predictive-maintenance-xgboost.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1> Predictive Maintainance MVP"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Only run this notebook after you have ran Dataset_preprocess.ipynb to create a cleaned dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook we train an ML model to predict whether a part is faulty or not based on sensor readings from the part. For training, we use a dataset which is adapted from a real predictive maintenance use case. The columns and rows have been anonymized, but the overall statistics of the dataset has been preserved. This will allow you to see some of the typical challenges that arise when dealing with ML for predictive maintenance.\n",
    "\n",
    "We will go over the key aspects of a typical data science pipeline:\n",
    "\n",
    "1. data ingest\n",
    "2. data exploration <br/>\n",
    "3. model training <br/>\n",
    "4. model deployment <br/>\n",
    "\n",
    "Here we train an XGBoost model, using Amazon SageMaker's built in algorithms to predict whether a part is faulty or not in assembly line for predictive maintainance. Once the faulty part is found, it will be removed from the line and taken for repair.\n",
    "\n",
    "Instead of deploying a live endpoint here, we will deploy the model locally on the Greengrass core. But to test the model, we will test the model predictions on a test dataset and plot relevant metrics."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> Import libraries and get data into dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "arn:aws:iam::389535300735:role/service-role/AmazonSageMaker-ExecutionRole-20190823T110499\n"
     ]
    }
   ],
   "source": [
    "import sagemaker\n",
    "import numpy as np\n",
    "import os\n",
    "import json\n",
    "import boto3\n",
    "\n",
    "sagemaker_session = sagemaker.Session()\n",
    "bucket = 'your-S3-bucket' #NOTE: Replace with the bucket name created for you by CloudFormation or any other bucket\n",
    "# if you are just running the notebook, not the entire lab\n",
    "prefix = 'xgbdata'\n",
    "LOCAL_DIR = 'training_data'\n",
    "role = sagemaker.get_execution_role()\n",
    "print(role)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "s3://iot-ml-predictive-maintenance/xgb\n"
     ]
    }
   ],
   "source": [
    "# Path to upload the trained model\n",
    "xgb_upload_location = os.path.join('s3://{}/{}'.format(bucket, 'xgb'))\n",
    "print(xgb_upload_location)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:0.90-1-cpu-py3\n"
     ]
    }
   ],
   "source": [
    "# Grab the latest XGBoost container from ECR\n",
    "region = sagemaker_session.boto_region_name\n",
    "from sagemaker.amazon.amazon_estimator import get_image_uri\n",
    "container = get_image_uri(region, 'xgboost', '0.90-1')\n",
    "print(container)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The training data doesn't have any column name headers so lets include those first\n",
    "with open('cols.txt', 'r') as f:\n",
    "    cols = json.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read the data directly locally\n",
    "import pandas as pd\n",
    "\n",
    "DATANAME = 'train.csv'\n",
    "\n",
    "datorig = pd.read_csv('{}/{}'.format(LOCAL_DIR, DATANAME), names = cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>Response</th>\n",
       "      <th>Sensor_1</th>\n",
       "      <th>Sensor_2</th>\n",
       "      <th>Sensor_3</th>\n",
       "      <th>Sensor_4</th>\n",
       "      <th>Sensor_5</th>\n",
       "      <th>Sensor_6</th>\n",
       "      <th>Sensor_7</th>\n",
       "      <th>Sensor_8</th>\n",
       "      <th>Sensor_9</th>\n",
       "      <th>...</th>\n",
       "      <th>Sensor_159</th>\n",
       "      <th>Sensor_160</th>\n",
       "      <th>Sensor_161</th>\n",
       "      <th>Sensor_162</th>\n",
       "      <th>Sensor_163</th>\n",
       "      <th>Sensor_164</th>\n",
       "      <th>Sensor_165</th>\n",
       "      <th>Sensor_166</th>\n",
       "      <th>Sensor_167</th>\n",
       "      <th>Sensor_168</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>59696</td>\n",
       "      <td>0</td>\n",
       "      <td>166</td>\n",
       "      <td>150</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>439836</td>\n",
       "      <td>612972</td>\n",
       "      <td>354848</td>\n",
       "      <td>190784</td>\n",
       "      <td>524262</td>\n",
       "      <td>915936</td>\n",
       "      <td>583296</td>\n",
       "      <td>116054</td>\n",
       "      <td>13562</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>466</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>19536</td>\n",
       "      <td>7650</td>\n",
       "      <td>2432</td>\n",
       "      <td>856</td>\n",
       "      <td>2370</td>\n",
       "      <td>832</td>\n",
       "      <td>758</td>\n",
       "      <td>11760</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>56552</td>\n",
       "      <td>0</td>\n",
       "      <td>2308</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2552</td>\n",
       "      <td>...</td>\n",
       "      <td>908340</td>\n",
       "      <td>2242318</td>\n",
       "      <td>870188</td>\n",
       "      <td>418716</td>\n",
       "      <td>774598</td>\n",
       "      <td>277358</td>\n",
       "      <td>52496</td>\n",
       "      <td>7796</td>\n",
       "      <td>3232</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>59726</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>409488</td>\n",
       "      <td>442272</td>\n",
       "      <td>340830</td>\n",
       "      <td>201640</td>\n",
       "      <td>583574</td>\n",
       "      <td>708870</td>\n",
       "      <td>582644</td>\n",
       "      <td>227474</td>\n",
       "      <td>70738</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>41418</td>\n",
       "      <td>0</td>\n",
       "      <td>218</td>\n",
       "      <td>192</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>491524</td>\n",
       "      <td>719128</td>\n",
       "      <td>216170</td>\n",
       "      <td>115270</td>\n",
       "      <td>289302</td>\n",
       "      <td>363574</td>\n",
       "      <td>346694</td>\n",
       "      <td>203702</td>\n",
       "      <td>213272</td>\n",
       "      <td>184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>1438</td>\n",
       "      <td>0</td>\n",
       "      <td>66</td>\n",
       "      <td>62</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>19290</td>\n",
       "      <td>16326</td>\n",
       "      <td>4938</td>\n",
       "      <td>2878</td>\n",
       "      <td>38666</td>\n",
       "      <td>20286</td>\n",
       "      <td>112</td>\n",
       "      <td>48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>49234</td>\n",
       "      <td>0</td>\n",
       "      <td>202</td>\n",
       "      <td>172</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>383518</td>\n",
       "      <td>755716</td>\n",
       "      <td>428538</td>\n",
       "      <td>249462</td>\n",
       "      <td>592652</td>\n",
       "      <td>555244</td>\n",
       "      <td>301762</td>\n",
       "      <td>95258</td>\n",
       "      <td>22370</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0</td>\n",
       "      <td>800</td>\n",
       "      <td>0</td>\n",
       "      <td>2130706432</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>43220</td>\n",
       "      <td>58078</td>\n",
       "      <td>6908</td>\n",
       "      <td>1992</td>\n",
       "      <td>2616</td>\n",
       "      <td>1478</td>\n",
       "      <td>6410</td>\n",
       "      <td>12080</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0</td>\n",
       "      <td>648922</td>\n",
       "      <td>0</td>\n",
       "      <td>2908</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>26168</td>\n",
       "      <td>...</td>\n",
       "      <td>3190138</td>\n",
       "      <td>3045518</td>\n",
       "      <td>2457236</td>\n",
       "      <td>1351014</td>\n",
       "      <td>4394380</td>\n",
       "      <td>2047636</td>\n",
       "      <td>5956570</td>\n",
       "      <td>11832902</td>\n",
       "      <td>135994</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>24</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>5426</td>\n",
       "      <td>1248</td>\n",
       "      <td>24</td>\n",
       "      <td>6</td>\n",
       "      <td>20</td>\n",
       "      <td>18</td>\n",
       "      <td>34</td>\n",
       "      <td>28</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 169 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Response  Sensor_1  Sensor_2    Sensor_3  Sensor_4  Sensor_5  Sensor_6  \\\n",
       "0         0     59696         0         166       150         0         0   \n",
       "1         0       466         0          22        14         0         0   \n",
       "2         0     56552         0        2308         0         0         0   \n",
       "3         0     59726         0          78        40         0         0   \n",
       "4         0     41418         0         218       192         0         0   \n",
       "5         0      1438         0          66        62         0         0   \n",
       "6         0     49234         0         202       172         0         0   \n",
       "7         0       800         0  2130706432        50         0         0   \n",
       "8         0    648922         0        2908         0         0         0   \n",
       "9         0         8         0          24        24         0         0   \n",
       "\n",
       "   Sensor_7  Sensor_8  Sensor_9  ...  Sensor_159  Sensor_160  Sensor_161  \\\n",
       "0         0         0         0  ...      439836      612972      354848   \n",
       "1         0         0         0  ...       19536        7650        2432   \n",
       "2         0         0      2552  ...      908340     2242318      870188   \n",
       "3         0         0         0  ...      409488      442272      340830   \n",
       "4         0         0         0  ...      491524      719128      216170   \n",
       "5         0         0         0  ...       19290       16326        4938   \n",
       "6         0         0         0  ...      383518      755716      428538   \n",
       "7         0         0         0  ...       43220       58078        6908   \n",
       "8         0         0     26168  ...     3190138     3045518     2457236   \n",
       "9         0         0         0  ...        5426        1248          24   \n",
       "\n",
       "   Sensor_162  Sensor_163  Sensor_164  Sensor_165  Sensor_166  Sensor_167  \\\n",
       "0      190784      524262      915936      583296      116054       13562   \n",
       "1         856        2370         832         758       11760           0   \n",
       "2      418716      774598      277358       52496        7796        3232   \n",
       "3      201640      583574      708870      582644      227474       70738   \n",
       "4      115270      289302      363574      346694      203702      213272   \n",
       "5        2878       38666       20286         112          48           0   \n",
       "6      249462      592652      555244      301762       95258       22370   \n",
       "7        1992        2616        1478        6410       12080          20   \n",
       "8     1351014     4394380     2047636     5956570    11832902      135994   \n",
       "9           6          20          18          34          28           0   \n",
       "\n",
       "   Sensor_168  \n",
       "0           0  \n",
       "1           0  \n",
       "2           0  \n",
       "3          14  \n",
       "4         184  \n",
       "5           0  \n",
       "6           6  \n",
       "7           0  \n",
       "8           0  \n",
       "9           0  \n",
       "\n",
       "[10 rows x 169 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "datorig.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data consists of 2 parts:\n",
    "\n",
    "1. Response: this is the faulty (1) or not faulty (0) signal. <br/>\n",
    "2. Sensor readings: the remaining 168 columns correspond to Sensor readings.\n",
    "\n",
    "Note the large number of 0s in the dataset. This is because not all sensors fire at the same time, and each reading only comes from a subset of the sensors. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> Data Exploration"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before building an ML model, let's go ahead and explore the dataset and look for some correlations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Class Imbalance\n",
    "One of the first things you notice in predictive maintenance is CLASS IMBALANCE.\n",
    "\n",
    "CLASS IMBALANCE occurs when one class (in this case faulty labels) is much less prevalent than the not faulty label.\n",
    "\n",
    "This is a problem for ML models as this means that the model may not see enough examples of the faulty class to train accurately. Seen another way, this means that a model that always predicts a label to be not faulty is 98.4% accurate!\n",
    "\n",
    "Our ML model will need to beat this accuracy to be useful from a business perspective. There are a number of ways for dealing with imbalanced classification problems -- upsampling, downsampling, SMOTE to name a few. For this POC we will not use these approaches and train a minimal viable model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percent of Faulty Examples = 1.6\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZUAAAEKCAYAAADaa8itAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAESdJREFUeJzt3XusZWV5x/HvjxlRqkVQphRnkCFlWjveECaItRojFQZrhRhUqMqIRJqKVlNDi40RRUmx2iJ4S6iMXGpB1FaoxZIJaG0rCIMg1yCnqGUoyMhVi6KDT//Y7+AWzxk2M+8+ew7n+0l2zlrPetfazyJDfll73VJVSJLUwzaTbkCS9NhhqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1I2hIknqxlCRJHWzcNINzLaddtqpli5dOuk2JGnOuOKKK35QVYtGGTvvQmXp0qWsXbt20m1I0pyR5HujjvXnL0lSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndGCqSpG4MFUlSN/PujvottfcxZ066BW2FrvjQ4ZNuQdoqeKQiSerGUJEkdWOoSJK6MVQkSd0YKpKkbgwVSVI3hookqRtDRZLUjaEiSerGUJEkdWOoSJK6MVQkSd0YKpKkbgwVSVI3hookqRtDRZLUjaEiSerGUJEkdWOoSJK6MVQkSd0YKpKkbgwVSVI3hookqRtDRZLUjaEiSerGUJEkdWOoSJK6GXuoJFmQ5MokX2rzuyf5RpKpJJ9Nsm2rP77NT7XlS4e28a5WvzHJAUP1la02leTYce+LJGnTZuNI5e3ADUPzHwROqqo9gLuBI1v9SODuVj+pjSPJcuBQ4JnASuATLagWAB8HDgSWA4e1sZKkCRlrqCRZAvwh8Kk2H+ClwOfbkDOAg9v0QW2etny/Nv4g4JyqeqCqvgNMAfu0z1RV3VxVPwXOaWMlSRMy7iOVjwB/Afy8zT8VuKeqNrT5dcDiNr0YuAWgLb+3jX+o/rB1Zqr/iiRHJVmbZO369eu3dJ8kSTMYW6gkeQVwR1VdMa7vGFVVnVpVK6pqxaJFiybdjiQ9Zi0c47ZfCLwyycuBJwDbAycDOyRZ2I5GlgC3tvG3ArsC65IsBJ4M3DlU32h4nZnqkqQJGNuRSlW9q6qWVNVSBifaL66q1wFfAQ5pw1YB57Xp89s8bfnFVVWtfmi7Omx3YBlwGXA5sKxdTbZt+47zx7U/kqRHNs4jlZn8JXBOkg8AVwKntfppwFlJpoC7GIQEVXVdknOB64ENwNFV9SBAkrcCFwILgNVVdd2s7okk6ZfMSqhU1VeBr7bpmxlcufXwMT8BXj3D+icAJ0xTvwC4oGOrkqQt4B31kqRuDBVJUjeGiiSpG0NFktSNoSJJ6sZQkSR1Y6hIkroxVCRJ3RgqkqRuDBVJUjeGiiSpG0NFktSNoSJJ6sZQkSR1Y6hIkroxVCRJ3RgqkqRuDBVJUjeGiiSpG0NFktSNoSJJ6sZQkSR1Y6hIkroxVCRJ3RgqkqRuDBVJUjeGiiSpG0NFktSNoSJJ6sZQkSR1Y6hIkroxVCRJ3RgqkqRuDBVJUjeGiiSpm7GFSpInJLksybeSXJfkfa2+e5JvJJlK8tkk27b649v8VFu+dGhb72r1G5McMFRf2WpTSY4d175IkkYzziOVB4CXVtVzgT2BlUn2BT4InFRVewB3A0e28UcCd7f6SW0cSZYDhwLPBFYCn0iyIMkC4OPAgcBy4LA2VpI0IWMLlRr4UZt9XPsU8FLg861+BnBwmz6ozdOW75ckrX5OVT1QVd8BpoB92meqqm6uqp8C57SxkqQJGes5lXZEcRVwB7AG+G/gnqra0IasAxa36cXALQBt+b3AU4frD1tnpvp0fRyVZG2StevXr++xa5KkaYw1VKrqwaraE1jC4MjiGeP8vk30cWpVraiqFYsWLZpEC5I0L8zK1V9VdQ/wFeAFwA5JFrZFS4Bb2/StwK4AbfmTgTuH6w9bZ6a6JGlCxnn116IkO7Tp7YCXATcwCJdD2rBVwHlt+vw2T1t+cVVVqx/arg7bHVgGXAZcDixrV5Nty+Bk/vnj2h9J0iNb+MhDNtsuwBntKq1tgHOr6ktJrgfOSfIB4ErgtDb+NOCsJFPAXQxCgqq6Lsm5wPXABuDoqnoQIMlbgQuBBcDqqrpujPsjSXoEYwuVqroaeN409ZsZnF95eP0nwKtn2NYJwAnT1C8ALtjiZiVJXXhHvSSpG0NFktSNoSJJ6sZQkSR1Y6hIkroZKVSSXDRKTZI0v23ykuIkTwB+DdgpyY5A2qLtmeE5W5Kk+euR7lP5E+AdwNOAK/hFqNwHfGyMfUmS5qBNhkpVnQycnORtVfXRWepJkjRHjXRHfVV9NMnvAUuH16mqM8fUlyRpDhopVJKcBfwWcBXwYCsXYKhIkh4y6rO/VgDL21ODJUma1qj3qVwL/OY4G5EkzX2jHqnsBFyf5DLggY3FqnrlWLqSJM1Jo4bKe8fZhCTpsWHUq7/+fdyNSJLmvlGv/vohg6u9ALYFHgf8X1VtP67GJElzz6hHKr++cTpJgIOAfcfVlCRpbnrUTymugS8CB4yhH0nSHDbqz1+vGprdhsF9Kz8ZS0eSpDlr1Ku//mhoegPwXQY/gUmS9JBRz6kcMe5GJElz36gv6VqS5J+T3NE+X0iyZNzNSZLmllFP1H8aOJ/Be1WeBvxLq0mS9JBRQ2VRVX26qja0z+nAojH2JUmag0YNlTuTvD7JgvZ5PXDnOBuTJM09o4bKm4DXALcDtwGHAG8cU0+SpDlq1EuKjwdWVdXdAEmeAnyYQdhIkgSMfqTynI2BAlBVdwHPG09LkqS5atRQ2SbJjhtn2pHKqEc5kqR5YtRg+FvgkiSfa/OvBk4YT0uSpLlq1Dvqz0yyFnhpK72qqq4fX1uSpLlo5J+wWogYJJKkGT3qR99LkjSTsYVKkl2TfCXJ9UmuS/L2Vn9KkjVJbmp/d2z1JDklyVSSq5PsNbStVW38TUlWDdX3TnJNW+eU9gIxSdKEjPNIZQPwzqpazuAtkUcnWQ4cC1xUVcuAi9o8wIHAsvY5CvgkPHSl2XHA84F9gOOGrkT7JPDmofVWjnF/JEmPYGyhUlW3VdU32/QPgRuAxQzew3JGG3YGcHCbPgg4s71Z8lJghyS7MHjD5JqquqvdK7MGWNmWbV9Vl1ZVAWcObUuSNAGzck4lyVIGN0t+A9i5qm5ri24Hdm7Ti4FbhlZb12qbqq+bpi5JmpCxh0qSJwFfAN5RVfcNL2tHGDULPRyVZG2StevXrx/310nSvDXWUEnyOAaB8pmq+qdW/n776Yr2945WvxXYdWj1Ja22qfqSaeq/oqpOraoVVbVi0SKf2C9J4zLOq78CnAbcUFV/N7TofGDjFVyrgPOG6oe3q8D2Be5tP5NdCOyfZMd2gn5/4MK27L4k+7bvOnxoW5KkCRjn87teCLwBuCbJVa32V8CJwLlJjgS+x+CR+gAXAC8HpoD7gSNg8PDKJO8HLm/jjm8PtAR4C3A6sB3w5faRJE3I2EKlqv4TmOm+kf2mGV/A0TNsazWwepr6WuBZW9CmJKkj76iXJHVjqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndjC1UkqxOckeSa4dqT0myJslN7e+OrZ4kpySZSnJ1kr2G1lnVxt+UZNVQfe8k17R1TkmSce2LJGk04zxSOR1Y+bDascBFVbUMuKjNAxwILGufo4BPwiCEgOOA5wP7AMdtDKI25s1D6z38uyRJs2xsoVJVXwPuelj5IOCMNn0GcPBQ/cwauBTYIckuwAHAmqq6q6ruBtYAK9uy7avq0qoq4MyhbUmSJmS2z6nsXFW3tenbgZ3b9GLglqFx61ptU/V109QlSRM0sRP17QijZuO7khyVZG2StevXr5+Nr5SkeWm2Q+X77acr2t87Wv1WYNehcUtabVP1JdPUp1VVp1bViqpasWjRoi3eCUnS9GY7VM4HNl7BtQo4b6h+eLsKbF/g3vYz2YXA/kl2bCfo9wcubMvuS7Jvu+rr8KFtSZImZOG4NpzkbOAlwE5J1jG4iutE4NwkRwLfA17Thl8AvByYAu4HjgCoqruSvB+4vI07vqo2nvx/C4MrzLYDvtw+kqQJGluoVNVhMyzab5qxBRw9w3ZWA6unqa8FnrUlPUqS+vKOeklSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1I2hIknqxlCRJHVjqEiSujFUJEndGCqSpG4MFUlSN4aKJKkbQ0WS1M3CSTewpZKsBE4GFgCfqqoTJ9ySNDH/c/yzJ92CtkJPf881s/Zdc/pIJckC4OPAgcBy4LAkyyfblSTNX3M6VIB9gKmqurmqfgqcAxw04Z4kad6a66GyGLhlaH5dq0mSJmDOn1MZRZKjgKPa7I+S3DjJfh5DdgJ+MOkmtgb58KpJt6Bf5b/PjY7Llm5ht1EHzvVQuRXYdWh+Sav9kqo6FTh1tpqaL5KsraoVk+5Dmo7/Pidjrv/8dTmwLMnuSbYFDgXOn3BPkjRvzekjlarakOStwIUMLileXVXXTbgtSZq35nSoAFTVBcAFk+5jnvInRW3N/Pc5AamqSfcgSXqMmOvnVCRJWxFDRZslycokNyaZSnLspPuRNkqyOskdSa6ddC/zkaGiR83H42grdzqwctJNzFeGijaHj8fRVquqvgbcNek+5itDRZvDx+NImpahIknqxlDR5hjp8TiS5h9DRZvDx+NImpahoketqjYAGx+PcwNwro/H0dYiydnAJcDvJFmX5MhJ9zSfeEe9JKkbj1QkSd0YKpKkbgwVSVI3hookqRtDRZLUzZx/SZc0m5I8CFzD4P+d7wBvqKp7JtuVtPXwSEV6dH5cVXtW1bMYPLTw6Ek3JG1NDBVp813C0IM0kxyT5PIkVyd5X6s9Mcm/JvlWkmuTvLbVv5vkb5Jck+SyJHu0+tIkF7dtXJTk6a1+epJTknw9yc1JDmn1XZJ8LclVbfsvavX9k1yS5JtJPpfkSbP830bzlKEibYb2Tpn9aI+nSbI/sIzBawH2BPZO8mIG7/X436p6bju6+behzdxbVc8GPgZ8pNU+CpxRVc8BPgOcMjR+F+D3gVcAJ7baHwMXVtWewHOBq5LsBLwb+IOq2gtYC/x5z/2XZmKoSI/OdkmuAm4HdgbWtPr+7XMl8E3gGQxC5hrgZUk+mORFVXXv0LbOHvr7gjb9AuAf2/RZDEJkoy9W1c+r6vr23TB4DtsRSd4LPLuqfgjsy+Dlaf/Vel0F7LbFey6NwFCRHp0ft6OC3YDwi3MqAf66nW/Zs6r2qKrTqurbwF4MwuUDSd4ztK2aYXomDwxNBx56IdWLGTwl+vQkh7dla4Z6WV5VPv9Ks8JQkTZDVd0P/BnwziQLGTxc800bz10kWZzkN5I8Dbi/qv4B+BCDgNnotUN/L2nTX2fw1GeA1wH/sak+kuwGfL+q/h74VNv+pcALh87TPDHJb2/RDksj8pJiaTNV1ZVJrgYOq6qzkvwucEkSgB8Brwf2AD6U5OfAz4A/HdrEjm39B4DDWu1twKeTHAOsB454hDZeAhyT5GftOw+vqvVJ3gicneTxbdy7gW9v0Q5LI/ApxdIEJPkusKKqfjDpXqSe/PlLktSNRyqSpG48UpEkdWOoSJK6MVQkSd0YKpKkbgwVSVI3hookqZv/B9Ikd22dJmo2AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(datorig.Response)\n",
    "print(\"Percent of Faulty Examples = {:.1f}\".format(len(datorig[datorig['Response']==1])/len(datorig)*100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Correlations\n",
    "\n",
    "Next let's extract some correlations between the Response label and the Sensor features. Since we have so many features, we can plot a heatmap for the correlations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "corr = datorig.corr()\n",
    "\n",
    "plt.figure(figsize= (10, 10))\n",
    "sns.heatmap(np.abs(corr))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Look at the correlation heatmap above: what do you notice?\n",
    "\n",
    "There are striking blocks of high correlations amidst relatively little correlations between other features. \n",
    "Physically these blocks could be sensors that are all firing at once on a single assembly line, or assigned to a single part. Our ML model must be able to deal with these strong and weak correlations well. A linear model may not be able to perform well when features are strongly correlated. For this reason, we choose a tree based gradient boosting model like XGBoost.\n",
    "\n",
    "Traditionally, we might want to consider methods like PCA to perform some dimensionality reduction, but this comes at a cost of interpretability. In our factory, we want to determine which sensor is giving the anomalous signal corresponding to the faulty part, and PCA will mix the sensor signals together. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# identify what sensors correlate strongly with the faulty label\n",
    "index_list = list(corr['Response'].dropna().index)\n",
    "val_list = np.argsort(np.abs(corr['Response'].dropna().values))[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Response', 'Sensor_95', 'Sensor_1', 'Sensor_81', 'Sensor_63']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get most corelated variables to the Response label. The response variable is ofcourse most correlated with itself \n",
    "#and removed\n",
    "top_corrs = [index_list[x] for x in val_list[:5]]\n",
    "top_corrs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>Response</th>\n",
       "      <th>Sensor_95</th>\n",
       "      <th>Sensor_1</th>\n",
       "      <th>Sensor_81</th>\n",
       "      <th>Sensor_63</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>3352088.64</td>\n",
       "      <td>59696</td>\n",
       "      <td>59695.03</td>\n",
       "      <td>4067232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>27352.32</td>\n",
       "      <td>466</td>\n",
       "      <td>465.20</td>\n",
       "      <td>77066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>4668317.76</td>\n",
       "      <td>56552</td>\n",
       "      <td>56552.38</td>\n",
       "      <td>5737506</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>3172006.08</td>\n",
       "      <td>59726</td>\n",
       "      <td>59726.45</td>\n",
       "      <td>3729806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>2437453.44</td>\n",
       "      <td>41418</td>\n",
       "      <td>41418.67</td>\n",
       "      <td>3160338</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Response   Sensor_95  Sensor_1  Sensor_81  Sensor_63\n",
       "0         0  3352088.64     59696   59695.03    4067232\n",
       "1         0    27352.32       466     465.20      77066\n",
       "2         0  4668317.76     56552   56552.38    5737506\n",
       "3         0  3172006.08     59726   59726.45    3729806\n",
       "4         0  2437453.44     41418   41418.67    3160338"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reduceddat = datorig[top_corrs]\n",
    "reduceddat.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 956.25x900 with 30 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot how the top correlated sensor readings are correlated with each other and the Faulty and Not Faulty Labels\n",
    "g = sns.PairGrid(reduceddat, hue=\"Response\")\n",
    "g.map_diag(plt.hist)\n",
    "g.map_offdiag(plt.scatter)\n",
    "g.add_legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> Model Training and Deployment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Having explored the dataset some, we can now build a model.\n",
    "Our guiding principles for Model choice are as follows:\n",
    "\n",
    "1. There are multiple sensors with weak correlations to the final output variable <br/>\n",
    "2. Outputs are highly skewed and imbalanced, so our baseline model already has 98.4% accuracy. The ML model will need to do better to drive business value <br/>\n",
    "3. Sensor data may itself be correlated with each other and not all sensors may fire at once <br/>\n",
    "\n",
    "A linear model may not work as well in this situation as the outputs are imbalanced and in real life sensor data may be highly correlated. For this reason, we choose a gradient boosting based tree model called XGBoost. \n",
    "\n",
    "SageMaker has a built in implementation of the XGBoost algorithm designed to work at scale. The default data type for XGBoost is LIBSVM format. But our data is in csv and XGBoost also accepts a csv input but it needs to be specified in the 'content_type' variable. We do this first.\n",
    "\n",
    "Next we train a model: This should take less than 5 minutes and incur about 1 minute of training cost. Remember that SageMaker charges you for the time it takes the train and the infrastructure you train on. There are no upfront charges or recurring costs. Once training is complete, SageMaker tears it down automatically."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# NExt we need to push the training and test data to S3\n",
    "train_channel = prefix + '/train'\n",
    "validation_channel = prefix + '/validation'\n",
    "\n",
    "sagemaker_session.upload_data(path='training_data', bucket=bucket, key_prefix=train_channel)\n",
    "sagemaker_session.upload_data(path='test_data', bucket=bucket, key_prefix=validation_channel)\n",
    "\n",
    "#s3_train_data = 's3://{}/{}'.format(bucket, train_channel)\n",
    "#s3_validation_data = 's3://{}/{}'.format(bucket, validation_channel)\n",
    "s3_train_channel = sagemaker.session.s3_input('s3://{}/{}'.format(bucket, train_channel), content_type ='csv')\n",
    "s3_valid_channel = sagemaker.session.s3_input('s3://{}/{}'.format(bucket, validation_channel), content_type ='csv')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2019-10-29 16:24:40 Starting - Starting the training job...\n",
      "2019-10-29 16:24:46 Starting - Launching requested ML instances.........\n",
      "2019-10-29 16:26:24 Starting - Preparing the instances for training......\n",
      "2019-10-29 16:27:17 Downloading - Downloading input data...\n",
      "2019-10-29 16:28:11 Training - Training image download completed. Training in progress.\n",
      "2019-10-29 16:28:11 Uploading - Uploading generated training model\n",
      "2019-10-29 16:28:11 Completed - Training job completed\n",
      "\u001b[31mINFO:sagemaker-containers:Imported framework sagemaker_xgboost_container.training\u001b[0m\n",
      "\u001b[31mINFO:sagemaker-containers:Failed to parse hyperparameter objective value binary:logistic to Json.\u001b[0m\n",
      "\u001b[31mReturning the value itself\u001b[0m\n",
      "\u001b[31mINFO:sagemaker-containers:No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[31mINFO:sagemaker_xgboost_container.training:Running XGBoost Sagemaker in algorithm mode\u001b[0m\n",
      "\u001b[31mINFO:root:Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31mINFO:root:Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31mINFO:root:Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31m[16:27:58] 48000x168 matrix with 8064000 entries loaded from /opt/ml/input/data/train?format=csv&label_column=0&delimiter=,\u001b[0m\n",
      "\u001b[31mINFO:root:Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31m[16:27:58] 12000x168 matrix with 2016000 entries loaded from /opt/ml/input/data/validation?format=csv&label_column=0&delimiter=,\u001b[0m\n",
      "\u001b[31mINFO:root:Single node training.\u001b[0m\n",
      "\u001b[31mINFO:root:Train matrix has 48000 rows\u001b[0m\n",
      "\u001b[31mINFO:root:Validation matrix has 12000 rows\u001b[0m\n",
      "\u001b[31m[0]#011train-error:0.009354#011validation-error:0.01075\u001b[0m\n",
      "\u001b[31m[1]#011train-error:0.008562#011validation-error:0.010167\u001b[0m\n",
      "\u001b[31m[2]#011train-error:0.008104#011validation-error:0.009917\u001b[0m\n",
      "\u001b[31m[3]#011train-error:0.007604#011validation-error:0.009\u001b[0m\n",
      "\u001b[31m[4]#011train-error:0.007208#011validation-error:0.009\u001b[0m\n",
      "\u001b[31m[5]#011train-error:0.007021#011validation-error:0.008667\u001b[0m\n",
      "\u001b[31m[6]#011train-error:0.006667#011validation-error:0.00875\u001b[0m\n",
      "\u001b[31m[7]#011train-error:0.006479#011validation-error:0.008667\u001b[0m\n",
      "\u001b[31m[8]#011train-error:0.006229#011validation-error:0.00825\u001b[0m\n",
      "\u001b[31m[9]#011train-error:0.006104#011validation-error:0.008417\u001b[0m\n",
      "\u001b[31m[10]#011train-error:0.005937#011validation-error:0.007667\u001b[0m\n",
      "\u001b[31m[11]#011train-error:0.006021#011validation-error:0.007917\u001b[0m\n",
      "\u001b[31m[12]#011train-error:0.005833#011validation-error:0.007667\u001b[0m\n",
      "\u001b[31m[13]#011train-error:0.005729#011validation-error:0.007667\u001b[0m\n",
      "\u001b[31m[14]#011train-error:0.005542#011validation-error:0.007583\u001b[0m\n",
      "\u001b[31m[15]#011train-error:0.005354#011validation-error:0.0075\u001b[0m\n",
      "\u001b[31m[16]#011train-error:0.00525#011validation-error:0.0075\u001b[0m\n",
      "\u001b[31m[17]#011train-error:0.005083#011validation-error:0.007583\u001b[0m\n",
      "\u001b[31m[18]#011train-error:0.004875#011validation-error:0.0075\u001b[0m\n",
      "\u001b[31m[19]#011train-error:0.004896#011validation-error:0.00775\u001b[0m\n",
      "\u001b[31m[20]#011train-error:0.004979#011validation-error:0.007667\u001b[0m\n",
      "\u001b[31m[21]#011train-error:0.004917#011validation-error:0.007667\u001b[0m\n",
      "\u001b[31m[22]#011train-error:0.004521#011validation-error:0.007333\u001b[0m\n",
      "\u001b[31m[23]#011train-error:0.004333#011validation-error:0.007333\u001b[0m\n",
      "\u001b[31m[24]#011train-error:0.004438#011validation-error:0.007333\u001b[0m\n",
      "\u001b[31m[25]#011train-error:0.004333#011validation-error:0.007167\u001b[0m\n",
      "\u001b[31m[26]#011train-error:0.004271#011validation-error:0.007083\u001b[0m\n",
      "\u001b[31m[27]#011train-error:0.004208#011validation-error:0.00725\u001b[0m\n",
      "\u001b[31m[28]#011train-error:0.00425#011validation-error:0.00725\u001b[0m\n",
      "\u001b[31m[29]#011train-error:0.004167#011validation-error:0.007167\u001b[0m\n",
      "\u001b[31m[30]#011train-error:0.003958#011validation-error:0.007083\u001b[0m\n",
      "\u001b[31m[31]#011train-error:0.003979#011validation-error:0.007167\u001b[0m\n",
      "\u001b[31m[32]#011train-error:0.004#011validation-error:0.00725\u001b[0m\n",
      "\u001b[31m[33]#011train-error:0.003896#011validation-error:0.00725\u001b[0m\n",
      "\u001b[31m[34]#011train-error:0.003792#011validation-error:0.007083\u001b[0m\n",
      "\u001b[31m[35]#011train-error:0.00375#011validation-error:0.006917\u001b[0m\n",
      "\u001b[31m[36]#011train-error:0.003729#011validation-error:0.007\u001b[0m\n",
      "\u001b[31m[37]#011train-error:0.003667#011validation-error:0.007\u001b[0m\n",
      "\u001b[31m[38]#011train-error:0.003437#011validation-error:0.007083\u001b[0m\n",
      "\u001b[31m[39]#011train-error:0.0035#011validation-error:0.006833\u001b[0m\n",
      "\u001b[31m[40]#011train-error:0.003354#011validation-error:0.0065\u001b[0m\n",
      "\u001b[31m[41]#011train-error:0.003125#011validation-error:0.006583\u001b[0m\n",
      "\u001b[31m[42]#011train-error:0.00325#011validation-error:0.006667\u001b[0m\n",
      "\u001b[31m[43]#011train-error:0.003146#011validation-error:0.006583\u001b[0m\n",
      "\u001b[31m[44]#011train-error:0.003125#011validation-error:0.006583\u001b[0m\n",
      "\u001b[31m[45]#011train-error:0.003062#011validation-error:0.0065\u001b[0m\n",
      "\u001b[31m[46]#011train-error:0.003042#011validation-error:0.0065\u001b[0m\n",
      "\u001b[31m[47]#011train-error:0.002813#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[48]#011train-error:0.002771#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[49]#011train-error:0.002646#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[50]#011train-error:0.002542#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[51]#011train-error:0.002438#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[52]#011train-error:0.002438#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[53]#011train-error:0.002438#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[54]#011train-error:0.002458#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[55]#011train-error:0.002458#011validation-error:0.006583\u001b[0m\n",
      "\u001b[31m[56]#011train-error:0.002417#011validation-error:0.0065\u001b[0m\n",
      "\u001b[31m[57]#011train-error:0.002333#011validation-error:0.006583\u001b[0m\n",
      "\u001b[31m[58]#011train-error:0.002333#011validation-error:0.0065\u001b[0m\n",
      "\u001b[31m[59]#011train-error:0.002313#011validation-error:0.0065\u001b[0m\n",
      "\u001b[31m[60]#011train-error:0.002292#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[61]#011train-error:0.002229#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[62]#011train-error:0.002229#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[63]#011train-error:0.002229#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[64]#011train-error:0.002146#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[65]#011train-error:0.002083#011validation-error:0.0065\u001b[0m\n",
      "\u001b[31m[66]#011train-error:0.002042#011validation-error:0.0065\u001b[0m\n",
      "\u001b[31m[67]#011train-error:0.001958#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[68]#011train-error:0.001896#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[69]#011train-error:0.001917#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[70]#011train-error:0.001917#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[71]#011train-error:0.001917#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[72]#011train-error:0.001896#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[73]#011train-error:0.001896#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[74]#011train-error:0.001833#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[75]#011train-error:0.001917#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[76]#011train-error:0.001896#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[77]#011train-error:0.001833#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[78]#011train-error:0.001792#011validation-error:0.006167\u001b[0m\n",
      "\u001b[31m[79]#011train-error:0.001792#011validation-error:0.006167\u001b[0m\n",
      "\u001b[31m[80]#011train-error:0.001708#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[81]#011train-error:0.001729#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[82]#011train-error:0.001729#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[83]#011train-error:0.001729#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[84]#011train-error:0.001646#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[85]#011train-error:0.001646#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[86]#011train-error:0.001646#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[87]#011train-error:0.001646#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[88]#011train-error:0.001646#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[89]#011train-error:0.001646#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[90]#011train-error:0.001542#011validation-error:0.006167\u001b[0m\n",
      "\u001b[31m[91]#011train-error:0.001583#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[92]#011train-error:0.001583#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[93]#011train-error:0.001583#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[94]#011train-error:0.001542#011validation-error:0.00625\u001b[0m\n",
      "\u001b[31m[95]#011train-error:0.001563#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[96]#011train-error:0.001563#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[97]#011train-error:0.001563#011validation-error:0.006333\u001b[0m\n",
      "\u001b[31m[98]#011train-error:0.001563#011validation-error:0.006417\u001b[0m\n",
      "\u001b[31m[99]#011train-error:0.0015#011validation-error:0.00625\u001b[0m\n",
      "Training seconds: 54\n",
      "Billable seconds: 54\n",
      "CPU times: user 447 ms, sys: 17.9 ms, total: 465 ms\n",
      "Wall time: 3min 42s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "xgb = sagemaker.estimator.Estimator(container,\n",
    "                                    role, \n",
    "                                    train_instance_count=1, \n",
    "                                    train_instance_type='ml.c4.8xlarge',\n",
    "                                    output_path=xgb_upload_location,\n",
    "                                    sagemaker_session=sagemaker_session)\n",
    "xgb.set_hyperparameters(max_depth=5,\n",
    "                        eta=0.2,\n",
    "                        gamma=4,\n",
    "                        min_child_weight=6,\n",
    "                        subsample=0.8,\n",
    "                        silent=0,\n",
    "                        objective='binary:logistic',\n",
    "                        num_round=100)\n",
    "\n",
    "xgb.fit({'train': s3_train_channel, 'validation': s3_valid_channel})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> Model deployment and predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are several options to deploy the model:\n",
    "\n",
    "1. Live Endpoint: this approach is suitable for low latency inference against live data in the cloud. Since we want to make inferences in our factory environment, we don't choose this option. <br/>\n",
    "\n",
    "\n",
    "2. Batch Transform: to test out our model accuracy, we run inferences against some test data. As our dataset is small, the test set here is just the training set + validation set. In reality, this should be completely unseen data, but this will suffice for our MVP. <br/>\n",
    "\n",
    "\n",
    "3. Deploy on Greengrass: Instead of deploying a live endpoint in the AWS Cloud, we will deploy the model locally on our Greengrass core. Follow the steps in the lab to accomplish this. <br/>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> Batch Transform "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Instead of deploying a live endpoint, let's run a Batch Transform job to test the model accuracy and other metrics on the entire dataset. This dataset is saved locally, so we need to first port it over to S3.\n",
    "\n",
    "\n",
    "Once the dataset is in S3, we can call SageMaker Batch Transform to perform inference on the entire dataset and save the output to file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop the Response column from the dataset.\n",
    "test_set = pd.read_csv('rawdataset.csv')\n",
    "resp = test_set['0']\n",
    "test_set = test_set.drop(columns = ['0'])\n",
    "test_set.to_csv('test.csv', index =False, header = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'s3://iot-ml-predictive-maintenance/xgbdata/test/test.csv'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_channel = prefix + '/test'\n",
    "sagemaker_session.upload_data('test.csv', bucket=bucket, key_prefix=test_channel)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Run the batch transform job. Specify the input file location and point batch transform to a path for storing the model outputs.\n",
    "\n",
    "This might take some time again as SageMaker needs to provision the infrastructure needed to perform the batch transform, copy the model artifacts from the trained model s3 path, copy the data to perform batch inferences on and finally, run the Batch transform job."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ".....................\u001b[31m[2019-10-29 16:36:17 +0000] [15] [INFO] Starting gunicorn 19.9.0\u001b[0m\n",
      "\u001b[31m[2019-10-29 16:36:17 +0000] [15] [INFO] Listening at: unix:/tmp/gunicorn.sock (15)\u001b[0m\n",
      "\u001b[31m[2019-10-29 16:36:17 +0000] [15] [INFO] Using worker: gevent\u001b[0m\n",
      "\u001b[31m[2019-10-29 16:36:17 +0000] [22] [INFO] Booting worker with pid: 22\u001b[0m\n",
      "\u001b[31m[2019-10-29 16:36:17 +0000] [26] [INFO] Booting worker with pid: 26\u001b[0m\n",
      "\u001b[31m[2019-10-29 16:36:17 +0000] [27] [INFO] Booting worker with pid: 27\u001b[0m\n",
      "\u001b[31m[2019-10-29 16:36:17 +0000] [28] [INFO] Booting worker with pid: 28\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:38:INFO] No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:38 +0000] \"GET /ping HTTP/1.1\" 200 0 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:38:INFO] No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:38 +0000] \"GET /execution-parameters HTTP/1.1\" 200 84 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:39:INFO] No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:39:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:39:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:39:INFO] No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:39:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:39:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:41 +0000] \"POST /invocations HTTP/1.1\" 200 192009 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:41 +0000] \"POST /invocations HTTP/1.1\" 200 192466 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:41 +0000] \"POST /invocations HTTP/1.1\" 200 192270 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:41:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:41 +0000] \"POST /invocations HTTP/1.1\" 200 192737 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:41:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\u001b[31m[2019-10-29:16:36:41:INFO] Determined delimiter of CSV input is ','\u001b[0m\n",
      "\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:42 +0000] \"POST /invocations HTTP/1.1\" 200 175316 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:43 +0000] \"POST /invocations HTTP/1.1\" 200 192807 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "\u001b[31m169.254.255.130 - - [29/Oct/2019:16:36:43 +0000] \"POST /invocations HTTP/1.1\" 200 192294 \"-\" \"Go-http-client/1.1\"\u001b[0m\n",
      "CPU times: user 586 ms, sys: 85 ms, total: 671 ms\n",
      "Wall time: 4min 14s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "batch_input = 's3://{}/{}'.format(bucket, test_channel) # The location of the test dataset\n",
    "\n",
    "batch_output = 's3://{}/{}/batch-inference'.format(bucket, test_channel) # The location to store the results of the batch transform job\n",
    "\n",
    "transformer = xgb.transformer(instance_count=1, instance_type='ml.m4.xlarge', output_path=batch_output)\n",
    "\n",
    "transformer.transform(data=batch_input, data_type='S3Prefix', content_type='text/csv', split_type='Line')\n",
    "\n",
    "try:\n",
    "    transformer.wait()\n",
    "except Exception as e:\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Completed 256.0 KiB/1.3 MiB (2.3 MiB/s) with 1 file(s) remaining\r",
      "Completed 512.0 KiB/1.3 MiB (4.4 MiB/s) with 1 file(s) remaining\r",
      "Completed 768.0 KiB/1.3 MiB (6.4 MiB/s) with 1 file(s) remaining\r",
      "Completed 1.0 MiB/1.3 MiB (8.3 MiB/s) with 1 file(s) remaining  \r",
      "Completed 1.2 MiB/1.3 MiB (10.1 MiB/s) with 1 file(s) remaining \r",
      "Completed 1.3 MiB/1.3 MiB (10.3 MiB/s) with 1 file(s) remaining \r",
      "download: s3://iot-ml-predictive-maintenance/xgbdata/test/batch-inference/test.csv.out to data/inference/test.csv.out\r\n"
     ]
    }
   ],
   "source": [
    "# Let's download the inference into a local directory\n",
    "data_dir = './data/inference'\n",
    "if not os.path.exists(data_dir):\n",
    "    os.makedirs(data_dir)\n",
    "\n",
    "!aws s3 cp --recursive $transformer.output_path $data_dir"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_test = pd.read_csv(os.path.join(data_dir, 'test.csv.out'), header=None)\n",
    "y_vals = np.round(y_test.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Metrics and Economics\n",
    "\n",
    "Lets get some model metrics.\n",
    "\n",
    "From a business perspective, we need to consider more than just the model accuracy but let's get that first.\n",
    "\n",
    "Two other metrics to look at are precision and recall. How do these manifest themselves in IoT Predictive maintenance?\n",
    "\n",
    "Precision is a measure of the number of false positives. High precision indicates low false positives. The added cost of high false positives is that the factory engineers and workers have to spend time and money troubleshooting and performing maintenance when its not required.\n",
    "\n",
    "Imagine the other situation: if the part is actually in need of maintenance and our model predicts that it is fine. \n",
    "In IoT settings, particularly in heavy industries, a large cost is incurred when there is \"unexpected downtime\". Especially for mission critical equipment, this can shut down the entire site. A high recall score (low false negatives) will help minimize this as it will ensure that when a part *is* in need of maintenance, it is flagged.\n",
    "\n",
    "An understanding of precision and recall must be tied in with the economics of the particular use case. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model Accuracy = 99.75499591659862 %\n"
     ]
    }
   ],
   "source": [
    "print(\"Model Accuracy = {} %\".format(accuracy_score(resp.values, y_vals)*100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall that the baseline model was 98.4% accurate. We are getting 99.75% which beats the baseline accuracy with our XGBoost model. But the business may care more about Recall and Precision. Let's extract those next"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     58999\n",
      "           1       0.97      0.88      0.92      1000\n",
      "\n",
      "   micro avg       1.00      1.00      1.00     59999\n",
      "   macro avg       0.98      0.94      0.96     59999\n",
      "weighted avg       1.00      1.00      1.00     59999\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(resp.values, y_vals))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[58969    30]\n",
      " [  117   883]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_confusion_matrix(y_true, y_pred, classes,\n",
    "                          normalize=False,\n",
    "                          title=None,\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    if not title:\n",
    "        if normalize:\n",
    "            title = 'Normalized confusion matrix'\n",
    "        else:\n",
    "            title = 'Confusion matrix, without normalization'\n",
    "\n",
    "    # Compute confusion matrix\n",
    "    cm = confusion_matrix(y_true, y_pred)\n",
    "    # Only use the labels that appear in the data\n",
    "    classes = classes\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    fig, ax = plt.subplots(figsize = (8, 8))\n",
    "    im = ax.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    ax.figure.colorbar(im, ax=ax)\n",
    "    # We want to show all ticks...\n",
    "    ax.set(xticks=np.arange(cm.shape[1]),\n",
    "           yticks=np.arange(cm.shape[0]),\n",
    "     # ... and label them with the respective list entries\n",
    "           xticklabels=classes, yticklabels=classes)\n",
    "    plt.tick_params(labelsize=15)  \n",
    "    plt.xlabel('Predicted label', fontsize=18)\n",
    "    plt.ylabel('True label',fontsize =18)\n",
    "    plt.title(title, fontsize=18)\n",
    "    # Rotate the tick labels and set their alignment.\n",
    "    plt.setp(ax.get_xticklabels(), rotation=45, ha=\"right\",\n",
    "             rotation_mode=\"anchor\")\n",
    "\n",
    "    # Loop over data dimensions and create text annotations.\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i in range(cm.shape[0]):\n",
    "        for j in range(cm.shape[1]):\n",
    "            ax.text(j, i, format(cm[i, j], fmt),\n",
    "                    ha=\"center\", va=\"center\", fontsize=20,\n",
    "                    color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "    fig.tight_layout()\n",
    "    return ax\n",
    "\n",
    "\n",
    "np.set_printoptions(precision=2)\n",
    "\n",
    "# Plot non-normalized confusion matrix\n",
    "plot_confusion_matrix(resp.values, y_vals, classes=['Normal', 'Faulty'],\n",
    "                      title='Confusion matrix, without normalization')\n",
    "\n",
    "# Plot normalized confusion matrix\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our model is not perfect but does much better than the naive model with 98.4% accuracy. What this means is that although the model still misses some examples of faulty labels, it correctly captures a large number of them that the naive model would not be able to.\n",
    "\n",
    "From a business standpoint, the business now has a much smaller number of recalled parts (883), at the expense of only 30 parts which the model incorrectly classified as Faulty, even though they were actually okay. It is these tradeoffs between business metrics and machine learning that requires multiple stakeholders to regularly engage in order to achieve good results.\n",
    "\n",
    "We can now deploy this model on to Greengrass.\n",
    "\n",
    "Go back to the predictive maintenance tutorial to continue building out this use case"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Bonus Steps"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As a next step, we will use SageMaker Neo to reduce the size of the ML model we have trained. Neo minimizes models to run up to twice as fast, with no loss in accuracy. This can be crucial for deploying on edge devices such as a RaspberryPi, NVIDIA Jetson Nano etc. Here we will compile the model for deploying on an EC2 instance we're using to represent Greengrass.\n",
    "\n",
    "\n",
    "As an exercise, once this model is trained, deploy this to your Greengrass Core and tag it with the Lambda function as shown in the tutorial."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "?..!"
     ]
    }
   ],
   "source": [
    "compiled_model = xgb\n",
    "try:\n",
    "    xgb.create_model()._neo_image_account(sagemaker_session.boto_region_name)\n",
    "except:\n",
    "    print('Neo is not currently supported in', sagemaker_session.boto_region_name)\n",
    "else:\n",
    "    output_path = '/'.join(xgb.output_path.split('/')[:-1]) + '/neo'\n",
    "    compiled_model = xgb.compile_model(target_instance_family='ml_m4', \n",
    "                                   input_shape={'data':[1, 168]},\n",
    "                                   role=role,\n",
    "                                   framework='xgboost',\n",
    "                                   framework_version='0.90-1',\n",
    "                                   output_path=output_path)\n",
    "    compiled_model.name = 'deployed-xgboost-customer-churn'\n",
    "    compiled_model.image = get_image_uri(region, 'xgboost-neo', repo_version='latest')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's write a simple function to get the Model size in both cases"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_model_size(estimator):\n",
    "    out= !aws s3 ls {estimator.model_data} --human-readable\n",
    "    return out[0].split(' ')[-3]+' MB'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>Targets</th>\n",
       "      <th>Locations</th>\n",
       "      <th>Sizes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Original</td>\n",
       "      <td>s3://iot-ml-predictive-maintenance/xgb/sagemaker-xgboost-2019-10-29-16-24-40-212/output/model.tar.gz</td>\n",
       "      <td>48.9 MB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Neo for ml.m4</td>\n",
       "      <td>s3://iot-ml-predictive-maintenance/neo/model-ml_m4.tar.gz</td>\n",
       "      <td>24.2 MB</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Targets  \\\n",
       "0  Original        \n",
       "1  Neo for ml.m4   \n",
       "\n",
       "                                                                                              Locations  \\\n",
       "0  s3://iot-ml-predictive-maintenance/xgb/sagemaker-xgboost-2019-10-29-16-24-40-212/output/model.tar.gz   \n",
       "1  s3://iot-ml-predictive-maintenance/neo/model-ml_m4.tar.gz                                              \n",
       "\n",
       "     Sizes  \n",
       "0  48.9 MB  \n",
       "1  24.2 MB  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "estimators = [xgb, compiled_model] \n",
    "targets = ['Original', 'Neo for ml.m4']\n",
    "locations = [e.model_data for e in estimators]\n",
    "sizes = [get_model_size(e) for e in estimators]\n",
    "pd.set_option('display.max_colwidth', 0)\n",
    "pd.DataFrame(list(zip(targets,locations,sizes)), columns =['Targets', 'Locations','Sizes'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notice that SageMaker Neo reduces the model size dramatically, which can make it easier for deploying on edge devides for limited RAM and Memory."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Thank you!"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}