{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Notebook 5: Clean Up Resources\n",
    "\n",
    "Specify \"Python 3\" Kernel  and \"Data Science\" Image.\n",
    "\n",
    "### Background\n",
    "\n",
    "In this notebook, we will clean up the resources we provisioned during this workshop:\n",
    "\n",
    "- SageMaker Feature Groups\n",
    "- SageMaker Endpoints\n",
    "- Amazon Kinesis Data Stream\n",
    "- Amazon Kinesis Data Analytics application"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from parameter_store import ParameterStore\n",
    "from utils import *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Session variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "role = sagemaker.get_execution_role()\n",
    "sagemaker_session = sagemaker.Session()\n",
    "region = sagemaker_session.boto_region_name\n",
    "boto_session = boto3.Session()\n",
    "kinesis_client = boto_session.client(service_name='kinesis', region_name=region)\n",
    "kinesis_analytics_client = boto_session.client('kinesisanalytics')\n",
    "\n",
    "ps = ParameterStore(verbose=False)\n",
    "ps.set_namespace('feature-store-workshop')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load variables from previous notebooks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters = ps.read()\n",
    "\n",
    "customers_feature_group_name = parameters['customers_feature_group_name']\n",
    "products_feature_group_name = parameters['products_feature_group_name']\n",
    "orders_feature_group_name = parameters['orders_feature_group_name']\n",
    "click_stream_historical_feature_group_name = parameters['click_stream_historical_feature_group_name']\n",
    "click_stream_feature_group_name = parameters['click_stream_feature_group_name']\n",
    "\n",
    "cf_model_endpoint_name = parameters['cf_model_endpoint_name']\n",
    "ranking_model_endpoint_name = parameters['ranking_model_endpoint_name']\n",
    "\n",
    "kinesis_stream_name = parameters['kinesis_stream_name']\n",
    "kinesis_analytics_application_name = parameters['kinesis_analytics_application_name']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Delete feature groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_group_list = [customers_feature_group_name, products_feature_group_name,\n",
    "                      orders_feature_group_name, click_stream_historical_feature_group_name,\n",
    "                      click_stream_feature_group_name]\n",
    "\n",
    "for feature_group in feature_group_list:\n",
    "    print(f'Deleting feature group: {feature_group}')\n",
    "    delete_feature_group(feature_group)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Delete endpoints and endpoint configurations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_up_endpoint(endpoint_name):\n",
    "    response = sagemaker_session.sagemaker_client.describe_endpoint(EndpointName=endpoint_name)\n",
    "    endpoint_config_name = response['EndpointConfigName']\n",
    "    print(f'Deleting endpoint: {endpoint_name}')\n",
    "    print(f'Deleting endpoint configuration : {endpoint_config_name}')\n",
    "    sagemaker_session.sagemaker_client.delete_endpoint(EndpointName=endpoint_name)\n",
    "    sagemaker_session.sagemaker_client.delete_endpoint_config(EndpointConfigName=endpoint_config_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "endpoint_list = [cf_model_endpoint_name, ranking_model_endpoint_name]\n",
    "\n",
    "for endpoint in endpoint_list:\n",
    "    clean_up_endpoint(endpoint)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Delete Kinesis Data Stream"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "kinesis_client.delete_stream(StreamName=kinesis_stream_name,\n",
    "                             EnforceConsumerDeletion=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Delete Kinesis Data Analytics application"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "response = kinesis_analytics_client.describe_application(ApplicationName=kinesis_analytics_application_name)\n",
    "create_ts = response['ApplicationDetail']['CreateTimestamp']\n",
    "kinesis_analytics_client.delete_application(ApplicationName=kinesis_analytics_application_name, CreateTimestamp=create_ts)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Go back to Workshop Studio and click on \"Next\"."
   ]
  }
 ],
 "metadata": {
  "instance_type": "ml.t3.medium",
  "interpreter": {
   "hash": "fea7262dfaa662dc7ea8f1b256cf975fd886d2f868152164ef15877318a1e322"
  },
  "kernelspec": {
   "display_name": "Python 3 (Data Science)",
   "language": "python",
   "name": "python3__SAGEMAKER_INTERNAL__arn:aws:sagemaker:ca-central-1:310906938811:image/datascience-1.0"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}