PROGRESS ...........................\n",
      "ACTIVE \n"
     ]
    }
   ],
   "source": [
    "status = util.wait(lambda: forecast.describe_dataset_import_job(DatasetImportJobArn=meta_dataset_import_job_arn))\n",
    "assert status\n",
    "status = util.wait(lambda: forecast.describe_dataset_import_job(DatasetImportJobArn=ts_dataset_import_job_arn))\n",
    "assert status"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 3. Create  Predictor with the datasets<a class=\"anchor\" id=\"algo\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read about the [Auto Predictor](https://github.com/aws-samples/amazon-forecast-samples/blob/main/library/content/AutoPredictor.md) to learn more."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Waiting for Predictor to become ACTIVE. Depending on data size and predictor setting，it can take several hours to be ACTIVE.\n",
      "\n",
      "Current Status:\n",
      "CREATE_PENDING ..\n",
      "CREATE_IN_PROGRESS ...............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................\n",
      "ACTIVE \n",
      "\n",
      "\n",
      "The Predictor is now ACTIVE.\n"
     ]
    }
   ],
   "source": [
    "PREDICTOR_NAME = project + \"_predictor\"\n",
    "\n",
    "create_auto_predictor_response = \\\n",
    "    forecast.create_auto_predictor(PredictorName = PREDICTOR_NAME,\n",
    "                                   ForecastHorizon = forecast_horizon,\n",
    "                                   ForecastFrequency = freq,\n",
    "                                   DataConfig = {\n",
    "                                       'DatasetGroupArn': dataset_group_arn\n",
    "                                    })\n",
    "\n",
    "predictor_arn = create_auto_predictor_response['PredictorArn']\n",
    "print(f\"Waiting for Predictor to become ACTIVE. Depending on data size and predictor setting，it can take several hours to be ACTIVE.\\n\\nCurrent Status:\")\n",
    "\n",
    "status = util.wait(lambda: forecast.describe_auto_predictor(PredictorArn=predictor_arn))\n",
    "\n",
    "describe_auto_predictor_response = forecast.describe_auto_predictor(PredictorArn=predictor_arn)\n",
    "print(f\"\\n\\nThe Predictor is now {describe_auto_predictor_response['Status']}.\")\n",
    "                                                   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 4. Creating a Forecast<a class=\"anchor\" id=\"forecast\">\n",
    "\n",
    "Next we re-train with the full dataset, and create the forecast."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Waiting for Forecast to become ACTIVE.\n",
      "\n",
      "Current Status:\n",
      "CREATE_PENDING \n",
      "CREATE_IN_PROGRESS ........................................................................\n",
      "ACTIVE \n"
     ]
    }
   ],
   "source": [
    "FORECAST_NAME = project + \"_forecast\"\n",
    "\n",
    "create_forecast_response = forecast.create_forecast(ForecastName=FORECAST_NAME,\n",
    "                                                    PredictorArn=predictor_arn)\n",
    "\n",
    "forecast_arn = create_forecast_response['ForecastArn']\n",
    "\n",
    "print(f\"Waiting for Forecast to become ACTIVE.\\n\\nCurrent Status:\")\n",
    "\n",
    "status = util.wait(lambda: forecast.describe_forecast(ForecastArn=forecast_arn))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 5. Querying the ColdStart Item Forecast<a class=\"anchor\" id=\"query\">\n",
    "    \n",
    "Now we plot the forecast.  First, we see there is <b>no historical data</b> for client_373 in the tdf dataframe.  Next, despite evidence of no history, a future prediction is made as shown in the plot.  client_373 is type D; therefore, the prediction is based on the distribution of values seen in other type D clients."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>timestamp</th>\n",
       "      <th>target_value</th>\n",
       "      <th>item_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [timestamp, target_value, item_id]\n",
       "Index: []"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tdf[tdf['item_id'] == 'client_373']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "forecast_response = forecast_query.query_forecast(\n",
    "    ForecastArn=forecast_arn,\n",
    "    Filters={\"item_id\": \"client_373\"})\n",
    "\n",
    "fcst = forecast_response['Forecast']['Predictions']\n",
    "time_stamp = list(map(lambda x: pd.to_datetime(x['Timestamp']), fcst['p10']))\n",
    "p10_fcst = list(map(lambda x: x['Value'], fcst['p10']))\n",
    "p50_fcst = list(map(lambda x: x['Value'], fcst['p50']))\n",
    "p90_fcst = list(map(lambda x: x['Value'], fcst['p90']))\n",
    "\n",
    "plt.figure(figsize=(15, 10))\n",
    "plt.plot(time_stamp, p50_fcst)\n",
    "plt.fill_between(time_stamp, p10_fcst, p90_fcst, alpha=0.2)\n",
    "plt.title(\"Cold-Start Forecast client_373\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 6. Exporting your Forecasts<a class=\"anchor\" id=\"export\">\n",
    "    \n",
    "In step 5, the Forecast Query API is used to review single items for quality assurance and testing purposes.  For production scale, you should export the data in bulk to S3.  From there, you can use the data for many reasons including using in your BI platform or querying from a low latency database."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "forecast_export_name = f'{project}_cold_start_forecast_export'\n",
    "forecast_export_name_path = f\"s3://{bucketName}/{forecast_export_name}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CREATE_PENDING .\n",
      "CREATE_IN_PROGRESS ...........\n",
      "ACTIVE \n"
     ]
    }
   ],
   "source": [
    "create_forecast_export_response = forecast.create_forecast_export_job(ForecastExportJobName=forecast_export_name,\n",
    "                                                        ForecastArn=forecast_arn,\n",
    "                                                        Destination={\n",
    "                                                            \"S3Config\" : {\n",
    "                                                                \"Path\": forecast_export_name_path,\n",
    "                                                                \"RoleArn\": role_arn\n",
    "                                                            }\n",
    "                                                        })\n",
    "forecast_export_arn = create_forecast_export_response['ForecastExportJobArn']\n",
    "\n",
    "status = util.wait(lambda: forecast.describe_forecast_export_job(ForecastExportJobArn = forecast_export_arn))\n",
    "assert status"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 7. Cleaning up your Resources<a class=\"anchor\" id=\"cleanup\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once you have completed the above steps, you can start to cleanup the resources we created. All delete jobs, except for `delete_dataset_group` are asynchronous, so we have added the helpful `wait_till_delete` function. \n",
    "Resource Limits documented <a href=\"https://docs.aws.amazon.com/forecast/latest/dg/limits.html\">here</a>. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Delete forecast export\n",
    "util.wait_till_delete(lambda: forecast.delete_forecast_export_job(ForecastExportJobArn = forecast_export_arn))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Delete forecast\n",
    "util.wait_till_delete(lambda: forecast.delete_forecast(ForecastArn = forecast_arn))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Delete predictor\n",
    "util.wait_till_delete(lambda: forecast.delete_predictor(PredictorArn = predictor_arn))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Delete the target time series and related time series dataset import jobs\n",
    "util.wait_till_delete(lambda: forecast.delete_dataset_import_job(DatasetImportJobArn=ts_dataset_import_job_arn))\n",
    "util.wait_till_delete(lambda: forecast.delete_dataset_import_job(DatasetImportJobArn=meta_dataset_import_job_arn))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Delete the target time series and related time series datasets\n",
    "util.wait_till_delete(lambda: forecast.delete_dataset(DatasetArn=ts_dataset_arn))\n",
    "util.wait_till_delete(lambda: forecast.delete_dataset(DatasetArn=meta_dataset_arn))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Delete dataset group\n",
    "forecast.delete_dataset_group(DatasetGroupArn=dataset_group_arn)"
   ]
  }
 ],
 "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": "3.6.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}