Human_Activity_Tracker.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!wget 'http://archive.ics.uci.edu/ml/machine-learning-databases/00507/wisdm-dataset.zip'\n",
    "# Loading the dataset from the repository"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import zipfile #for extracting zip file\n",
    "with zipfile.ZipFile('./wisdm-dataset.zip', 'r') as zip_ref:\n",
    "    zip_ref.extractall()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "        <script type=\"text/javascript\">\n",
       "        window.PlotlyConfig = {MathJaxConfig: 'local'};\n",
       "        if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}\n",
       "        if (typeof require !== 'undefined') {\n",
       "        require.undef(\"plotly\");\n",
       "        requirejs.config({\n",
       "            paths: {\n",
       "                'plotly': ['https://cdn.plot.ly/plotly-latest.min']\n",
       "            }\n",
       "        });\n",
       "        require(['plotly'], function(Plotly) {\n",
       "            window._Plotly = Plotly;\n",
       "        });\n",
       "        }\n",
       "        </script>\n",
       "        "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from scipy.io import arff\n",
    "import os\n",
    "\n",
    "import numpy\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import preprocessing\n",
    "\n",
    "# To store data\n",
    "import pandas as pd\n",
    "\n",
    "# To do linear algebra\n",
    "import numpy as np\n",
    "from numpy import pi\n",
    "\n",
    "# To create plots\n",
    "import seaborn as sns\n",
    "from matplotlib.colors import rgb2hex\n",
    "from matplotlib.cm import get_cmap\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "# To create interactive plots\n",
    "from plotly.offline import init_notebook_mode, iplot\n",
    "import plotly.graph_objs as go\n",
    "init_notebook_mode(connected=True)\n",
    "\n",
    "# To get new datatypes and functions\n",
    "from collections import Counter\n",
    "from cycler import cycler\n",
    "\n",
    "# To investigate distributions\n",
    "from scipy.stats import norm, skew, probplot\n",
    "from scipy.optimize import curve_fit\n",
    "\n",
    "# To build models\n",
    "from sklearn import preprocessing\n",
    "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.manifold import TSNE\n",
    "\n",
    "# To gbm light\n",
    "# from lightgbm import LGBMClassifier\n",
    "\n",
    "# To measure time\n",
    "from time import time\n",
    "\n",
    "sns.set(style='whitegrid', palette=\"deep\", font_scale=1.1, rc={\"figure.figsize\": [8, 5]}) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import boto3\n",
    "import sagemaker\n",
    "\n",
    "# session and role\n",
    "sagemaker_session = sagemaker.Session()\n",
    "role = sagemaker.get_execution_role()\n",
    "\n",
    "# create an S3 bucket\n",
    "bucket = sagemaker_session.default_bucket()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def arff_file_to_df(filename):\n",
    "    #convert arff files to pandas data frame\n",
    "    data = arff.loadarff(filename)\n",
    "    df = pd.DataFrame(data[0])\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = 'wisdm-dataset'\n",
    "files = []\n",
    "frames = []\n",
    "\n",
    "# r=root, d=directories, f = files\n",
    "for r, d, f in os.walk(path):\n",
    "    for file in f:\n",
    "        if '.arff' in file:\n",
    "            files.append(os.path.join(r, file))\n",
    "for file in files:\n",
    "    df = arff_file_to_df(file)\n",
    "    frames.append(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat(frames)\n",
    "#Appendding all the dataframes row wise to create a single dataframe df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(75099, 93)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>\"ACTIVITY\"</th>\n",
       "      <th>\"X0\"</th>\n",
       "      <th>\"X1\"</th>\n",
       "      <th>\"X2\"</th>\n",
       "      <th>\"X3\"</th>\n",
       "      <th>\"X4\"</th>\n",
       "      <th>\"X5\"</th>\n",
       "      <th>\"X6\"</th>\n",
       "      <th>\"X7\"</th>\n",
       "      <th>\"X8\"</th>\n",
       "      <th>...</th>\n",
       "      <th>\"ZMFCC11\"</th>\n",
       "      <th>\"ZMFCC12\"</th>\n",
       "      <th>\"XYCOS\"</th>\n",
       "      <th>\"XZCOS\"</th>\n",
       "      <th>\"YZCOS\"</th>\n",
       "      <th>\"XYCOR\"</th>\n",
       "      <th>\"XZCOR\"</th>\n",
       "      <th>\"YZCOR\"</th>\n",
       "      <th>\"RESULTANT\"</th>\n",
       "      <th>\"class\"</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b'A'</td>\n",
       "      <td>0.170</td>\n",
       "      <td>0.315</td>\n",
       "      <td>0.435</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.010</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.433341</td>\n",
       "      <td>0.427773</td>\n",
       "      <td>-0.188714</td>\n",
       "      <td>0.141920</td>\n",
       "      <td>-0.062945</td>\n",
       "      <td>-0.204853</td>\n",
       "      <td>0.140685</td>\n",
       "      <td>-0.067840</td>\n",
       "      <td>3.86209</td>\n",
       "      <td>b'1619'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b'A'</td>\n",
       "      <td>0.125</td>\n",
       "      <td>0.310</td>\n",
       "      <td>0.485</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.386114</td>\n",
       "      <td>0.381152</td>\n",
       "      <td>-0.371368</td>\n",
       "      <td>0.198086</td>\n",
       "      <td>0.442477</td>\n",
       "      <td>-0.372734</td>\n",
       "      <td>0.199274</td>\n",
       "      <td>0.446024</td>\n",
       "      <td>3.23334</td>\n",
       "      <td>b'1619'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>b'A'</td>\n",
       "      <td>0.150</td>\n",
       "      <td>0.320</td>\n",
       "      <td>0.485</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.005</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.405939</td>\n",
       "      <td>0.400722</td>\n",
       "      <td>-0.522678</td>\n",
       "      <td>-0.038708</td>\n",
       "      <td>0.519698</td>\n",
       "      <td>-0.517238</td>\n",
       "      <td>-0.038878</td>\n",
       "      <td>0.525614</td>\n",
       "      <td>2.93205</td>\n",
       "      <td>b'1619'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>b'A'</td>\n",
       "      <td>0.190</td>\n",
       "      <td>0.220</td>\n",
       "      <td>0.430</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.426505</td>\n",
       "      <td>0.421024</td>\n",
       "      <td>-0.642675</td>\n",
       "      <td>0.057995</td>\n",
       "      <td>0.128622</td>\n",
       "      <td>-0.643005</td>\n",
       "      <td>0.057987</td>\n",
       "      <td>0.129048</td>\n",
       "      <td>3.83345</td>\n",
       "      <td>b'1619'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>b'A'</td>\n",
       "      <td>0.205</td>\n",
       "      <td>0.240</td>\n",
       "      <td>0.370</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.015</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.469978</td>\n",
       "      <td>0.463939</td>\n",
       "      <td>-0.603190</td>\n",
       "      <td>-0.031606</td>\n",
       "      <td>0.137222</td>\n",
       "      <td>-0.603988</td>\n",
       "      <td>-0.031405</td>\n",
       "      <td>0.138634</td>\n",
       "      <td>4.36682</td>\n",
       "      <td>b'1619'</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 93 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  \"ACTIVITY\"   \"X0\"   \"X1\"   \"X2\"  \"X3\"   \"X4\"  \"X5\"  \"X6\"  \"X7\"  \"X8\"  ...  \\\n",
       "0       b'A'  0.170  0.315  0.435  0.06  0.010   0.0  0.01   0.0   0.0  ...   \n",
       "1       b'A'  0.125  0.310  0.485  0.08  0.000   0.0  0.00   0.0   0.0  ...   \n",
       "2       b'A'  0.150  0.320  0.485  0.04  0.005   0.0  0.00   0.0   0.0  ...   \n",
       "3       b'A'  0.190  0.220  0.430  0.16  0.000   0.0  0.00   0.0   0.0  ...   \n",
       "4       b'A'  0.205  0.240  0.370  0.17  0.015   0.0  0.00   0.0   0.0  ...   \n",
       "\n",
       "   \"ZMFCC11\"  \"ZMFCC12\"   \"XYCOS\"   \"XZCOS\"   \"YZCOS\"   \"XYCOR\"   \"XZCOR\"  \\\n",
       "0   0.433341   0.427773 -0.188714  0.141920 -0.062945 -0.204853  0.140685   \n",
       "1   0.386114   0.381152 -0.371368  0.198086  0.442477 -0.372734  0.199274   \n",
       "2   0.405939   0.400722 -0.522678 -0.038708  0.519698 -0.517238 -0.038878   \n",
       "3   0.426505   0.421024 -0.642675  0.057995  0.128622 -0.643005  0.057987   \n",
       "4   0.469978   0.463939 -0.603190 -0.031606  0.137222 -0.603988 -0.031405   \n",
       "\n",
       "    \"YZCOR\"  \"RESULTANT\"  \"class\"  \n",
       "0 -0.067840      3.86209  b'1619'  \n",
       "1  0.446024      3.23334  b'1619'  \n",
       "2  0.525614      2.93205  b'1619'  \n",
       "3  0.129048      3.83345  b'1619'  \n",
       "4  0.138634      4.36682  b'1619'  \n",
       "\n",
       "[5 rows x 93 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.columns=df.columns.str.replace('\"','')\n",
    "# Remove quotes from column names\n",
    "def clean_column(column_name):\n",
    "    #remmove quotes and unnecessary characters from the column values\n",
    "    df[column_name] = df[column_name].astype(str).str.replace('b','')\n",
    "    df[column_name] = df[column_name].astype(str).str.replace(\"'\",'')\n",
    "    \n",
    "    return df[column_name]\n",
    "\n",
    "df['ACTIVITY'] = clean_column('ACTIVITY')\n",
    "df['class'] = clean_column('class')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>ACTIVITY</th>\n",
       "      <th>X0</th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "      <th>X7</th>\n",
       "      <th>X8</th>\n",
       "      <th>...</th>\n",
       "      <th>ZMFCC11</th>\n",
       "      <th>ZMFCC12</th>\n",
       "      <th>XYCOS</th>\n",
       "      <th>XZCOS</th>\n",
       "      <th>YZCOS</th>\n",
       "      <th>XYCOR</th>\n",
       "      <th>XZCOR</th>\n",
       "      <th>YZCOR</th>\n",
       "      <th>RESULTANT</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A</td>\n",
       "      <td>0.170</td>\n",
       "      <td>0.315</td>\n",
       "      <td>0.435</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.010</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.433341</td>\n",
       "      <td>0.427773</td>\n",
       "      <td>-0.188714</td>\n",
       "      <td>0.141920</td>\n",
       "      <td>-0.062945</td>\n",
       "      <td>-0.204853</td>\n",
       "      <td>0.140685</td>\n",
       "      <td>-0.067840</td>\n",
       "      <td>3.86209</td>\n",
       "      <td>1619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A</td>\n",
       "      <td>0.125</td>\n",
       "      <td>0.310</td>\n",
       "      <td>0.485</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.386114</td>\n",
       "      <td>0.381152</td>\n",
       "      <td>-0.371368</td>\n",
       "      <td>0.198086</td>\n",
       "      <td>0.442477</td>\n",
       "      <td>-0.372734</td>\n",
       "      <td>0.199274</td>\n",
       "      <td>0.446024</td>\n",
       "      <td>3.23334</td>\n",
       "      <td>1619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A</td>\n",
       "      <td>0.150</td>\n",
       "      <td>0.320</td>\n",
       "      <td>0.485</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.005</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.405939</td>\n",
       "      <td>0.400722</td>\n",
       "      <td>-0.522678</td>\n",
       "      <td>-0.038708</td>\n",
       "      <td>0.519698</td>\n",
       "      <td>-0.517238</td>\n",
       "      <td>-0.038878</td>\n",
       "      <td>0.525614</td>\n",
       "      <td>2.93205</td>\n",
       "      <td>1619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>A</td>\n",
       "      <td>0.190</td>\n",
       "      <td>0.220</td>\n",
       "      <td>0.430</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.426505</td>\n",
       "      <td>0.421024</td>\n",
       "      <td>-0.642675</td>\n",
       "      <td>0.057995</td>\n",
       "      <td>0.128622</td>\n",
       "      <td>-0.643005</td>\n",
       "      <td>0.057987</td>\n",
       "      <td>0.129048</td>\n",
       "      <td>3.83345</td>\n",
       "      <td>1619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A</td>\n",
       "      <td>0.205</td>\n",
       "      <td>0.240</td>\n",
       "      <td>0.370</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.015</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.469978</td>\n",
       "      <td>0.463939</td>\n",
       "      <td>-0.603190</td>\n",
       "      <td>-0.031606</td>\n",
       "      <td>0.137222</td>\n",
       "      <td>-0.603988</td>\n",
       "      <td>-0.031405</td>\n",
       "      <td>0.138634</td>\n",
       "      <td>4.36682</td>\n",
       "      <td>1619</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 93 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  ACTIVITY     X0     X1     X2    X3     X4   X5    X6   X7   X8  ...  \\\n",
       "0        A  0.170  0.315  0.435  0.06  0.010  0.0  0.01  0.0  0.0  ...   \n",
       "1        A  0.125  0.310  0.485  0.08  0.000  0.0  0.00  0.0  0.0  ...   \n",
       "2        A  0.150  0.320  0.485  0.04  0.005  0.0  0.00  0.0  0.0  ...   \n",
       "3        A  0.190  0.220  0.430  0.16  0.000  0.0  0.00  0.0  0.0  ...   \n",
       "4        A  0.205  0.240  0.370  0.17  0.015  0.0  0.00  0.0  0.0  ...   \n",
       "\n",
       "    ZMFCC11   ZMFCC12     XYCOS     XZCOS     YZCOS     XYCOR     XZCOR  \\\n",
       "0  0.433341  0.427773 -0.188714  0.141920 -0.062945 -0.204853  0.140685   \n",
       "1  0.386114  0.381152 -0.371368  0.198086  0.442477 -0.372734  0.199274   \n",
       "2  0.405939  0.400722 -0.522678 -0.038708  0.519698 -0.517238 -0.038878   \n",
       "3  0.426505  0.421024 -0.642675  0.057995  0.128622 -0.643005  0.057987   \n",
       "4  0.469978  0.463939 -0.603190 -0.031606  0.137222 -0.603988 -0.031405   \n",
       "\n",
       "      YZCOR  RESULTANT  class  \n",
       "0 -0.067840    3.86209   1619  \n",
       "1  0.446024    3.23334   1619  \n",
       "2  0.525614    2.93205   1619  \n",
       "3  0.129048    3.83345   1619  \n",
       "4  0.138634    4.36682   1619  \n",
       "\n",
       "[5 rows x 93 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "activty_map = {\n",
    "    'A': 'walking',\n",
    "    'B': 'jogging',\n",
    "    'C': 'stairs',\n",
    "    'D': 'sitting',\n",
    "    'E': 'standing',\n",
    "    'F': 'typing',\n",
    "    'G': 'teeth',\n",
    "    'H': 'soup',\n",
    "    'I': 'chips',\n",
    "    'J': 'pasta',\n",
    "    'K': 'drinking',\n",
    "    'L': 'sandwich',\n",
    "    'M': 'kicking',\n",
    "    'O': 'catch',\n",
    "    'P': 'dribbling',\n",
    "    'Q': 'writing',\n",
    "    'R': 'clapping',\n",
    "    'S': 'folding'\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# temp_activity_df = df.iloc[:,0]\n",
    "# temp_class_df = df.iloc[:,92]\n",
    "# temp_df = df.iloc[:,1:92].astype(numpy.float32)\n",
    "# df = pd.concat([temp_df, temp_class_df, temp_activity_df], axis=1)\n",
    "# print(df.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp = df.ACTIVITY.value_counts()\n",
    "values = temp.values\n",
    "labels = list(temp.index)\n",
    "\n",
    "labels = [activty_map[label] for label in labels]\n",
    "\n",
    "plt.pie(values, labels=labels, autopct='%1.1f%%',counterclock=False, radius=2)\n",
    "plt.title('Activity Data Distribution')\n",
    "plt.show()\n",
    "# Visualizing the activity data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The data is uniformly distributed amongst the activity labels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<BarContainer object of 50 artists>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp = df['class'].value_counts()\n",
    "values = temp.values\n",
    "labels = list(temp.index)\n",
    "plt.figure(figsize=(10,10))\n",
    "plt.xticks(rotation=90)\n",
    "plt.xlabel('Subjects', fontsize=10)\n",
    "plt.ylabel('Count', fontsize=10)\n",
    "plt.title('Count of activities by Individuals', fontsize=20)\n",
    "plt.bar(labels, values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Null Values In DataFrame: 0\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 75099 entries, 0 to 407\n",
      "Data columns (total 93 columns):\n",
      "ACTIVITY     75099 non-null object\n",
      "X0           75099 non-null float64\n",
      "X1           75099 non-null float64\n",
      "X2           75099 non-null float64\n",
      "X3           75099 non-null float64\n",
      "X4           75099 non-null float64\n",
      "X5           75099 non-null float64\n",
      "X6           75099 non-null float64\n",
      "X7           75099 non-null float64\n",
      "X8           75099 non-null float64\n",
      "X9           75099 non-null float64\n",
      "Y0           75099 non-null float64\n",
      "Y1           75099 non-null float64\n",
      "Y2           75099 non-null float64\n",
      "Y3           75099 non-null float64\n",
      "Y4           75099 non-null float64\n",
      "Y5           75099 non-null float64\n",
      "Y6           75099 non-null float64\n",
      "Y7           75099 non-null float64\n",
      "Y8           75099 non-null float64\n",
      "Y9           75099 non-null float64\n",
      "Z0           75099 non-null float64\n",
      "Z1           75099 non-null float64\n",
      "Z2           75099 non-null float64\n",
      "Z3           75099 non-null float64\n",
      "Z4           75099 non-null float64\n",
      "Z5           75099 non-null float64\n",
      "Z6           75099 non-null float64\n",
      "Z7           75099 non-null float64\n",
      "Z8           75099 non-null float64\n",
      "Z9           75099 non-null float64\n",
      "XAVG         75099 non-null float64\n",
      "YAVG         75099 non-null float64\n",
      "ZAVG         75099 non-null float64\n",
      "XPEAK        75099 non-null float64\n",
      "YPEAK        75099 non-null float64\n",
      "ZPEAK        75099 non-null float64\n",
      "XABSOLDEV    75099 non-null float64\n",
      "YABSOLDEV    75099 non-null float64\n",
      "ZABSOLDEV    75099 non-null float64\n",
      "XSTANDDEV    75099 non-null float64\n",
      "YSTANDDEV    75099 non-null float64\n",
      "ZSTANDDEV    75099 non-null float64\n",
      "XVAR         75099 non-null float64\n",
      "YVAR         75099 non-null float64\n",
      "ZVAR         75099 non-null float64\n",
      "XMFCC0       75099 non-null float64\n",
      "XMFCC1       75099 non-null float64\n",
      "XMFCC2       75099 non-null float64\n",
      "XMFCC3       75099 non-null float64\n",
      "XMFCC4       75099 non-null float64\n",
      "XMFCC5       75099 non-null float64\n",
      "XMFCC6       75099 non-null float64\n",
      "XMFCC7       75099 non-null float64\n",
      "XMFCC8       75099 non-null float64\n",
      "XMFCC9       75099 non-null float64\n",
      "XMFCC10      75099 non-null float64\n",
      "XMFCC11      75099 non-null float64\n",
      "XMFCC12      75099 non-null float64\n",
      "YMFCC0       75099 non-null float64\n",
      "YMFCC1       75099 non-null float64\n",
      "YMFCC2       75099 non-null float64\n",
      "YMFCC3       75099 non-null float64\n",
      "YMFCC4       75099 non-null float64\n",
      "YMFCC5       75099 non-null float64\n",
      "YMFCC6       75099 non-null float64\n",
      "YMFCC7       75099 non-null float64\n",
      "YMFCC8       75099 non-null float64\n",
      "YMFCC9       75099 non-null float64\n",
      "YMFCC10      75099 non-null float64\n",
      "YMFCC11      75099 non-null float64\n",
      "YMFCC12      75099 non-null float64\n",
      "ZMFCC0       75099 non-null float64\n",
      "ZMFCC1       75099 non-null float64\n",
      "ZMFCC2       75099 non-null float64\n",
      "ZMFCC3       75099 non-null float64\n",
      "ZMFCC4       75099 non-null float64\n",
      "ZMFCC5       75099 non-null float64\n",
      "ZMFCC6       75099 non-null float64\n",
      "ZMFCC7       75099 non-null float64\n",
      "ZMFCC8       75099 non-null float64\n",
      "ZMFCC9       75099 non-null float64\n",
      "ZMFCC10      75099 non-null float64\n",
      "ZMFCC11      75099 non-null float64\n",
      "ZMFCC12      75099 non-null float64\n",
      "XYCOS        75099 non-null float64\n",
      "XZCOS        75099 non-null float64\n",
      "YZCOS        75099 non-null float64\n",
      "XYCOR        75099 non-null float64\n",
      "XZCOR        75099 non-null float64\n",
      "YZCOR        75099 non-null float64\n",
      "RESULTANT    75099 non-null float64\n",
      "class        75099 non-null object\n",
      "dtypes: float64(91), object(2)\n",
      "memory usage: 53.9+ MB\n"
     ]
    }
   ],
   "source": [
    "# Get the datatypes of columns\n",
    "df.describe()\n",
    "\n",
    "# Get null values and dataframe information\n",
    "print('Null Values In DataFrame: {}\\n'.format(df.isna().sum().sum()))\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def activty_mapper(activity):\n",
    "    if(activity in ['A', 'B', 'C', 'D', 'E', 'M']):\n",
    "        return 0\n",
    "    elif(activity in ['P', 'O', 'F', 'Q', 'R','G', 'S']):\n",
    "        return 1\n",
    "    else:\n",
    "        return 2\n",
    "\n",
    "df['ACTIVITY'] = df['ACTIVITY'].apply(activty_mapper)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     ACTIVITY    X0    X1    X2   X3   X4   X5   X6   X7   X8  ...   ZMFCC11  \\\n",
      "481         0  0.04  0.85  0.11  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.398749   \n",
      "300         1  0.00  0.46  0.54  0.0  0.0  0.0  0.0  0.0  0.0  ... -0.518413   \n",
      "84          0  0.00  0.58  0.42  0.0  0.0  0.0  0.0  0.0  0.0  ... -0.599415   \n",
      "58          0  0.00  1.00  0.00  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.438649   \n",
      "154         1  1.00  0.00  0.00  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.437647   \n",
      "\n",
      "      ZMFCC12     XYCOS     XZCOS     YZCOS     XYCOR     XZCOR     YZCOR  \\\n",
      "481  0.393625 -0.739168 -0.574235  0.576139  0.226855 -0.211418 -0.596378   \n",
      "300 -0.511751 -0.758820 -0.079382  0.485555 -0.755616 -0.082952  0.490093   \n",
      "84  -0.591712 -0.056533  0.159908  0.085791 -0.070513  0.152080  0.077642   \n",
      "58   0.433012  0.990842 -0.998585 -0.991308  0.104353  0.031198  0.169282   \n",
      "154  0.432024 -0.701279 -0.998312  0.702214  0.016179  0.111496 -0.099057   \n",
      "\n",
      "     RESULTANT  class  \n",
      "481  10.144800   1642  \n",
      "300   0.031457   1610  \n",
      "84    0.016921   1639  \n",
      "58    9.217230   1632  \n",
      "154   9.671100   1619  \n",
      "\n",
      "[5 rows x 93 columns]\n"
     ]
    }
   ],
   "source": [
    "df = df.sample(frac=1, random_state=1)\n",
    "# Shuffle the data inside the dataframe\n",
    "print(df.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "individuals = df.pop('class')\n",
    "activity = df.pop('ACTIVITY')\n",
    "#get the ids of the indiviual in a separate list and remove that colummn from data frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "activty_map = {\n",
    "    0: 'Non Hand Oriented',\n",
    "    1: 'Hand Oriented General',\n",
    "    2: 'Hand Oriented Eating'\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "linkText": "Export to plot.ly",
        "plotlyServerURL": "https://plot.ly",
        "showLink": false
       },
       "data": [
        {
         "marker": {
          "color": [
           "#440154",
           "#21918c",
           "#fde725"
          ]
         },
         "type": "bar",
         "x": [
          "Hand Oriented General",
          "Non Hand Oriented",
          "Hand Oriented Eating"
         ],
         "y": [
          29342,
          25107,
          20650
         ]
        }
       ],
       "layout": {
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "scatter": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Human Activity Label Distribution"
        },
        "xaxis": {
         "title": {
          "text": "Activity"
         }
        },
        "yaxis": {
         "title": {
          "text": "Count"
         }
        }
       }
      },
      "text/html": [
       "<div>\n",
       "        \n",
       "        \n",
       "            <div id=\"206557f2-2e71-460b-83fa-8d47ffc35e83\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>\n",
       "            <script type=\"text/javascript\">\n",
       "                require([\"plotly\"], function(Plotly) {\n",
       "                    window.PLOTLYENV=window.PLOTLYENV || {};\n",
       "                    \n",
       "                if (document.getElementById(\"206557f2-2e71-460b-83fa-8d47ffc35e83\")) {\n",
       "                    Plotly.newPlot(\n",
       "                        '206557f2-2e71-460b-83fa-8d47ffc35e83',\n",
       "                        [{\"marker\": {\"color\": [\"#440154\", \"#21918c\", \"#fde725\"]}, \"type\": \"bar\", \"x\": [\"Hand Oriented General\", \"Non Hand Oriented\", \"Hand Oriented Eating\"], \"y\": [29342, 25107, 20650]}],\n",
       "                        {\"template\": {\"data\": {\"bar\": [{\"error_x\": {\"color\": \"#2a3f5f\"}, \"error_y\": {\"color\": \"#2a3f5f\"}, \"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"bar\"}], \"barpolar\": [{\"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"barpolar\"}], \"carpet\": [{\"aaxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"baxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"type\": \"carpet\"}], \"choropleth\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"choropleth\"}], \"contour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"contour\"}], \"contourcarpet\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"contourcarpet\"}], \"heatmap\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmap\"}], \"heatmapgl\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmapgl\"}], \"histogram\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"histogram\"}], \"histogram2d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2d\"}], \"histogram2dcontour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2dcontour\"}], \"mesh3d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"mesh3d\"}], \"parcoords\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"parcoords\"}], \"scatter\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter\"}], \"scatter3d\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter3d\"}], \"scattercarpet\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattercarpet\"}], \"scattergeo\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergeo\"}], \"scattergl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergl\"}], \"scattermapbox\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattermapbox\"}], \"scatterpolar\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolar\"}], \"scatterpolargl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolargl\"}], \"scatterternary\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterternary\"}], \"surface\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"surface\"}], \"table\": [{\"cells\": {\"fill\": {\"color\": \"#EBF0F8\"}, \"line\": {\"color\": \"white\"}}, \"header\": {\"fill\": {\"color\": \"#C8D4E3\"}, \"line\": {\"color\": \"white\"}}, \"type\": \"table\"}]}, \"layout\": {\"annotationdefaults\": {\"arrowcolor\": \"#2a3f5f\", \"arrowhead\": 0, \"arrowwidth\": 1}, \"colorscale\": {\"diverging\": [[0, \"#8e0152\"], [0.1, \"#c51b7d\"], [0.2, \"#de77ae\"], [0.3, \"#f1b6da\"], [0.4, \"#fde0ef\"], [0.5, \"#f7f7f7\"], [0.6, \"#e6f5d0\"], [0.7, \"#b8e186\"], [0.8, \"#7fbc41\"], [0.9, \"#4d9221\"], [1, \"#276419\"]], \"sequential\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"sequentialminus\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"colorway\": [\"#636efa\", \"#EF553B\", \"#00cc96\", \"#ab63fa\", \"#FFA15A\", \"#19d3f3\", \"#FF6692\", \"#B6E880\", \"#FF97FF\", \"#FECB52\"], \"font\": {\"color\": \"#2a3f5f\"}, \"geo\": {\"bgcolor\": \"white\", \"lakecolor\": \"white\", \"landcolor\": \"#E5ECF6\", \"showlakes\": true, \"showland\": true, \"subunitcolor\": \"white\"}, \"hoverlabel\": {\"align\": \"left\"}, \"hovermode\": \"closest\", \"mapbox\": {\"style\": \"light\"}, \"paper_bgcolor\": \"white\", \"plot_bgcolor\": \"#E5ECF6\", \"polar\": {\"angularaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"radialaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"scene\": {\"xaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"yaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"zaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}}, \"shapedefaults\": {\"line\": {\"color\": \"#2a3f5f\"}}, \"ternary\": {\"aaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"baxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"caxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"title\": {\"x\": 0.05}, \"xaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}, \"yaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}}}, \"title\": {\"text\": \"Human Activity Label Distribution\"}, \"xaxis\": {\"title\": {\"text\": \"Activity\"}}, \"yaxis\": {\"title\": {\"text\": \"Count\"}}},\n",
       "                        {\"responsive\": true}\n",
       "                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('206557f2-2e71-460b-83fa-8d47ffc35e83');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })\n",
       "                };\n",
       "                });\n",
       "            </script>\n",
       "        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting data\n",
    "label_counts = activity.value_counts()\n",
    "\n",
    "x_labels = [activty_map[index] for index in label_counts.index]\n",
    "# Get colors\n",
    "n = label_counts.shape[0]\n",
    "colormap = get_cmap('viridis')\n",
    "colors = [rgb2hex(colormap(col)) for col in np.arange(0, 1.01, 1/(n-1))]\n",
    "\n",
    "# Create plot\n",
    "data = go.Bar(x = x_labels,\n",
    "              y = label_counts,\n",
    "              marker = dict(color = colors))\n",
    "\n",
    "layout = go.Layout(title = 'Human Activity Label Distribution',\n",
    "                   xaxis = dict(title = 'Activity'),\n",
    "                   yaxis = dict(title = 'Count'))\n",
    "\n",
    "fig = go.Figure(data=[data], layout=layout)\n",
    "iplot(fig)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### We observe that the distribution of data in the three groups is not entirely uniform. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Normalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "# we scale the values of columns so that all the values are between 0 and 1\n",
    "scaler = MinMaxScaler()\n",
    "df[df.columns] = scaler.fit_transform(df[df.columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = df.to_numpy(dtype=np.float32)\n",
    "activity_labels = activity.to_numpy(dtype=np.float32)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "xTrain, xTest, yTrain, yTest = train_test_split(features, activity_labels, test_size = 0.2, random_state = 0)\n",
    "xVal, xTest, yVal, yTest = train_test_split(xTest, yTest, test_size = 0.5, random_state = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training size:  60079\n",
      "Test size:  7510\n",
      "Validation size:  7510\n",
      "\n",
      "Training df sample: \n",
      " [[0.00000000e+00 5.60000002e-01 4.39999998e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 6.49999976e-01\n",
      "  3.49999994e-01 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 3.35000008e-01 6.65000021e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 4.58920598e-01 6.53594494e-01\n",
      "  5.20891130e-01 5.20741567e-02 8.50413963e-02 6.15954632e-03\n",
      "  9.02114494e-04 7.02926714e-04 2.60425790e-04 8.41249654e-04\n",
      "  7.13959394e-04 2.32849212e-04 2.41695400e-02 2.28345431e-02\n",
      "  1.15161026e-02 2.72834957e-01 2.72834122e-01 2.72836149e-01\n",
      "  2.72836417e-01 2.72833943e-01 2.72834420e-01 2.72835195e-01\n",
      "  2.72834599e-01 2.72836179e-01 2.72833377e-01 2.72837132e-01\n",
      "  2.72834837e-01 2.72833675e-01 3.11300963e-01 3.11301887e-01\n",
      "  3.11301559e-01 3.11299592e-01 3.11301172e-01 3.11301947e-01\n",
      "  3.11302364e-01 3.11300784e-01 3.11300635e-01 3.11302841e-01\n",
      "  3.11302245e-01 3.11302662e-01 3.11302423e-01 2.27058649e-01\n",
      "  2.27057070e-01 2.27058083e-01 2.27058440e-01 2.27059469e-01\n",
      "  2.27060497e-01 2.27058411e-01 2.27058411e-01 2.27057040e-01\n",
      "  2.27057695e-01 2.27060303e-01 2.27059886e-01 2.27058664e-01\n",
      "  1.95035577e-01 3.92918676e-01 5.68347514e-01 1.38668954e-01\n",
      "  4.40076172e-01 6.21279597e-01 6.05705136e-04]\n",
      " [0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  2.24999994e-01 7.74999976e-01 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 9.25000012e-01 7.50000030e-02\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  1.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 7.41693795e-01 5.57328641e-01\n",
      "  2.60276169e-01 3.98397781e-02 1.04677670e-01 1.99475549e-02\n",
      "  1.55168790e-02 9.80561040e-03 1.53820664e-02 1.39809018e-02\n",
      "  9.64574981e-03 1.32752443e-02 1.13507830e-01 9.43827257e-02\n",
      "  1.11373097e-01 7.50389397e-01 7.50388980e-01 7.50389516e-01\n",
      "  7.50389814e-01 7.50388801e-01 7.50388980e-01 7.50389397e-01\n",
      "  7.50389278e-01 7.50389755e-01 7.50388682e-01 7.50389934e-01\n",
      "  7.50389338e-01 7.50388801e-01 6.52312934e-01 6.52312875e-01\n",
      "  6.52312815e-01 6.52311623e-01 6.52312636e-01 6.52313054e-01\n",
      "  6.52313530e-01 6.52312517e-01 6.52312279e-01 6.52313292e-01\n",
      "  6.52313054e-01 6.52313232e-01 6.52313292e-01 7.15116799e-01\n",
      "  7.15116501e-01 7.15116978e-01 7.15116918e-01 7.15117157e-01\n",
      "  7.15117633e-01 7.15117335e-01 7.15116858e-01 7.15116620e-01\n",
      "  7.15116918e-01 7.15117335e-01 7.15117455e-01 7.15117276e-01\n",
      "  2.73301499e-03 1.88350561e-03 9.98062968e-01 7.00007021e-01\n",
      "  9.52780008e-01 6.41110063e-01 2.64197350e-01]\n",
      " [0.00000000e+00 1.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 1.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 1.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 4.25747991e-01 8.64437163e-01\n",
      "  8.86337936e-01 1.48116602e-02 3.29654962e-02 9.75294970e-03\n",
      "  1.09825993e-03 1.64558389e-03 1.86777464e-03 1.09579728e-03\n",
      "  1.64519111e-03 1.66909059e-03 2.82275565e-02 3.66166830e-02\n",
      "  3.68338116e-02 5.41385114e-01 5.41384101e-01 5.41385293e-01\n",
      "  5.41385531e-01 5.41384041e-01 5.41384280e-01 5.41384876e-01\n",
      "  5.41384578e-01 5.41385531e-01 5.41383862e-01 5.41386187e-01\n",
      "  5.41384757e-01 5.41383922e-01 7.24756479e-01 7.24756479e-01\n",
      "  7.24756241e-01 7.24755228e-01 7.24756002e-01 7.24756300e-01\n",
      "  7.24756837e-01 7.24755824e-01 7.24756062e-01 7.24756896e-01\n",
      "  7.24756837e-01 7.24756598e-01 7.24756777e-01 7.44234204e-01\n",
      "  7.44233489e-01 7.44233787e-01 7.44233847e-01 7.44234025e-01\n",
      "  7.44234502e-01 7.44233787e-01 7.44233608e-01 7.44233727e-01\n",
      "  7.44233549e-01 7.44234383e-01 7.44234145e-01 7.44233787e-01\n",
      "  1.47000814e-04 1.34500398e-04 9.99978006e-01 5.23505330e-01\n",
      "  4.55664963e-01 3.21034580e-01 2.62697011e-01]\n",
      " [2.04999998e-01 2.04999998e-01 1.80000007e-01 1.55000001e-01\n",
      "  1.40000001e-01 5.99999987e-02 2.50000004e-02 1.05263162e-02\n",
      "  3.50877196e-02 4.05405387e-02 8.70000005e-01 7.00000003e-02\n",
      "  5.99999987e-02 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  2.00000003e-01 3.14999998e-01 2.94999987e-01 9.00000036e-02\n",
      "  5.99999987e-02 1.49999997e-02 2.56410260e-02 5.12820520e-02\n",
      "  3.12500000e-02 0.00000000e+00 5.21331072e-01 3.15758675e-01\n",
      "  5.07665634e-01 7.88502917e-02 1.65964887e-01 3.85241695e-02\n",
      "  2.74129272e-01 2.31389761e-01 1.93174139e-01 2.81120509e-01\n",
      "  2.45879769e-01 2.15941444e-01 5.27638853e-01 4.93682802e-01\n",
      "  4.62335587e-01 8.54308784e-01 8.54308009e-01 8.54308486e-01\n",
      "  8.54308248e-01 8.54307950e-01 8.54307890e-01 8.54308367e-01\n",
      "  8.54308307e-01 8.54308367e-01 8.54308128e-01 8.54308784e-01\n",
      "  8.54308248e-01 8.54307771e-01 8.77255440e-01 8.77255440e-01\n",
      "  8.77254963e-01 8.77254903e-01 8.77255261e-01 8.77255201e-01\n",
      "  8.77255619e-01 8.77254903e-01 8.77254784e-01 8.77255678e-01\n",
      "  8.77255380e-01 8.77255201e-01 8.77255321e-01 8.44687462e-01\n",
      "  8.44687164e-01 8.44687223e-01 8.44687462e-01 8.44687521e-01\n",
      "  8.44687998e-01 8.44687581e-01 8.44687641e-01 8.44687223e-01\n",
      "  8.44687581e-01 8.44687700e-01 8.44687462e-01 8.44687223e-01\n",
      "  2.51187891e-01 2.67770797e-01 6.60138786e-01 2.64227957e-01\n",
      "  2.64787376e-01 7.56703734e-01 3.29090565e-01]\n",
      " [0.00000000e+00 4.74999994e-01 5.24999976e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 3.75000000e-01\n",
      "  6.25000000e-01 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 5.00000000e-01 5.00000000e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 4.59155798e-01 6.53810859e-01\n",
      "  5.20763874e-01 8.26253965e-02 9.01010782e-02 4.07754034e-02\n",
      "  7.90587452e-04 5.89964853e-04 2.62009795e-04 7.29840307e-04\n",
      "  5.74741280e-04 2.33741404e-04 2.22070329e-02 2.01158747e-02\n",
      "  1.15442872e-02 2.46281296e-01 2.46280164e-01 2.46282145e-01\n",
      "  2.46282697e-01 2.46280000e-01 2.46280462e-01 2.46281534e-01\n",
      "  2.46280596e-01 2.46282607e-01 2.46279970e-01 2.46283218e-01\n",
      "  2.46281177e-01 2.46280074e-01 2.83203214e-01 2.83203870e-01\n",
      "  2.83203691e-01 2.83201456e-01 2.83203125e-01 2.83203810e-01\n",
      "  2.83204198e-01 2.83202648e-01 2.83203006e-01 2.83204496e-01\n",
      "  2.83204556e-01 2.83204943e-01 2.83204764e-01 2.09812894e-01\n",
      "  2.09810972e-01 2.09812358e-01 2.09812120e-01 2.09813505e-01\n",
      "  2.09814474e-01 2.09812462e-01 2.09812403e-01 2.09810853e-01\n",
      "  2.09811494e-01 2.09814385e-01 2.09813684e-01 2.09812865e-01\n",
      "  6.99612856e-01 2.55375773e-01 3.01149547e-01 6.95403039e-01\n",
      "  2.57039219e-01 3.03526253e-01 5.18689048e-04]\n",
      " [0.00000000e+00 5.69999993e-01 4.30000007e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 4.85000014e-01\n",
      "  5.14999986e-01 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 5.14999986e-01 4.85000014e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 4.58217591e-01 6.53757572e-01\n",
      "  5.21085262e-01 2.43827850e-02 6.22728430e-02 1.89645085e-02\n",
      "  3.74016259e-03 7.51773012e-04 1.21238548e-03 9.03122313e-03\n",
      "  8.39796965e-04 2.76760501e-03 9.02034417e-02 2.50764657e-02\n",
      "  4.85861376e-02 5.47211170e-01 5.47210157e-01 5.47211349e-01\n",
      "  5.47211587e-01 5.47210097e-01 5.47210276e-01 5.47210872e-01\n",
      "  5.47210634e-01 5.47211587e-01 5.47209918e-01 5.47212183e-01\n",
      "  5.47210813e-01 5.47209978e-01 3.35240155e-01 3.35240483e-01\n",
      "  3.35240424e-01 3.35238427e-01 3.35240394e-01 3.35240602e-01\n",
      "  3.35241258e-01 3.35240096e-01 3.35239857e-01 3.35241675e-01\n",
      "  3.35241258e-01 3.35241556e-01 3.35241765e-01 4.61702555e-01\n",
      "  4.61701334e-01 4.61702198e-01 4.61702526e-01 4.61703032e-01\n",
      "  4.61703777e-01 4.61702734e-01 4.61702257e-01 4.61701512e-01\n",
      "  4.61702019e-01 4.61703897e-01 4.61703479e-01 4.61702734e-01\n",
      "  3.04303676e-01 2.53005754e-02 6.63124800e-01 3.04040402e-01\n",
      "  2.52666343e-02 6.62729681e-01 1.65685290e-03]\n",
      " [1.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 6.44999981e-01 2.80000001e-01\n",
      "  7.50000030e-02 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  5.99999987e-02 3.89999986e-01 4.69999999e-01 7.99999982e-02\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 1.19009688e-01 5.45783877e-01\n",
      "  5.26756644e-01 7.33132511e-02 1.85054377e-01 4.05403823e-02\n",
      "  5.82602061e-02 7.66970739e-02 8.02783370e-02 6.15981072e-02\n",
      "  7.59143233e-02 7.42983967e-02 2.44103730e-01 2.72403598e-01\n",
      "  2.69381225e-01 7.87199020e-01 7.87198424e-01 7.87198782e-01\n",
      "  7.87198842e-01 7.87198424e-01 7.87198484e-01 7.87198663e-01\n",
      "  7.87198901e-01 7.87198842e-01 7.87198544e-01 7.87199318e-01\n",
      "  7.87198722e-01 7.87198484e-01 7.46914446e-01 7.46914625e-01\n",
      "  7.46914625e-01 7.46914029e-01 7.46914625e-01 7.46914804e-01\n",
      "  7.46914983e-01 7.46914387e-01 7.46914506e-01 7.46914983e-01\n",
      "  7.46915221e-01 7.46915042e-01 7.46914923e-01 7.34436512e-01\n",
      "  7.34436154e-01 7.34436333e-01 7.34436452e-01 7.34436750e-01\n",
      "  7.34437585e-01 7.34436929e-01 7.34436631e-01 7.34436035e-01\n",
      "  7.34436572e-01 7.34437287e-01 7.34437048e-01 7.34436572e-01\n",
      "  9.11497533e-01 4.75769222e-01 2.96065539e-01 3.02229702e-01\n",
      "  6.03803992e-01 1.67389050e-01 2.72241294e-01]\n",
      " [5.14999986e-01 1.94999993e-01 2.89999992e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 7.20000029e-01 1.85000002e-01\n",
      "  9.49999988e-02 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 1.19999997e-01 5.99999987e-02 1.59999996e-01\n",
      "  5.69999993e-01 7.50000030e-02 3.84615399e-02 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 3.32809180e-01 4.73768622e-01\n",
      "  7.51537204e-01 3.78377140e-02 4.27441671e-02 2.81863678e-02\n",
      "  2.44958758e-01 1.46039367e-01 1.22010171e-01 2.06559598e-01\n",
      "  1.37211010e-01 1.22131765e-01 4.51507807e-01 3.67701143e-01\n",
      "  3.46609592e-01 7.54852176e-01 7.54851639e-01 7.54851937e-01\n",
      "  7.54851997e-01 7.54851103e-01 7.54851162e-01 7.54851878e-01\n",
      "  7.54851818e-01 7.54852057e-01 7.54851043e-01 7.54852533e-01\n",
      "  7.54851818e-01 7.54851282e-01 8.01401973e-01 8.01401913e-01\n",
      "  8.01401734e-01 8.01401317e-01 8.01401913e-01 8.01402092e-01\n",
      "  8.01402271e-01 8.01402032e-01 8.01401675e-01 8.01402330e-01\n",
      "  8.01402092e-01 8.01402032e-01 8.01402509e-01 7.51633465e-01\n",
      "  7.51633465e-01 7.51633644e-01 7.51633763e-01 7.51634061e-01\n",
      "  7.51634657e-01 7.51634121e-01 7.51634002e-01 7.51633644e-01\n",
      "  7.51633763e-01 7.51634121e-01 7.51634121e-01 7.51633644e-01\n",
      "  6.33139968e-01 2.90865362e-01 1.57664046e-01 1.49297148e-01\n",
      "  7.14829385e-01 5.76948285e-01 2.62350172e-01]\n",
      " [1.29999995e-01 3.84999990e-01 3.79999995e-01 7.50000030e-02\n",
      "  9.99999978e-03 9.99999978e-03 9.99999978e-03 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 5.00000007e-02 4.35000002e-01\n",
      "  4.95000005e-01 1.99999996e-02 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  1.50000006e-01 4.79999989e-01 1.94999993e-01 1.25000000e-01\n",
      "  4.50000018e-02 4.99999989e-03 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 4.54628021e-01 6.46437109e-01\n",
      "  5.18453300e-01 7.78087452e-02 1.10885561e-01 9.36049819e-02\n",
      "  1.23826027e-01 4.25015204e-02 1.23880975e-01 1.45487979e-01\n",
      "  4.72662933e-02 1.15461171e-01 3.78053367e-01 2.14043126e-01\n",
      "  3.36890399e-01 7.99956739e-01 7.99955904e-01 7.99956739e-01\n",
      "  7.99956739e-01 7.99956143e-01 7.99955964e-01 7.99956083e-01\n",
      "  7.99956560e-01 7.99956799e-01 7.99955904e-01 7.99956858e-01\n",
      "  7.99956262e-01 7.99956143e-01 7.06835091e-01 7.06835091e-01\n",
      "  7.06834793e-01 7.06833959e-01 7.06834614e-01 7.06835151e-01\n",
      "  7.06835389e-01 7.06834495e-01 7.06834614e-01 7.06835151e-01\n",
      "  7.06835330e-01 7.06834972e-01 7.06835091e-01 7.53994823e-01\n",
      "  7.53994465e-01 7.53994823e-01 7.53995240e-01 7.53995121e-01\n",
      "  7.53996074e-01 7.53995121e-01 7.53995240e-01 7.53994823e-01\n",
      "  7.53994942e-01 7.53995538e-01 7.53995359e-01 7.53995240e-01\n",
      "  5.02669513e-01 3.78610641e-01 4.86559898e-01 4.98692840e-01\n",
      "  3.78064454e-01 4.84655201e-01 8.92242044e-02]\n",
      " [0.00000000e+00 5.64999998e-01 4.35000002e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 5.04999995e-01\n",
      "  4.95000005e-01 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 5.45000017e-01 4.55000013e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00 4.57552701e-01 6.54526830e-01\n",
      "  5.15666604e-01 2.14494504e-02 9.60040316e-02 2.32244134e-02\n",
      "  8.48946162e-03 4.18777345e-03 1.13698142e-02 1.06361611e-02\n",
      "  6.47366932e-03 1.52269462e-02 9.83347520e-02 7.65732005e-02\n",
      "  1.19582489e-01 5.43630421e-01 5.43629527e-01 5.43630660e-01\n",
      "  5.43630898e-01 5.43629408e-01 5.43629646e-01 5.43630183e-01\n",
      "  5.43629885e-01 5.43630898e-01 5.43629229e-01 5.43631554e-01\n",
      "  5.43630123e-01 5.43629289e-01 5.14228523e-01 5.14228821e-01\n",
      "  5.14228642e-01 5.14227271e-01 5.14228463e-01 5.14228880e-01\n",
      "  5.14229298e-01 5.14228225e-01 5.14228165e-01 5.14229536e-01\n",
      "  5.14229298e-01 5.14229417e-01 5.14229476e-01 5.57762682e-01\n",
      "  5.57761669e-01 5.57762325e-01 5.57762504e-01 5.57763100e-01\n",
      "  5.57763755e-01 5.57762623e-01 5.57762444e-01 5.57761729e-01\n",
      "  5.57762146e-01 5.57763517e-01 5.57763338e-01 5.57762682e-01\n",
      "  2.26854250e-01 8.21947992e-01 1.28588453e-01 2.32879460e-01\n",
      "  8.11826587e-01 1.28995672e-01 6.13461109e-03]]\n"
     ]
    }
   ],
   "source": [
    "# check that division of samples seems correct\n",
    "# these should add up to 95 (100 - 5 original files)\n",
    "print('Training size: ', len(xTrain))\n",
    "print('Test size: ', len(xTest))\n",
    "print('Validation size: ', len(xVal))\n",
    "\n",
    "print()\n",
    "print('Training df sample: \\n', xTrain[:10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "linear_learner_estimator = sagemaker.LinearLearner(role=sagemaker.get_execution_role(),\n",
    "                                               train_instance_count=1,\n",
    "                                               train_instance_type='ml.m4.xlarge',\n",
    "                                               predictor_type='multiclass_classifier',\n",
    "                                               num_classes=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create RecordSet\n",
    "train_records = linear_learner_estimator.record_set(xTrain, yTrain, channel='train')\n",
    "val_records = linear_learner_estimator.record_set(xVal, yVal, channel='validation')\n",
    "test_records = linear_learner_estimator.record_set(xTest, yTest, channel='test')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2020-01-24 17:45:39 Starting - Starting the training job...\n",
      "2020-01-24 17:45:40 Starting - Launching requested ML instances...\n",
      "2020-01-24 17:46:37 Starting - Preparing the instances for training......\n",
      "2020-01-24 17:47:38 Downloading - Downloading input data...\n",
      "2020-01-24 17:47:53 Training - Downloading the training image..\u001b[34mDocker entrypoint called with argument(s): train\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] Reading default configuration from /opt/amazon/lib/python2.7/site-packages/algorithm/resources/default-input.json: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'auto', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'1000', u'huber_delta': u'1.0', u'num_classes': u'1', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] Reading provided configuration from /opt/ml/input/config/hyperparameters.json: {u'feature_dim': u'91', u'mini_batch_size': u'1000', u'predictor_type': u'multiclass_classifier', u'num_classes': u'3'}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] Final configuration: {u'loss_insensitivity': u'0.01', u'epochs': u'15', u'feature_dim': u'91', u'init_bias': u'0.0', u'lr_scheduler_factor': u'auto', u'num_calibration_samples': u'10000000', u'accuracy_top_k': u'3', u'_num_kv_servers': u'auto', u'use_bias': u'true', u'num_point_for_scaler': u'10000', u'_log_level': u'info', u'quantile': u'0.5', u'bias_lr_mult': u'auto', u'lr_scheduler_step': u'auto', u'init_method': u'uniform', u'init_sigma': u'0.01', u'lr_scheduler_minimum_lr': u'auto', u'target_recall': u'0.8', u'num_models': u'auto', u'early_stopping_patience': u'3', u'momentum': u'auto', u'unbias_label': u'auto', u'wd': u'auto', u'optimizer': u'auto', u'_tuning_objective_metric': u'', u'early_stopping_tolerance': u'0.001', u'learning_rate': u'auto', u'_kvstore': u'auto', u'normalize_data': u'true', u'binary_classifier_model_selection_criteria': u'accuracy', u'use_lr_scheduler': u'true', u'target_precision': u'0.8', u'unbias_data': u'auto', u'init_scale': u'0.07', u'bias_wd_mult': u'auto', u'f_beta': u'1.0', u'mini_batch_size': u'1000', u'huber_delta': u'1.0', u'num_classes': u'3', u'predictor_type': u'multiclass_classifier', u'beta_1': u'auto', u'loss': u'auto', u'beta_2': u'auto', u'_enable_profiler': u'false', u'normalize_label': u'auto', u'_num_gpus': u'auto', u'balance_multiclass_weights': u'false', u'positive_example_weight_mult': u'1.0', u'l1': u'auto', u'margin': u'1.0'}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Loggers have already been setup.\u001b[0m\n",
      "\u001b[34mProcess 1 is a worker.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] Using default worker.\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:16.657] [tensorio] [warning] TensorIO is already initialized; ignoring the initialization routine.\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:16.664] [tensorio] [warning] TensorIO is already initialized; ignoring the initialization routine.\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:16.707] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 0, \"duration\": 55, \"num_examples\": 1, \"num_bytes\": 412000}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] Create Store: local\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:16.776] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 1, \"duration\": 67, \"num_examples\": 11, \"num_bytes\": 4532000}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] Scaler algorithm parameters\n",
      " <algorithm.scaler.ScalerAlgorithmStable object at 0x7f945b15d350>\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] Scaling model computed with parameters:\n",
      " {'stdev_weight': \u001b[0m\n",
      "\u001b[34m[0.33562687 0.2805359  0.26561087 0.1791667  0.17462423 0.2114733\n",
      " 0.05832912 0.05410581 0.04519978 0.06581129 0.4140131  0.27967215\n",
      " 0.26674724 0.14489874 0.10180167 0.16227338 0.0440416  0.05029359\n",
      " 0.03902063 0.04619246 0.27505833 0.26410356 0.2524848  0.19208325\n",
      " 0.19429605 0.19512995 0.04161161 0.03680304 0.03064851 0.03771748\n",
      " 0.15275978 0.14654133 0.15305209 0.0325244  0.05426938 0.02575855\n",
      " 0.11022007 0.08432315 0.08548461 0.10813726 0.0854397  0.08265294\n",
      " 0.17244068 0.15204793 0.15263724 0.17873491 0.1787352  0.17873463\n",
      " 0.17873453 0.17873512 0.17873505 0.1787348  0.17873506 0.17873457\n",
      " 0.1787353  0.17873445 0.17873488 0.17873515 0.17480156 0.1748014\n",
      " 0.17480138 0.17480189 0.1748015  0.17480141 0.17480122 0.17480162\n",
      " 0.17480159 0.17480119 0.17480122 0.17480107 0.17480117 0.17356203\n",
      " 0.1735625  0.1735622  0.1735622  0.17356187 0.17356177 0.1735622\n",
      " 0.1735622  0.17356257 0.17356236 0.17356174 0.1735618  0.17356203\n",
      " 0.3256555  0.3125532  0.3191923  0.21070611 0.22251475 0.21867669\n",
      " 0.12925674]\u001b[0m\n",
      "\u001b[34m<NDArray 91 @cpu(0)>, 'stdev_label': None, 'mean_label': None, 'mean_weight': \u001b[0m\n",
      "\u001b[34m[0.18877411 0.31200725 0.28356406 0.06516909 0.05333364 0.06778908\n",
      " 0.01521364 0.01344306 0.01043222 0.01294717 0.3132364  0.30852273\n",
      " 0.26566955 0.040345   0.02053591 0.03785864 0.008381   0.00887762\n",
      " 0.00620996 0.0059697  0.13593501 0.3320218  0.31967998 0.08525137\n",
      " 0.06532909 0.05357908 0.01036247 0.00775991 0.00565625 0.00611962\n",
      " 0.4705214  0.6020736  0.5461286  0.04583655 0.09048669 0.03305996\n",
      " 0.07400002 0.05311849 0.06017501 0.0764074  0.05649736 0.05899986\n",
      " 0.2125352  0.17973867 0.18594219 0.66444457 0.66444373 0.6644446\n",
      " 0.66444474 0.6644437  0.66444373 0.66444427 0.6644441  0.66444474\n",
      " 0.66444355 0.66444534 0.6644442  0.66444355 0.6567853  0.6567855\n",
      " 0.6567853  0.6567843  0.6567853  0.6567855  0.65678585 0.6567851\n",
      " 0.65678495 0.656786   0.65678585 0.65678585 0.65678596 0.6452872\n",
      " 0.64528656 0.64528704 0.6452872  0.6452876  0.6452881  0.6452872\n",
      " 0.6452872  0.6452867  0.64528686 0.645288   0.64528775 0.6452872\n",
      " 0.5151189  0.48929808 0.4987596  0.47302192 0.51803035 0.51380754\n",
      " 0.16494292]\u001b[0m\n",
      "\u001b[34m<NDArray 91 @cpu(0)>}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] nvidia-smi took: 0.0251801013947 secs to identify 0 gpus\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 INFO 140276589545280] Number of GPUs being used: 0\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:16 WARNING 140276589545280] Setting num_models to 32.\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 0, \"sum\": 0.0, \"min\": 0}, \"Total Batches Seen\": {\"count\": 1, \"max\": 12, \"sum\": 12.0, \"min\": 12}, \"Total Records Seen\": {\"count\": 1, \"max\": 12000, \"sum\": 12000.0, \"min\": 12000}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 11000, \"sum\": 11000.0, \"min\": 11000}, \"Reset Count\": {\"count\": 1, \"max\": 2, \"sum\": 2.0, \"min\": 2}}, \"EndTime\": 1579888096.987173, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"init_train_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1579888096.987132}\n",
      "\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:19.708] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 3, \"duration\": 2720, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.924326205444336, \"sum\": 0.924326205444336, \"min\": 0.924326205444336}}, \"EndTime\": 1579888099.708594, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.708501}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.107098846435547, \"sum\": 1.107098846435547, \"min\": 1.107098846435547}}, \"EndTime\": 1579888099.708702, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.70868}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5113025634765624, \"sum\": 1.5113025634765624, \"min\": 1.5113025634765624}}, \"EndTime\": 1579888099.708766, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.708752}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9518657796223958, \"sum\": 0.9518657796223958, \"min\": 0.9518657796223958}}, \"EndTime\": 1579888099.708825, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.708813}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0973142791748047, \"sum\": 1.0973142791748047, \"min\": 1.0973142791748047}}, \"EndTime\": 1579888099.708884, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.708867}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.992035801188151, \"sum\": 0.992035801188151, \"min\": 0.992035801188151}}, \"EndTime\": 1579888099.708958, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.708938}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9531870198567708, \"sum\": 0.9531870198567708, \"min\": 0.9531870198567708}}, \"EndTime\": 1579888099.709031, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709013}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0019986450195313, \"sum\": 1.0019986450195313, \"min\": 1.0019986450195313}}, \"EndTime\": 1579888099.709097, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.70908}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1632694030761719, \"sum\": 1.1632694030761719, \"min\": 1.1632694030761719}}, \"EndTime\": 1579888099.70917, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709153}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.5247825439453124, \"sum\": 3.5247825439453124, \"min\": 3.5247825439453124}}, \"EndTime\": 1579888099.709242, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709224}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9681121419270833, \"sum\": 0.9681121419270833, \"min\": 0.9681121419270833}}, \"EndTime\": 1579888099.709314, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709295}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.030328086344401, \"sum\": 1.030328086344401, \"min\": 1.030328086344401}}, \"EndTime\": 1579888099.709375, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709358}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9374443725585937, \"sum\": 0.9374443725585937, \"min\": 0.9374443725585937}}, \"EndTime\": 1579888099.709432, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709418}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1359935221354167, \"sum\": 1.1359935221354167, \"min\": 1.1359935221354167}}, \"EndTime\": 1579888099.709472, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709458}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.0638847412109373, \"sum\": 3.0638847412109373, \"min\": 3.0638847412109373}}, \"EndTime\": 1579888099.70953, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709514}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9549829956054687, \"sum\": 0.9549829956054687, \"min\": 0.9549829956054687}}, \"EndTime\": 1579888099.709573, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709557}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9447216044108073, \"sum\": 0.9447216044108073, \"min\": 0.9447216044108073}}, \"EndTime\": 1579888099.709635, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709618}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9526611592610678, \"sum\": 0.9526611592610678, \"min\": 0.9526611592610678}}, \"EndTime\": 1579888099.709696, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709679}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.189287803141276, \"sum\": 2.189287803141276, \"min\": 2.189287803141276}}, \"EndTime\": 1579888099.709761, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709743}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7634132731119792, \"sum\": 1.7634132731119792, \"min\": 1.7634132731119792}}, \"EndTime\": 1579888099.709831, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709814}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.3623472534179686, \"sum\": 2.3623472534179686, \"min\": 2.3623472534179686}}, \"EndTime\": 1579888099.709904, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709886}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.4509052998860676, \"sum\": 3.4509052998860676, \"min\": 3.4509052998860676}}, \"EndTime\": 1579888099.709976, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.709958}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0277337148030599, \"sum\": 1.0277337148030599, \"min\": 1.0277337148030599}}, \"EndTime\": 1579888099.710049, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710031}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9287416097005209, \"sum\": 0.9287416097005209, \"min\": 0.9287416097005209}}, \"EndTime\": 1579888099.710121, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710102}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7794925170898437, \"sum\": 1.7794925170898437, \"min\": 1.7794925170898437}}, \"EndTime\": 1579888099.710191, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710174}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9436209248860677, \"sum\": 0.9436209248860677, \"min\": 0.9436209248860677}}, \"EndTime\": 1579888099.71026, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710243}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2754494201660156, \"sum\": 1.2754494201660156, \"min\": 1.2754494201660156}}, \"EndTime\": 1579888099.71032, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710303}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9322130045572916, \"sum\": 0.9322130045572916, \"min\": 0.9322130045572916}}, \"EndTime\": 1579888099.710378, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710361}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1732801584879557, \"sum\": 1.1732801584879557, \"min\": 1.1732801584879557}}, \"EndTime\": 1579888099.710448, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.71043}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.058207953898112, \"sum\": 1.058207953898112, \"min\": 1.058207953898112}}, \"EndTime\": 1579888099.710512, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710494}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9883454701741536, \"sum\": 0.9883454701741536, \"min\": 0.9883454701741536}}, \"EndTime\": 1579888099.710574, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710557}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9399204783121745, \"sum\": 0.9399204783121745, \"min\": 0.9399204783121745}}, \"EndTime\": 1579888099.710641, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.710624}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:19 INFO 140276589545280] #quality_metric: host=algo-1, epoch=0, train multiclass_cross_entropy_objective <loss>=0.924326205444\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:19.732] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 0, \"duration\": 3074, \"num_examples\": 1, \"num_bytes\": 412000}\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:19.931] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 2, \"duration\": 198, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9088934196136922, \"sum\": 0.9088934196136922, \"min\": 0.9088934196136922}}, \"EndTime\": 1579888099.938049, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.937981}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1515646130679926, \"sum\": 1.1515646130679926, \"min\": 1.1515646130679926}}, \"EndTime\": 1579888099.938131, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938115}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.362016009903462, \"sum\": 1.362016009903462, \"min\": 1.362016009903462}}, \"EndTime\": 1579888099.938202, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938182}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.914208236674018, \"sum\": 0.914208236674018, \"min\": 0.914208236674018}}, \"EndTime\": 1579888099.938277, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938257}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0391357340603153, \"sum\": 1.0391357340603153, \"min\": 1.0391357340603153}}, \"EndTime\": 1579888099.938354, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938334}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9740255627587695, \"sum\": 0.9740255627587695, \"min\": 0.9740255627587695}}, \"EndTime\": 1579888099.938433, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938413}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.903973693441297, \"sum\": 0.903973693441297, \"min\": 0.903973693441297}}, \"EndTime\": 1579888099.938507, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938488}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9676761672912679, \"sum\": 0.9676761672912679, \"min\": 0.9676761672912679}}, \"EndTime\": 1579888099.938575, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938556}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9065705361601198, \"sum\": 0.9065705361601198, \"min\": 0.9065705361601198}}, \"EndTime\": 1579888099.938646, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938627}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.308391188051348, \"sum\": 2.308391188051348, \"min\": 2.308391188051348}}, \"EndTime\": 1579888099.938717, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938698}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9437440183921438, \"sum\": 0.9437440183921438, \"min\": 0.9437440183921438}}, \"EndTime\": 1579888099.938787, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938768}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9629361087885424, \"sum\": 0.9629361087885424, \"min\": 0.9629361087885424}}, \"EndTime\": 1579888099.938858, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938839}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9140169349396118, \"sum\": 0.9140169349396118, \"min\": 0.9140169349396118}}, \"EndTime\": 1579888099.938925, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938906}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0627803553595205, \"sum\": 1.0627803553595205, \"min\": 1.0627803553595205}}, \"EndTime\": 1579888099.938997, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.938977}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.6373984020654753, \"sum\": 1.6373984020654753, \"min\": 1.6373984020654753}}, \"EndTime\": 1579888099.939069, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.93905}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9219607458609874, \"sum\": 0.9219607458609874, \"min\": 0.9219607458609874}}, \"EndTime\": 1579888099.939135, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939116}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9178696064752205, \"sum\": 0.9178696064752205, \"min\": 0.9178696064752205}}, \"EndTime\": 1579888099.939205, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939187}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9244955293983023, \"sum\": 0.9244955293983023, \"min\": 0.9244955293983023}}, \"EndTime\": 1579888099.939269, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.93925}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4977169763232356, \"sum\": 1.4977169763232356, \"min\": 1.4977169763232356}}, \"EndTime\": 1579888099.939332, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939315}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.3356397037023553, \"sum\": 1.3356397037023553, \"min\": 1.3356397037023553}}, \"EndTime\": 1579888099.939395, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939377}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.4217912737443825, \"sum\": 2.4217912737443825, \"min\": 2.4217912737443825}}, \"EndTime\": 1579888099.939459, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939441}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.863247839144162, \"sum\": 1.863247839144162, \"min\": 1.863247839144162}}, \"EndTime\": 1579888099.939523, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939505}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9375512540895993, \"sum\": 0.9375512540895993, \"min\": 0.9375512540895993}}, \"EndTime\": 1579888099.939586, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939568}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9119399841552409, \"sum\": 0.9119399841552409, \"min\": 0.9119399841552409}}, \"EndTime\": 1579888099.939654, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939636}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.304631820658393, \"sum\": 1.304631820658393, \"min\": 1.304631820658393}}, \"EndTime\": 1579888099.939718, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.9397}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9155571786446197, \"sum\": 0.9155571786446197, \"min\": 0.9155571786446197}}, \"EndTime\": 1579888099.939783, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939764}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1956513304526257, \"sum\": 1.1956513304526257, \"min\": 1.1956513304526257}}, \"EndTime\": 1579888099.939854, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939835}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9132411504712467, \"sum\": 0.9132411504712467, \"min\": 0.9132411504712467}}, \"EndTime\": 1579888099.939918, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.9399}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9741675609278774, \"sum\": 0.9741675609278774, \"min\": 0.9741675609278774}}, \"EndTime\": 1579888099.939982, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.939964}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0203065224557362, \"sum\": 1.0203065224557362, \"min\": 1.0203065224557362}}, \"EndTime\": 1579888099.940045, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.940028}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9578486537806362, \"sum\": 0.9578486537806362, \"min\": 0.9578486537806362}}, \"EndTime\": 1579888099.940109, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.94009}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9239628940384176, \"sum\": 0.9239628940384176, \"min\": 0.9239628940384176}}, \"EndTime\": 1579888099.940172, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888099.940154}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:19 INFO 140276589545280] #quality_metric: host=algo-1, epoch=0, validation multiclass_cross_entropy_objective <loss>=0.908893419614\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:19 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=0, criteria=multiclass_cross_entropy_objective, value=0.903973693441\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:19 INFO 140276589545280] Epoch 0: Loss improved. Updating best model\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:19 INFO 140276589545280] #progress_metric: host=algo-1, completed 6 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 73, \"sum\": 73.0, \"min\": 73}, \"Total Records Seen\": {\"count\": 1, \"max\": 72079, \"sum\": 72079.0, \"min\": 72079}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 3, \"sum\": 3.0, \"min\": 3}}, \"EndTime\": 1579888099.94208, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 0}, \"StartTime\": 1579888096.987396}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:19 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=20332.6108785 records/second\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:22.895] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 5, \"duration\": 2952, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9015669769287109, \"sum\": 0.9015669769287109, \"min\": 0.9015669769287109}}, \"EndTime\": 1579888102.895641, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.895517}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0548296559651693, \"sum\": 1.0548296559651693, \"min\": 1.0548296559651693}}, \"EndTime\": 1579888102.895806, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.895783}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2952057525634766, \"sum\": 1.2952057525634766, \"min\": 1.2952057525634766}}, \"EndTime\": 1579888102.895903, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.895881}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9013351755777995, \"sum\": 0.9013351755777995, \"min\": 0.9013351755777995}}, \"EndTime\": 1579888102.895972, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.895953}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0376922505696615, \"sum\": 1.0376922505696615, \"min\": 1.0376922505696615}}, \"EndTime\": 1579888102.896052, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896034}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9535490325927735, \"sum\": 0.9535490325927735, \"min\": 0.9535490325927735}}, \"EndTime\": 1579888102.896119, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896101}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9028322031656901, \"sum\": 0.9028322031656901, \"min\": 0.9028322031656901}}, \"EndTime\": 1579888102.896182, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896163}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9575322377522787, \"sum\": 0.9575322377522787, \"min\": 0.9575322377522787}}, \"EndTime\": 1579888102.896244, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896226}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9149442626953125, \"sum\": 0.9149442626953125, \"min\": 0.9149442626953125}}, \"EndTime\": 1579888102.896306, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.89629}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.12556283976237, \"sum\": 3.12556283976237, \"min\": 3.12556283976237}}, \"EndTime\": 1579888102.896369, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896352}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9403844848632813, \"sum\": 0.9403844848632813, \"min\": 0.9403844848632813}}, \"EndTime\": 1579888102.896432, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896414}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9545375722249348, \"sum\": 0.9545375722249348, \"min\": 0.9545375722249348}}, \"EndTime\": 1579888102.896517, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896499}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9078884755452474, \"sum\": 0.9078884755452474, \"min\": 0.9078884755452474}}, \"EndTime\": 1579888102.896582, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896564}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0882629465738933, \"sum\": 1.0882629465738933, \"min\": 1.0882629465738933}}, \"EndTime\": 1579888102.896645, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896627}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.5164452799479164, \"sum\": 2.5164452799479164, \"min\": 2.5164452799479164}}, \"EndTime\": 1579888102.896706, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896689}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9061323740641276, \"sum\": 0.9061323740641276, \"min\": 0.9061323740641276}}, \"EndTime\": 1579888102.896768, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896751}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9041985778808593, \"sum\": 0.9041985778808593, \"min\": 0.9041985778808593}}, \"EndTime\": 1579888102.896828, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896812}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9081930735270183, \"sum\": 0.9081930735270183, \"min\": 0.9081930735270183}}, \"EndTime\": 1579888102.896889, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.896871}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7874916625976562, \"sum\": 1.7874916625976562, \"min\": 1.7874916625976562}}, \"EndTime\": 1579888102.896958, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.89694}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5343989156087239, \"sum\": 1.5343989156087239, \"min\": 1.5343989156087239}}, \"EndTime\": 1579888102.897024, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897007}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.043638336181641, \"sum\": 2.043638336181641, \"min\": 2.043638336181641}}, \"EndTime\": 1579888102.897089, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897071}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.733131972249349, \"sum\": 2.733131972249349, \"min\": 2.733131972249349}}, \"EndTime\": 1579888102.897156, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897137}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9242185963948568, \"sum\": 0.9242185963948568, \"min\": 0.9242185963948568}}, \"EndTime\": 1579888102.897219, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897202}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9034348429361979, \"sum\": 0.9034348429361979, \"min\": 0.9034348429361979}}, \"EndTime\": 1579888102.897283, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897264}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5476850341796875, \"sum\": 1.5476850341796875, \"min\": 1.5476850341796875}}, \"EndTime\": 1579888102.897345, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897327}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9040938079833984, \"sum\": 0.9040938079833984, \"min\": 0.9040938079833984}}, \"EndTime\": 1579888102.897413, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897395}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1561836588541667, \"sum\": 1.1561836588541667, \"min\": 1.1561836588541667}}, \"EndTime\": 1579888102.897488, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897469}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9030899322509766, \"sum\": 0.9030899322509766, \"min\": 0.9030899322509766}}, \"EndTime\": 1579888102.897553, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897534}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9724229675292969, \"sum\": 0.9724229675292969, \"min\": 0.9724229675292969}}, \"EndTime\": 1579888102.89762, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897601}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0087704416910808, \"sum\": 1.0087704416910808, \"min\": 1.0087704416910808}}, \"EndTime\": 1579888102.897683, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897665}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9607210184733073, \"sum\": 0.9607210184733073, \"min\": 0.9607210184733073}}, \"EndTime\": 1579888102.897749, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.89773}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9113351328531901, \"sum\": 0.9113351328531901, \"min\": 0.9113351328531901}}, \"EndTime\": 1579888102.897821, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888102.897802}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:22 INFO 140276589545280] #quality_metric: host=algo-1, epoch=1, train multiclass_cross_entropy_objective <loss>=0.901566976929\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:23.128] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 5, \"duration\": 206, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9073500753877642, \"sum\": 0.9073500753877642, \"min\": 0.9073500753877642}}, \"EndTime\": 1579888103.137214, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137148}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0941928827968959, \"sum\": 1.0941928827968959, \"min\": 1.0941928827968959}}, \"EndTime\": 1579888103.137307, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137284}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2061060106706682, \"sum\": 1.2061060106706682, \"min\": 1.2061060106706682}}, \"EndTime\": 1579888103.137386, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137366}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.909703427656354, \"sum\": 0.909703427656354, \"min\": 0.909703427656354}}, \"EndTime\": 1579888103.137452, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137434}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0390269801397616, \"sum\": 1.0390269801397616, \"min\": 1.0390269801397616}}, \"EndTime\": 1579888103.137516, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137499}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9467624059847287, \"sum\": 0.9467624059847287, \"min\": 0.9467624059847287}}, \"EndTime\": 1579888103.137572, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137558}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9078655136250624, \"sum\": 0.9078655136250624, \"min\": 0.9078655136250624}}, \"EndTime\": 1579888103.137627, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137612}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.971921101454571, \"sum\": 0.971921101454571, \"min\": 0.971921101454571}}, \"EndTime\": 1579888103.137705, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137688}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9435801174605734, \"sum\": 0.9435801174605734, \"min\": 0.9435801174605734}}, \"EndTime\": 1579888103.137766, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.13775}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4955495455928554, \"sum\": 1.4955495455928554, \"min\": 1.4955495455928554}}, \"EndTime\": 1579888103.137826, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.13781}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9344952416959679, \"sum\": 0.9344952416959679, \"min\": 0.9344952416959679}}, \"EndTime\": 1579888103.13788, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137866}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9549374312440184, \"sum\": 0.9549374312440184, \"min\": 0.9549374312440184}}, \"EndTime\": 1579888103.137933, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137919}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9196252733984895, \"sum\": 0.9196252733984895, \"min\": 0.9196252733984895}}, \"EndTime\": 1579888103.137988, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.137973}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1328702761552305, \"sum\": 1.1328702761552305, \"min\": 1.1328702761552305}}, \"EndTime\": 1579888103.13805, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138036}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.8569781957390146, \"sum\": 1.8569781957390146, \"min\": 1.8569781957390146}}, \"EndTime\": 1579888103.138102, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138089}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9074849297616199, \"sum\": 0.9074849297616199, \"min\": 0.9074849297616199}}, \"EndTime\": 1579888103.138154, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.13814}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9053047230970684, \"sum\": 0.9053047230970684, \"min\": 0.9053047230970684}}, \"EndTime\": 1579888103.138207, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138193}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9087052265273906, \"sum\": 0.9087052265273906, \"min\": 0.9087052265273906}}, \"EndTime\": 1579888103.138258, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138243}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4471069092121962, \"sum\": 1.4471069092121962, \"min\": 1.4471069092121962}}, \"EndTime\": 1579888103.138317, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138301}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.369127710903691, \"sum\": 1.369127710903691, \"min\": 1.369127710903691}}, \"EndTime\": 1579888103.138372, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138357}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.6997955444173394, \"sum\": 1.6997955444173394, \"min\": 1.6997955444173394}}, \"EndTime\": 1579888103.13843, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138413}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.3888518596298685, \"sum\": 2.3888518596298685, \"min\": 2.3888518596298685}}, \"EndTime\": 1579888103.138484, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138469}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9318180170579852, \"sum\": 0.9318180170579852, \"min\": 0.9318180170579852}}, \"EndTime\": 1579888103.138546, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138529}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9099891794664405, \"sum\": 0.9099891794664405, \"min\": 0.9099891794664405}}, \"EndTime\": 1579888103.138581, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138572}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.6667555726479277, \"sum\": 1.6667555726479277, \"min\": 1.6667555726479277}}, \"EndTime\": 1579888103.138626, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138611}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.90761010059504, \"sum\": 0.90761010059504, \"min\": 0.90761010059504}}, \"EndTime\": 1579888103.138692, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138674}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1340287494278463, \"sum\": 1.1340287494278463, \"min\": 1.1340287494278463}}, \"EndTime\": 1579888103.138753, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138736}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9077251246385029, \"sum\": 0.9077251246385029, \"min\": 0.9077251246385029}}, \"EndTime\": 1579888103.138816, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138798}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9496380030077085, \"sum\": 0.9496380030077085, \"min\": 0.9496380030077085}}, \"EndTime\": 1579888103.138879, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138861}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.006610050531583, \"sum\": 1.006610050531583, \"min\": 1.006610050531583}}, \"EndTime\": 1579888103.138942, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138924}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9550327956279648, \"sum\": 0.9550327956279648, \"min\": 0.9550327956279648}}, \"EndTime\": 1579888103.139003, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.138986}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9182211925440559, \"sum\": 0.9182211925440559, \"min\": 0.9182211925440559}}, \"EndTime\": 1579888103.139074, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888103.139056}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:23 INFO 140276589545280] #quality_metric: host=algo-1, epoch=1, validation multiclass_cross_entropy_objective <loss>=0.907350075388\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:23 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=1, criteria=multiclass_cross_entropy_objective, value=0.905304723097\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:23 INFO 140276589545280] #progress_metric: host=algo-1, completed 13 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 134, \"sum\": 134.0, \"min\": 134}, \"Total Records Seen\": {\"count\": 1, \"max\": 132158, \"sum\": 132158.0, \"min\": 132158}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 4, \"sum\": 4.0, \"min\": 4}}, \"EndTime\": 1579888103.140208, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 1}, \"StartTime\": 1579888099.942344}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:23 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=18786.4460939 records/second\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:25.760] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 7, \"duration\": 2619, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8997965362548828, \"sum\": 0.8997965362548828, \"min\": 0.8997965362548828}}, \"EndTime\": 1579888105.76016, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760055}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.061617280069987, \"sum\": 1.061617280069987, \"min\": 1.061617280069987}}, \"EndTime\": 1579888105.760276, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760254}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.266932509358724, \"sum\": 1.266932509358724, \"min\": 1.266932509358724}}, \"EndTime\": 1579888105.760344, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760325}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8979438588460287, \"sum\": 0.8979438588460287, \"min\": 0.8979438588460287}}, \"EndTime\": 1579888105.76041, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760391}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0341512115478515, \"sum\": 1.0341512115478515, \"min\": 1.0341512115478515}}, \"EndTime\": 1579888105.760502, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760482}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.942428076171875, \"sum\": 0.942428076171875, \"min\": 0.942428076171875}}, \"EndTime\": 1579888105.760569, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760551}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9012813283284505, \"sum\": 0.9012813283284505, \"min\": 0.9012813283284505}}, \"EndTime\": 1579888105.760632, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760615}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9465372324625652, \"sum\": 0.9465372324625652, \"min\": 0.9465372324625652}}, \"EndTime\": 1579888105.760698, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.76068}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9212094329833984, \"sum\": 0.9212094329833984, \"min\": 0.9212094329833984}}, \"EndTime\": 1579888105.760761, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760744}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.0988808634440104, \"sum\": 3.0988808634440104, \"min\": 3.0988808634440104}}, \"EndTime\": 1579888105.760824, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760806}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9376983591715495, \"sum\": 0.9376983591715495, \"min\": 0.9376983591715495}}, \"EndTime\": 1579888105.760885, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760868}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9590290293375651, \"sum\": 0.9590290293375651, \"min\": 0.9590290293375651}}, \"EndTime\": 1579888105.760947, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760929}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9087566884358724, \"sum\": 0.9087566884358724, \"min\": 0.9087566884358724}}, \"EndTime\": 1579888105.761012, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.760995}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0811436564127603, \"sum\": 1.0811436564127603, \"min\": 1.0811436564127603}}, \"EndTime\": 1579888105.761074, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761057}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.4860513997395834, \"sum\": 2.4860513997395834, \"min\": 2.4860513997395834}}, \"EndTime\": 1579888105.761135, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761117}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8979939748128255, \"sum\": 0.8979939748128255, \"min\": 0.8979939748128255}}, \"EndTime\": 1579888105.7612, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761182}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8971697408040364, \"sum\": 0.8971697408040364, \"min\": 0.8971697408040364}}, \"EndTime\": 1579888105.761266, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761247}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.899086610921224, \"sum\": 0.899086610921224, \"min\": 0.899086610921224}}, \"EndTime\": 1579888105.761327, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.76131}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.807148203531901, \"sum\": 1.807148203531901, \"min\": 1.807148203531901}}, \"EndTime\": 1579888105.761382, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761371}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.516539638264974, \"sum\": 1.516539638264974, \"min\": 1.516539638264974}}, \"EndTime\": 1579888105.761415, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761406}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.036740002441406, \"sum\": 2.036740002441406, \"min\": 2.036740002441406}}, \"EndTime\": 1579888105.761452, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761437}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.4395551595052085, \"sum\": 2.4395551595052085, \"min\": 2.4395551595052085}}, \"EndTime\": 1579888105.761513, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761494}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9306545227050781, \"sum\": 0.9306545227050781, \"min\": 0.9306545227050781}}, \"EndTime\": 1579888105.761578, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761562}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9037675974527994, \"sum\": 0.9037675974527994, \"min\": 0.9037675974527994}}, \"EndTime\": 1579888105.761642, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761624}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.531312255859375, \"sum\": 1.531312255859375, \"min\": 1.531312255859375}}, \"EndTime\": 1579888105.761704, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761686}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8996724395751953, \"sum\": 0.8996724395751953, \"min\": 0.8996724395751953}}, \"EndTime\": 1579888105.761763, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761746}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1624081634521484, \"sum\": 1.1624081634521484, \"min\": 1.1624081634521484}}, \"EndTime\": 1579888105.76183, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761812}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9006176401774089, \"sum\": 0.9006176401774089, \"min\": 0.9006176401774089}}, \"EndTime\": 1579888105.761895, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761878}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9798316090901693, \"sum\": 0.9798316090901693, \"min\": 0.9798316090901693}}, \"EndTime\": 1579888105.761958, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.76194}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0034512908935547, \"sum\": 1.0034512908935547, \"min\": 1.0034512908935547}}, \"EndTime\": 1579888105.762005, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.761989}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9599960072835286, \"sum\": 0.9599960072835286, \"min\": 0.9599960072835286}}, \"EndTime\": 1579888105.762066, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.762048}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9075830352783203, \"sum\": 0.9075830352783203, \"min\": 0.9075830352783203}}, \"EndTime\": 1579888105.762119, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.762102}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:25 INFO 140276589545280] #quality_metric: host=algo-1, epoch=2, train multiclass_cross_entropy_objective <loss>=0.899796536255\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:25.969] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 8, \"duration\": 187, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9061923498161306, \"sum\": 0.9061923498161306, \"min\": 0.9061923498161306}}, \"EndTime\": 1579888105.976237, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976173}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.020492228623554, \"sum\": 1.020492228623554, \"min\": 1.020492228623554}}, \"EndTime\": 1579888105.976329, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976306}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.224534678046459, \"sum\": 1.224534678046459, \"min\": 1.224534678046459}}, \"EndTime\": 1579888105.9764, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.97638}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9007090480921272, \"sum\": 0.9007090480921272, \"min\": 0.9007090480921272}}, \"EndTime\": 1579888105.976488, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976449}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0245875697637525, \"sum\": 1.0245875697637525, \"min\": 1.0245875697637525}}, \"EndTime\": 1579888105.976557, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976538}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9458290475979944, \"sum\": 0.9458290475979944, \"min\": 0.9458290475979944}}, \"EndTime\": 1579888105.976613, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976596}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8991966846938457, \"sum\": 0.8991966846938457, \"min\": 0.8991966846938457}}, \"EndTime\": 1579888105.976682, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976664}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9537374971392311, \"sum\": 0.9537374971392311, \"min\": 0.9537374971392311}}, \"EndTime\": 1579888105.976745, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976727}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9205086170912742, \"sum\": 0.9205086170912742, \"min\": 0.9205086170912742}}, \"EndTime\": 1579888105.976807, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976789}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.940833836135153, \"sum\": 2.940833836135153, \"min\": 2.940833836135153}}, \"EndTime\": 1579888105.976869, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976851}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9327997647017519, \"sum\": 0.9327997647017519, \"min\": 0.9327997647017519}}, \"EndTime\": 1579888105.976932, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976915}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9746373812780875, \"sum\": 0.9746373812780875, \"min\": 0.9746373812780875}}, \"EndTime\": 1579888105.976996, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.976978}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9200697003604569, \"sum\": 0.9200697003604569, \"min\": 0.9200697003604569}}, \"EndTime\": 1579888105.977062, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977045}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0446562786076898, \"sum\": 1.0446562786076898, \"min\": 1.0446562786076898}}, \"EndTime\": 1579888105.977128, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.97711}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.121415716544607, \"sum\": 2.121415716544607, \"min\": 2.121415716544607}}, \"EndTime\": 1579888105.977192, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977175}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9028101366147221, \"sum\": 0.9028101366147221, \"min\": 0.9028101366147221}}, \"EndTime\": 1579888105.977262, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977243}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9009761358228092, \"sum\": 0.9009761358228092, \"min\": 0.9009761358228092}}, \"EndTime\": 1579888105.977333, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977314}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9033066588298935, \"sum\": 0.9033066588298935, \"min\": 0.9033066588298935}}, \"EndTime\": 1579888105.977395, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977377}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5724161160770016, \"sum\": 1.5724161160770016, \"min\": 1.5724161160770016}}, \"EndTime\": 1579888105.977461, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977442}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.830890404590754, \"sum\": 1.830890404590754, \"min\": 1.830890404590754}}, \"EndTime\": 1579888105.977534, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977516}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.5128930956641145, \"sum\": 2.5128930956641145, \"min\": 2.5128930956641145}}, \"EndTime\": 1579888105.977609, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.97759}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.3104797444553096, \"sum\": 1.3104797444553096, \"min\": 1.3104797444553096}}, \"EndTime\": 1579888105.977682, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977662}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9521662969881304, \"sum\": 0.9521662969881304, \"min\": 0.9521662969881304}}, \"EndTime\": 1579888105.977752, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977733}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9094524317511548, \"sum\": 0.9094524317511548, \"min\": 0.9094524317511548}}, \"EndTime\": 1579888105.977824, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977805}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7993932406530875, \"sum\": 1.7993932406530875, \"min\": 1.7993932406530875}}, \"EndTime\": 1579888105.977889, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.97787}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9050968215881429, \"sum\": 0.9050968215881429, \"min\": 0.9050968215881429}}, \"EndTime\": 1579888105.977955, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.977938}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1407588140942286, \"sum\": 1.1407588140942286, \"min\": 1.1407588140942286}}, \"EndTime\": 1579888105.978027, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.978008}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9063766824897532, \"sum\": 0.9063766824897532, \"min\": 0.9063766824897532}}, \"EndTime\": 1579888105.978094, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.978077}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9791066379267748, \"sum\": 0.9791066379267748, \"min\": 0.9791066379267748}}, \"EndTime\": 1579888105.978161, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.978143}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9969034085737246, \"sum\": 0.9969034085737246, \"min\": 0.9969034085737246}}, \"EndTime\": 1579888105.978221, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.978205}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.954395043262629, \"sum\": 0.954395043262629, \"min\": 0.954395043262629}}, \"EndTime\": 1579888105.978282, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.978265}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9153582942469936, \"sum\": 0.9153582942469936, \"min\": 0.9153582942469936}}, \"EndTime\": 1579888105.97835, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888105.978331}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:25 INFO 140276589545280] #quality_metric: host=algo-1, epoch=2, validation multiclass_cross_entropy_objective <loss>=0.906192349816\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:25 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=2, criteria=multiclass_cross_entropy_objective, value=0.899196684694\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:25 INFO 140276589545280] Epoch 2: Loss improved. Updating best model\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:25 INFO 140276589545280] #progress_metric: host=algo-1, completed 20 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 195, \"sum\": 195.0, \"min\": 195}, \"Total Records Seen\": {\"count\": 1, \"max\": 192237, \"sum\": 192237.0, \"min\": 192237}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 5, \"sum\": 5.0, \"min\": 5}}, \"EndTime\": 1579888105.980106, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 2}, \"StartTime\": 1579888103.140501}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:25 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=21156.5140463 records/second\u001b[0m\n",
      "\n",
      "2020-01-24 17:48:14 Training - Training image download completed. Training in progress.\u001b[34m[2020-01-24 17:48:28.537] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 9, \"duration\": 2555, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.898541298421224, \"sum\": 0.898541298421224, \"min\": 0.898541298421224}}, \"EndTime\": 1579888108.537495, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.537398}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.050540625, \"sum\": 1.050540625, \"min\": 1.050540625}}, \"EndTime\": 1579888108.537596, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.537578}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2347003743489584, \"sum\": 1.2347003743489584, \"min\": 1.2347003743489584}}, \"EndTime\": 1579888108.537708, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.537687}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8953809509277344, \"sum\": 0.8953809509277344, \"min\": 0.8953809509277344}}, \"EndTime\": 1579888108.537764, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.537751}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.034381440226237, \"sum\": 1.034381440226237, \"min\": 1.034381440226237}}, \"EndTime\": 1579888108.537812, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.5378}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9346695821126302, \"sum\": 0.9346695821126302, \"min\": 0.9346695821126302}}, \"EndTime\": 1579888108.537858, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.537846}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8980591339111328, \"sum\": 0.8980591339111328, \"min\": 0.8980591339111328}}, \"EndTime\": 1579888108.537903, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.537891}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9383661214192708, \"sum\": 0.9383661214192708, \"min\": 0.9383661214192708}}, \"EndTime\": 1579888108.537957, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.537943}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9301388783772786, \"sum\": 0.9301388783772786, \"min\": 0.9301388783772786}}, \"EndTime\": 1579888108.53801, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.537997}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.9529402852376303, \"sum\": 2.9529402852376303, \"min\": 2.9529402852376303}}, \"EndTime\": 1579888108.538062, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.53805}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9364473999023437, \"sum\": 0.9364473999023437, \"min\": 0.9364473999023437}}, \"EndTime\": 1579888108.538115, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538103}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9587863576253255, \"sum\": 0.9587863576253255, \"min\": 0.9587863576253255}}, \"EndTime\": 1579888108.538168, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538156}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9089447438557943, \"sum\": 0.9089447438557943, \"min\": 0.9089447438557943}}, \"EndTime\": 1579888108.538221, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538208}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0749898752848308, \"sum\": 1.0749898752848308, \"min\": 1.0749898752848308}}, \"EndTime\": 1579888108.538279, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538266}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.4452654561360676, \"sum\": 2.4452654561360676, \"min\": 2.4452654561360676}}, \"EndTime\": 1579888108.538331, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538319}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8947385243733724, \"sum\": 0.8947385243733724, \"min\": 0.8947385243733724}}, \"EndTime\": 1579888108.538384, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538371}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8941554392496744, \"sum\": 0.8941554392496744, \"min\": 0.8941554392496744}}, \"EndTime\": 1579888108.538436, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538423}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.895662240600586, \"sum\": 0.895662240600586, \"min\": 0.895662240600586}}, \"EndTime\": 1579888108.538488, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538475}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7808914794921875, \"sum\": 1.7808914794921875, \"min\": 1.7808914794921875}}, \"EndTime\": 1579888108.53854, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538528}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5302054239908853, \"sum\": 1.5302054239908853, \"min\": 1.5302054239908853}}, \"EndTime\": 1579888108.538593, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.53858}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.0202433451334634, \"sum\": 2.0202433451334634, \"min\": 2.0202433451334634}}, \"EndTime\": 1579888108.538646, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538633}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.3146925048828124, \"sum\": 2.3146925048828124, \"min\": 2.3146925048828124}}, \"EndTime\": 1579888108.538698, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538685}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.930800493367513, \"sum\": 0.930800493367513, \"min\": 0.930800493367513}}, \"EndTime\": 1579888108.53875, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538737}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9040727345784505, \"sum\": 0.9040727345784505, \"min\": 0.9040727345784505}}, \"EndTime\": 1579888108.538802, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538789}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.533273979695638, \"sum\": 1.533273979695638, \"min\": 1.533273979695638}}, \"EndTime\": 1579888108.538849, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538837}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.898063818359375, \"sum\": 0.898063818359375, \"min\": 0.898063818359375}}, \"EndTime\": 1579888108.538901, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538888}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1563670166015625, \"sum\": 1.1563670166015625, \"min\": 1.1563670166015625}}, \"EndTime\": 1579888108.538952, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.53894}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8998885467529297, \"sum\": 0.8998885467529297, \"min\": 0.8998885467529297}}, \"EndTime\": 1579888108.539005, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.538992}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9813591959635417, \"sum\": 0.9813591959635417, \"min\": 0.9813591959635417}}, \"EndTime\": 1579888108.539057, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.539045}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0038550384521485, \"sum\": 1.0038550384521485, \"min\": 1.0038550384521485}}, \"EndTime\": 1579888108.53911, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.539097}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.959771474202474, \"sum\": 0.959771474202474, \"min\": 0.959771474202474}}, \"EndTime\": 1579888108.539161, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.539151}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9052552693684895, \"sum\": 0.9052552693684895, \"min\": 0.9052552693684895}}, \"EndTime\": 1579888108.5392, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.53919}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:28 INFO 140276589545280] #quality_metric: host=algo-1, epoch=3, train multiclass_cross_entropy_objective <loss>=0.898541298421\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:28.751] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 11, \"duration\": 188, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9051079477038428, \"sum\": 0.9051079477038428, \"min\": 0.9051079477038428}}, \"EndTime\": 1579888108.758471, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.758407}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2007305363681757, \"sum\": 1.2007305363681757, \"min\": 1.2007305363681757}}, \"EndTime\": 1579888108.758559, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.758537}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1540739056908498, \"sum\": 1.1540739056908498, \"min\": 1.1540739056908498}}, \"EndTime\": 1579888108.758631, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.758611}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8976262322437271, \"sum\": 0.8976262322437271, \"min\": 0.8976262322437271}}, \"EndTime\": 1579888108.758697, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.758679}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9986426910928657, \"sum\": 0.9986426910928657, \"min\": 0.9986426910928657}}, \"EndTime\": 1579888108.758764, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.758746}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9389280235402276, \"sum\": 0.9389280235402276, \"min\": 0.9389280235402276}}, \"EndTime\": 1579888108.758828, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.75881}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8957202408507406, \"sum\": 0.8957202408507406, \"min\": 0.8957202408507406}}, \"EndTime\": 1579888108.75889, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.758872}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9433754546347058, \"sum\": 0.9433754546347058, \"min\": 0.9433754546347058}}, \"EndTime\": 1579888108.758951, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.758934}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9299148620524197, \"sum\": 0.9299148620524197, \"min\": 0.9299148620524197}}, \"EndTime\": 1579888108.759019, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759002}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.7358370317441327, \"sum\": 2.7358370317441327, \"min\": 2.7358370317441327}}, \"EndTime\": 1579888108.759081, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759063}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.931833751287346, \"sum\": 0.931833751287346, \"min\": 0.931833751287346}}, \"EndTime\": 1579888108.759144, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759126}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.975627304869549, \"sum\": 0.975627304869549, \"min\": 0.975627304869549}}, \"EndTime\": 1579888108.759205, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759188}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9199102653168172, \"sum\": 0.9199102653168172, \"min\": 0.9199102653168172}}, \"EndTime\": 1579888108.759265, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759248}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1139334890718626, \"sum\": 1.1139334890718626, \"min\": 1.1139334890718626}}, \"EndTime\": 1579888108.759326, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759308}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.9565853276360685, \"sum\": 2.9565853276360685, \"min\": 2.9565853276360685}}, \"EndTime\": 1579888108.759385, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759369}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9002264558713382, \"sum\": 0.9002264558713382, \"min\": 0.9002264558713382}}, \"EndTime\": 1579888108.759445, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759429}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8984844101094375, \"sum\": 0.8984844101094375, \"min\": 0.8984844101094375}}, \"EndTime\": 1579888108.759507, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.75949}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.900625147102041, \"sum\": 0.900625147102041, \"min\": 0.900625147102041}}, \"EndTime\": 1579888108.759562, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759545}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4827945958123543, \"sum\": 1.4827945958123543, \"min\": 1.4827945958123543}}, \"EndTime\": 1579888108.75962, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.7596}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.636718571201939, \"sum\": 1.636718571201939, \"min\": 1.636718571201939}}, \"EndTime\": 1579888108.759681, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759665}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4962934267981234, \"sum\": 1.4962934267981234, \"min\": 1.4962934267981234}}, \"EndTime\": 1579888108.759742, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759725}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4404770120958514, \"sum\": 1.4404770120958514, \"min\": 1.4404770120958514}}, \"EndTime\": 1579888108.75979, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759775}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9570106800958098, \"sum\": 0.9570106800958098, \"min\": 0.9570106800958098}}, \"EndTime\": 1579888108.759846, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759829}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9101477611556986, \"sum\": 0.9101477611556986, \"min\": 0.9101477611556986}}, \"EndTime\": 1579888108.759907, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.75989}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2600034345482065, \"sum\": 1.2600034345482065, \"min\": 1.2600034345482065}}, \"EndTime\": 1579888108.759969, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759952}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9039077413383717, \"sum\": 0.9039077413383717, \"min\": 0.9039077413383717}}, \"EndTime\": 1579888108.760016, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.759999}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1453332161935128, \"sum\": 1.1453332161935128, \"min\": 1.1453332161935128}}, \"EndTime\": 1579888108.760065, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.760049}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.905855693385382, \"sum\": 0.905855693385382, \"min\": 0.905855693385382}}, \"EndTime\": 1579888108.760124, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.760107}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9878623261432673, \"sum\": 0.9878623261432673, \"min\": 0.9878623261432673}}, \"EndTime\": 1579888108.760176, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.76016}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9703347969309151, \"sum\": 0.9703347969309151, \"min\": 0.9703347969309151}}, \"EndTime\": 1579888108.760221, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.760206}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9541760684647033, \"sum\": 0.9541760684647033, \"min\": 0.9541760684647033}}, \"EndTime\": 1579888108.760273, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.760257}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9131405521804261, \"sum\": 0.9131405521804261, \"min\": 0.9131405521804261}}, \"EndTime\": 1579888108.760333, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888108.760316}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:28 INFO 140276589545280] #quality_metric: host=algo-1, epoch=3, validation multiclass_cross_entropy_objective <loss>=0.905107947704\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:28 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=3, criteria=multiclass_cross_entropy_objective, value=0.895720240851\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:28 INFO 140276589545280] Epoch 3: Loss improved. Updating best model\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:28 INFO 140276589545280] #progress_metric: host=algo-1, completed 26 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 256, \"sum\": 256.0, \"min\": 256}, \"Total Records Seen\": {\"count\": 1, \"max\": 252316, \"sum\": 252316.0, \"min\": 252316}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 6, \"sum\": 6.0, \"min\": 6}}, \"EndTime\": 1579888108.761886, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 3}, \"StartTime\": 1579888105.980932}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:28 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=21602.6807991 records/second\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:31.352] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 11, \"duration\": 2589, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8977575612386067, \"sum\": 0.8977575612386067, \"min\": 0.8977575612386067}}, \"EndTime\": 1579888111.352195, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352126}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0540142567952473, \"sum\": 1.0540142567952473, \"min\": 1.0540142567952473}}, \"EndTime\": 1579888111.352264, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352251}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.220074383544922, \"sum\": 1.220074383544922, \"min\": 1.220074383544922}}, \"EndTime\": 1579888111.352315, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.3523}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8940900390625, \"sum\": 0.8940900390625, \"min\": 0.8940900390625}}, \"EndTime\": 1579888111.352373, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352357}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0268894134521485, \"sum\": 1.0268894134521485, \"min\": 1.0268894134521485}}, \"EndTime\": 1579888111.352488, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352468}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9292251597086588, \"sum\": 0.9292251597086588, \"min\": 0.9292251597086588}}, \"EndTime\": 1579888111.352572, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352551}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8972577728271485, \"sum\": 0.8972577728271485, \"min\": 0.8972577728271485}}, \"EndTime\": 1579888111.352636, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352618}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9325777720133464, \"sum\": 0.9325777720133464, \"min\": 0.9325777720133464}}, \"EndTime\": 1579888111.352698, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.35268}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9290279469807943, \"sum\": 0.9290279469807943, \"min\": 0.9290279469807943}}, \"EndTime\": 1579888111.352762, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352744}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.9922413655598956, \"sum\": 2.9922413655598956, \"min\": 2.9922413655598956}}, \"EndTime\": 1579888111.35281, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352799}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9355743255615234, \"sum\": 0.9355743255615234, \"min\": 0.9355743255615234}}, \"EndTime\": 1579888111.352866, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352849}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9598941864013671, \"sum\": 0.9598941864013671, \"min\": 0.9598941864013671}}, \"EndTime\": 1579888111.352927, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.35291}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9082936197916667, \"sum\": 0.9082936197916667, \"min\": 0.9082936197916667}}, \"EndTime\": 1579888111.352989, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.352972}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0777692220052084, \"sum\": 1.0777692220052084, \"min\": 1.0777692220052084}}, \"EndTime\": 1579888111.353049, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353033}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.406058215332031, \"sum\": 2.406058215332031, \"min\": 2.406058215332031}}, \"EndTime\": 1579888111.353107, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.35309}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8927818868001302, \"sum\": 0.8927818868001302, \"min\": 0.8927818868001302}}, \"EndTime\": 1579888111.353166, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.35315}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8921749308268229, \"sum\": 0.8921749308268229, \"min\": 0.8921749308268229}}, \"EndTime\": 1579888111.353228, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353211}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8937582397460937, \"sum\": 0.8937582397460937, \"min\": 0.8937582397460937}}, \"EndTime\": 1579888111.353279, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353268}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7594539591471354, \"sum\": 1.7594539591471354, \"min\": 1.7594539591471354}}, \"EndTime\": 1579888111.353339, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353322}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5040996795654298, \"sum\": 1.5040996795654298, \"min\": 1.5040996795654298}}, \"EndTime\": 1579888111.353398, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353381}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.9064004302978517, \"sum\": 1.9064004302978517, \"min\": 1.9064004302978517}}, \"EndTime\": 1579888111.353459, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353442}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.1281579833984376, \"sum\": 2.1281579833984376, \"min\": 2.1281579833984376}}, \"EndTime\": 1579888111.353518, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353502}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9316430653889974, \"sum\": 0.9316430653889974, \"min\": 0.9316430653889974}}, \"EndTime\": 1579888111.353578, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353562}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9043907236735026, \"sum\": 0.9043907236735026, \"min\": 0.9043907236735026}}, \"EndTime\": 1579888111.35364, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353622}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5452479136149089, \"sum\": 1.5452479136149089, \"min\": 1.5452479136149089}}, \"EndTime\": 1579888111.353702, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353684}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8972686024983724, \"sum\": 0.8972686024983724, \"min\": 0.8972686024983724}}, \"EndTime\": 1579888111.353764, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353747}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1559841278076173, \"sum\": 1.1559841278076173, \"min\": 1.1559841278076173}}, \"EndTime\": 1579888111.353827, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.35381}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8995642537434896, \"sum\": 0.8995642537434896, \"min\": 0.8995642537434896}}, \"EndTime\": 1579888111.353888, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353872}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9827730967203776, \"sum\": 0.9827730967203776, \"min\": 0.9827730967203776}}, \"EndTime\": 1579888111.353926, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353916}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.997791904703776, \"sum\": 0.997791904703776, \"min\": 0.997791904703776}}, \"EndTime\": 1579888111.35397, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.353955}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9596819763183594, \"sum\": 0.9596819763183594, \"min\": 0.9596819763183594}}, \"EndTime\": 1579888111.354024, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.354013}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9036598419189453, \"sum\": 0.9036598419189453, \"min\": 0.9036598419189453}}, \"EndTime\": 1579888111.354064, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.354051}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:31 INFO 140276589545280] #quality_metric: host=algo-1, epoch=4, train multiclass_cross_entropy_objective <loss>=0.897757561239\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:31.563] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 14, \"duration\": 191, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9036006516051515, \"sum\": 0.9036006516051515, \"min\": 0.9036006516051515}}, \"EndTime\": 1579888111.57054, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.570474}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.002187854345248, \"sum\": 1.002187854345248, \"min\": 1.002187854345248}}, \"EndTime\": 1579888111.570624, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.570609}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2395073820843043, \"sum\": 1.2395073820843043, \"min\": 1.2395073820843043}}, \"EndTime\": 1579888111.570697, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.570676}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8970855956706163, \"sum\": 0.8970855956706163, \"min\": 0.8970855956706163}}, \"EndTime\": 1579888111.570778, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.570758}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0807254755703333, \"sum\": 1.0807254755703333, \"min\": 1.0807254755703333}}, \"EndTime\": 1579888111.57085, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.570831}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9321231440761276, \"sum\": 0.9321231440761276, \"min\": 0.9321231440761276}}, \"EndTime\": 1579888111.570924, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.570905}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.894122964627892, \"sum\": 0.894122964627892, \"min\": 0.894122964627892}}, \"EndTime\": 1579888111.570991, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.570973}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9362764686147637, \"sum\": 0.9362764686147637, \"min\": 0.9362764686147637}}, \"EndTime\": 1579888111.571059, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.57104}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9276808452987163, \"sum\": 0.9276808452987163, \"min\": 0.9276808452987163}}, \"EndTime\": 1579888111.571134, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571114}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.0153065606535034, \"sum\": 3.0153065606535034, \"min\": 3.0153065606535034}}, \"EndTime\": 1579888111.571201, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571182}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.929063571975647, \"sum\": 0.929063571975647, \"min\": 0.929063571975647}}, \"EndTime\": 1579888111.571266, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571248}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9724038155831287, \"sum\": 0.9724038155831287, \"min\": 0.9724038155831287}}, \"EndTime\": 1579888111.571329, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.57131}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9190185709418692, \"sum\": 0.9190185709418692, \"min\": 0.9190185709418692}}, \"EndTime\": 1579888111.571391, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571373}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0205315682605485, \"sum\": 1.0205315682605485, \"min\": 1.0205315682605485}}, \"EndTime\": 1579888111.571452, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571441}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7099697229865387, \"sum\": 1.7099697229865387, \"min\": 1.7099697229865387}}, \"EndTime\": 1579888111.571511, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571493}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8983384946373585, \"sum\": 0.8983384946373585, \"min\": 0.8983384946373585}}, \"EndTime\": 1579888111.571574, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571556}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8967130577199152, \"sum\": 0.8967130577199152, \"min\": 0.8967130577199152}}, \"EndTime\": 1579888111.571636, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571618}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8989142000119632, \"sum\": 0.8989142000119632, \"min\": 0.8989142000119632}}, \"EndTime\": 1579888111.571699, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571682}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.8795434456531916, \"sum\": 1.8795434456531916, \"min\": 1.8795434456531916}}, \"EndTime\": 1579888111.571749, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571734}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2955568031686917, \"sum\": 1.2955568031686917, \"min\": 1.2955568031686917}}, \"EndTime\": 1579888111.571817, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.5718}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.9697045215753994, \"sum\": 1.9697045215753994, \"min\": 1.9697045215753994}}, \"EndTime\": 1579888111.571874, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571862}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.6099248202916945, \"sum\": 1.6099248202916945, \"min\": 1.6099248202916945}}, \"EndTime\": 1579888111.57193, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571913}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.955234630193596, \"sum\": 0.955234630193596, \"min\": 0.955234630193596}}, \"EndTime\": 1579888111.571992, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.571974}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9110490415448672, \"sum\": 0.9110490415448672, \"min\": 0.9110490415448672}}, \"EndTime\": 1579888111.572054, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.572037}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.3600315546068784, \"sum\": 1.3600315546068784, \"min\": 1.3600315546068784}}, \"EndTime\": 1579888111.572116, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.5721}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9030861520576731, \"sum\": 0.9030861520576731, \"min\": 0.9030861520576731}}, \"EndTime\": 1579888111.572187, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.572169}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1027308965649967, \"sum\": 1.1027308965649967, \"min\": 1.1027308965649967}}, \"EndTime\": 1579888111.572313, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.572293}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9055756549860603, \"sum\": 0.9055756549860603, \"min\": 0.9055756549860603}}, \"EndTime\": 1579888111.572365, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.572354}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9765303610485498, \"sum\": 0.9765303610485498, \"min\": 0.9765303610485498}}, \"EndTime\": 1579888111.572424, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.572407}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9751142560245194, \"sum\": 0.9751142560245194, \"min\": 0.9751142560245194}}, \"EndTime\": 1579888111.572519, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.5725}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9540649779785807, \"sum\": 0.9540649779785807, \"min\": 0.9540649779785807}}, \"EndTime\": 1579888111.572586, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.572569}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9116946732156603, \"sum\": 0.9116946732156603, \"min\": 0.9116946732156603}}, \"EndTime\": 1579888111.572659, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888111.572641}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:31 INFO 140276589545280] #quality_metric: host=algo-1, epoch=4, validation multiclass_cross_entropy_objective <loss>=0.903600651605\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:31 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=4, criteria=multiclass_cross_entropy_objective, value=0.894122964628\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:31 INFO 140276589545280] Epoch 4: Loss improved. Updating best model\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:31 INFO 140276589545280] #progress_metric: host=algo-1, completed 33 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 317, \"sum\": 317.0, \"min\": 317}, \"Total Records Seen\": {\"count\": 1, \"max\": 312395, \"sum\": 312395.0, \"min\": 312395}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 7, \"sum\": 7.0, \"min\": 7}}, \"EndTime\": 1579888111.574435, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 4}, \"StartTime\": 1579888108.762592}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:31 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=21365.3125939 records/second\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:34.205] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 13, \"duration\": 2630, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8973208618164062, \"sum\": 0.8973208618164062, \"min\": 0.8973208618164062}}, \"EndTime\": 1579888114.205985, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.205894}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0585223154703776, \"sum\": 1.0585223154703776, \"min\": 1.0585223154703776}}, \"EndTime\": 1579888114.206082, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206062}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1896060546875, \"sum\": 1.1896060546875, \"min\": 1.1896060546875}}, \"EndTime\": 1579888114.206149, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.20613}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8937510091145834, \"sum\": 0.8937510091145834, \"min\": 0.8937510091145834}}, \"EndTime\": 1579888114.206213, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206195}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0302564534505207, \"sum\": 1.0302564534505207, \"min\": 1.0302564534505207}}, \"EndTime\": 1579888114.206278, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.20626}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9252955190022787, \"sum\": 0.9252955190022787, \"min\": 0.9252955190022787}}, \"EndTime\": 1579888114.206341, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206324}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8966277465820313, \"sum\": 0.8966277465820313, \"min\": 0.8966277465820313}}, \"EndTime\": 1579888114.206402, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206385}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.928422748819987, \"sum\": 0.928422748819987, \"min\": 0.928422748819987}}, \"EndTime\": 1579888114.206459, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206442}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9308502583821615, \"sum\": 0.9308502583821615, \"min\": 0.9308502583821615}}, \"EndTime\": 1579888114.206513, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206497}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.964362727864583, \"sum\": 2.964362727864583, \"min\": 2.964362727864583}}, \"EndTime\": 1579888114.206574, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206556}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9342426890055339, \"sum\": 0.9342426890055339, \"min\": 0.9342426890055339}}, \"EndTime\": 1579888114.206634, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206617}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9605365956624349, \"sum\": 0.9605365956624349, \"min\": 0.9605365956624349}}, \"EndTime\": 1579888114.206695, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206678}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9075406555175781, \"sum\": 0.9075406555175781, \"min\": 0.9075406555175781}}, \"EndTime\": 1579888114.206755, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206739}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.07169191792806, \"sum\": 1.07169191792806, \"min\": 1.07169191792806}}, \"EndTime\": 1579888114.206814, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206797}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.3742361022949217, \"sum\": 2.3742361022949217, \"min\": 2.3742361022949217}}, \"EndTime\": 1579888114.206874, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206857}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8912732065836588, \"sum\": 0.8912732065836588, \"min\": 0.8912732065836588}}, \"EndTime\": 1579888114.206935, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206918}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8906666717529297, \"sum\": 0.8906666717529297, \"min\": 0.8906666717529297}}, \"EndTime\": 1579888114.206996, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.206979}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8924564595540364, \"sum\": 0.8924564595540364, \"min\": 0.8924564595540364}}, \"EndTime\": 1579888114.207056, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207039}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7341093119303386, \"sum\": 1.7341093119303386, \"min\": 1.7341093119303386}}, \"EndTime\": 1579888114.207116, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.2071}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5244920369466146, \"sum\": 1.5244920369466146, \"min\": 1.5244920369466146}}, \"EndTime\": 1579888114.207179, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207161}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.8833234405517578, \"sum\": 1.8833234405517578, \"min\": 1.8833234405517578}}, \"EndTime\": 1579888114.207239, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207222}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.979422332763672, \"sum\": 1.979422332763672, \"min\": 1.979422332763672}}, \"EndTime\": 1579888114.207301, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207283}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9328943501790364, \"sum\": 0.9328943501790364, \"min\": 0.9328943501790364}}, \"EndTime\": 1579888114.207363, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207345}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9045658864339192, \"sum\": 0.9045658864339192, \"min\": 0.9045658864339192}}, \"EndTime\": 1579888114.207423, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207406}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.519097324625651, \"sum\": 1.519097324625651, \"min\": 1.519097324625651}}, \"EndTime\": 1579888114.207478, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207463}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8967052724202474, \"sum\": 0.8967052724202474, \"min\": 0.8967052724202474}}, \"EndTime\": 1579888114.207538, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207521}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1611283182779948, \"sum\": 1.1611283182779948, \"min\": 1.1611283182779948}}, \"EndTime\": 1579888114.207599, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207582}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8994117228190104, \"sum\": 0.8994117228190104, \"min\": 0.8994117228190104}}, \"EndTime\": 1579888114.207661, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207644}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9766993733723959, \"sum\": 0.9766993733723959, \"min\": 0.9766993733723959}}, \"EndTime\": 1579888114.207722, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207706}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0008544901529948, \"sum\": 1.0008544901529948, \"min\": 1.0008544901529948}}, \"EndTime\": 1579888114.207791, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207772}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9596407897949218, \"sum\": 0.9596407897949218, \"min\": 0.9596407897949218}}, \"EndTime\": 1579888114.207851, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207834}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9024145345052084, \"sum\": 0.9024145345052084, \"min\": 0.9024145345052084}}, \"EndTime\": 1579888114.207908, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.207891}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:34 INFO 140276589545280] #quality_metric: host=algo-1, epoch=5, train multiclass_cross_entropy_objective <loss>=0.897320861816\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:34.414] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 17, \"duration\": 187, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9012235598303824, \"sum\": 0.9012235598303824, \"min\": 0.9012235598303824}}, \"EndTime\": 1579888114.421314, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421245}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.004301540003953, \"sum\": 1.004301540003953, \"min\": 1.004301540003953}}, \"EndTime\": 1579888114.421408, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421384}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1167152780667444, \"sum\": 1.1167152780667444, \"min\": 1.1167152780667444}}, \"EndTime\": 1579888114.421482, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421462}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8972508132060898, \"sum\": 0.8972508132060898, \"min\": 0.8972508132060898}}, \"EndTime\": 1579888114.421541, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421523}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9936176284810357, \"sum\": 0.9936176284810357, \"min\": 0.9936176284810357}}, \"EndTime\": 1579888114.421606, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421588}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9283397933614556, \"sum\": 0.9283397933614556, \"min\": 0.9283397933614556}}, \"EndTime\": 1579888114.421669, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421651}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8930183908434587, \"sum\": 0.8930183908434587, \"min\": 0.8930183908434587}}, \"EndTime\": 1579888114.421741, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421723}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9323494414673664, \"sum\": 0.9323494414673664, \"min\": 0.9323494414673664}}, \"EndTime\": 1579888114.421808, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.42179}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9338453554440433, \"sum\": 0.9338453554440433, \"min\": 0.9338453554440433}}, \"EndTime\": 1579888114.421881, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421863}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.525128311990263, \"sum\": 3.525128311990263, \"min\": 3.525128311990263}}, \"EndTime\": 1579888114.421953, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.421934}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9301242589633093, \"sum\": 0.9301242589633093, \"min\": 0.9301242589633093}}, \"EndTime\": 1579888114.422018, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422001}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9713689039613849, \"sum\": 0.9713689039613849, \"min\": 0.9713689039613849}}, \"EndTime\": 1579888114.422092, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422073}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9174782581557923, \"sum\": 0.9174782581557923, \"min\": 0.9174782581557923}}, \"EndTime\": 1579888114.422155, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422139}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0408730741823402, \"sum\": 1.0408730741823402, \"min\": 1.0408730741823402}}, \"EndTime\": 1579888114.422218, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422202}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5168164127517476, \"sum\": 1.5168164127517476, \"min\": 1.5168164127517476}}, \"EndTime\": 1579888114.422281, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422264}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.896956007712373, \"sum\": 0.896956007712373, \"min\": 0.896956007712373}}, \"EndTime\": 1579888114.422353, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422334}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8953244754064893, \"sum\": 0.8953244754064893, \"min\": 0.8953244754064893}}, \"EndTime\": 1579888114.422472, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422449}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8976742679682299, \"sum\": 0.8976742679682299, \"min\": 0.8976742679682299}}, \"EndTime\": 1579888114.422543, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422524}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.470487234468625, \"sum\": 1.470487234468625, \"min\": 1.470487234468625}}, \"EndTime\": 1579888114.422613, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422594}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2924641412361644, \"sum\": 1.2924641412361644, \"min\": 1.2924641412361644}}, \"EndTime\": 1579888114.422679, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422662}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4806476455871338, \"sum\": 1.4806476455871338, \"min\": 1.4806476455871338}}, \"EndTime\": 1579888114.422736, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.42272}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.091542672920481, \"sum\": 2.091542672920481, \"min\": 2.091542672920481}}, \"EndTime\": 1579888114.4228, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422782}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.953351301454831, \"sum\": 0.953351301454831, \"min\": 0.953351301454831}}, \"EndTime\": 1579888114.422864, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422846}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9119866748306945, \"sum\": 0.9119866748306945, \"min\": 0.9119866748306945}}, \"EndTime\": 1579888114.422925, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.422907}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.3685473486523176, \"sum\": 1.3685473486523176, \"min\": 1.3685473486523176}}, \"EndTime\": 1579888114.422988, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.42297}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9028223802183026, \"sum\": 0.9028223802183026, \"min\": 0.9028223802183026}}, \"EndTime\": 1579888114.42305, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.423032}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.066990497045606, \"sum\": 1.066990497045606, \"min\": 1.066990497045606}}, \"EndTime\": 1579888114.423113, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.423095}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9054617931299933, \"sum\": 0.9054617931299933, \"min\": 0.9054617931299933}}, \"EndTime\": 1579888114.423168, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.423151}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9841415506863245, \"sum\": 0.9841415506863245, \"min\": 0.9841415506863245}}, \"EndTime\": 1579888114.42323, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.423212}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9698196309543006, \"sum\": 0.9698196309543006, \"min\": 0.9698196309543006}}, \"EndTime\": 1579888114.423294, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.423275}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.953999643541684, \"sum\": 0.953999643541684, \"min\": 0.953999643541684}}, \"EndTime\": 1579888114.423357, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.423339}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9105315755750145, \"sum\": 0.9105315755750145, \"min\": 0.9105315755750145}}, \"EndTime\": 1579888114.42342, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888114.423401}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:34 INFO 140276589545280] #quality_metric: host=algo-1, epoch=5, validation multiclass_cross_entropy_objective <loss>=0.90122355983\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:34 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=5, criteria=multiclass_cross_entropy_objective, value=0.893018390843\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:34 INFO 140276589545280] Epoch 5: Loss improved. Updating best model\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:34 INFO 140276589545280] #progress_metric: host=algo-1, completed 40 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 378, \"sum\": 378.0, \"min\": 378}, \"Total Records Seen\": {\"count\": 1, \"max\": 372474, \"sum\": 372474.0, \"min\": 372474}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}}, \"EndTime\": 1579888114.425344, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 5}, \"StartTime\": 1579888111.575221}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:34 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=21078.487738 records/second\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:36.979] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 15, \"duration\": 2553, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8970994008382162, \"sum\": 0.8970994008382162, \"min\": 0.8970994008382162}}, \"EndTime\": 1579888116.979858, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.979762}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0510479014078775, \"sum\": 1.0510479014078775, \"min\": 1.0510479014078775}}, \"EndTime\": 1579888116.979948, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.979932}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1728559438069661, \"sum\": 1.1728559438069661, \"min\": 1.1728559438069661}}, \"EndTime\": 1579888116.980001, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.979988}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.893452968343099, \"sum\": 0.893452968343099, \"min\": 0.893452968343099}}, \"EndTime\": 1579888116.980048, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980035}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0231776346842447, \"sum\": 1.0231776346842447, \"min\": 1.0231776346842447}}, \"EndTime\": 1579888116.980093, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980082}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9226353200276692, \"sum\": 0.9226353200276692, \"min\": 0.9226353200276692}}, \"EndTime\": 1579888116.980139, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980127}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8961428680419922, \"sum\": 0.8961428680419922, \"min\": 0.8961428680419922}}, \"EndTime\": 1579888116.980183, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980171}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9257180541992187, \"sum\": 0.9257180541992187, \"min\": 0.9257180541992187}}, \"EndTime\": 1579888116.980227, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980215}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9313262125651042, \"sum\": 0.9313262125651042, \"min\": 0.9313262125651042}}, \"EndTime\": 1579888116.980272, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.98026}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.9896907491048177, \"sum\": 2.9896907491048177, \"min\": 2.9896907491048177}}, \"EndTime\": 1579888116.980319, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980307}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9339546122233073, \"sum\": 0.9339546122233073, \"min\": 0.9339546122233073}}, \"EndTime\": 1579888116.980366, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980354}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.960893642171224, \"sum\": 0.960893642171224, \"min\": 0.960893642171224}}, \"EndTime\": 1579888116.980411, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980399}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9067071879069011, \"sum\": 0.9067071879069011, \"min\": 0.9067071879069011}}, \"EndTime\": 1579888116.980479, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980445}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0687877888997397, \"sum\": 1.0687877888997397, \"min\": 1.0687877888997397}}, \"EndTime\": 1579888116.980529, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980516}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.3894626607259113, \"sum\": 2.3894626607259113, \"min\": 2.3894626607259113}}, \"EndTime\": 1579888116.980582, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980569}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8901780212402344, \"sum\": 0.8901780212402344, \"min\": 0.8901780212402344}}, \"EndTime\": 1579888116.980628, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980616}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.889443667602539, \"sum\": 0.889443667602539, \"min\": 0.889443667602539}}, \"EndTime\": 1579888116.980673, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980661}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8914711405436198, \"sum\": 0.8914711405436198, \"min\": 0.8914711405436198}}, \"EndTime\": 1579888116.980719, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980707}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.777418983968099, \"sum\": 1.777418983968099, \"min\": 1.777418983968099}}, \"EndTime\": 1579888116.980771, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980758}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5397405598958334, \"sum\": 1.5397405598958334, \"min\": 1.5397405598958334}}, \"EndTime\": 1579888116.980818, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980806}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.903484297688802, \"sum\": 1.903484297688802, \"min\": 1.903484297688802}}, \"EndTime\": 1579888116.980862, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980851}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.8884189270019531, \"sum\": 1.8884189270019531, \"min\": 1.8884189270019531}}, \"EndTime\": 1579888116.980906, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980894}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9336929178873697, \"sum\": 0.9336929178873697, \"min\": 0.9336929178873697}}, \"EndTime\": 1579888116.98095, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980939}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9046371663411459, \"sum\": 0.9046371663411459, \"min\": 0.9046371663411459}}, \"EndTime\": 1579888116.980993, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.980982}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5129201405843098, \"sum\": 1.5129201405843098, \"min\": 1.5129201405843098}}, \"EndTime\": 1579888116.981039, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.981028}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8964593221028646, \"sum\": 0.8964593221028646, \"min\": 0.8964593221028646}}, \"EndTime\": 1579888116.981084, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.981072}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1593520324707032, \"sum\": 1.1593520324707032, \"min\": 1.1593520324707032}}, \"EndTime\": 1579888116.981144, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.981132}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8993507588704427, \"sum\": 0.8993507588704427, \"min\": 0.8993507588704427}}, \"EndTime\": 1579888116.981189, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.981177}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9785551239013672, \"sum\": 0.9785551239013672, \"min\": 0.9785551239013672}}, \"EndTime\": 1579888116.981232, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.981221}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9946411834716797, \"sum\": 0.9946411834716797, \"min\": 0.9946411834716797}}, \"EndTime\": 1579888116.981275, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.981264}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.959620166015625, \"sum\": 0.959620166015625, \"min\": 0.959620166015625}}, \"EndTime\": 1579888116.981318, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.981307}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.901256010945638, \"sum\": 0.901256010945638, \"min\": 0.901256010945638}}, \"EndTime\": 1579888116.981363, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888116.981352}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:36 INFO 140276589545280] #quality_metric: host=algo-1, epoch=6, train multiclass_cross_entropy_objective <loss>=0.897099400838\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:37.198] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 20, \"duration\": 193, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8999985208524052, \"sum\": 0.8999985208524052, \"min\": 0.8999985208524052}}, \"EndTime\": 1579888117.204719, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.204654}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0298894524098077, \"sum\": 1.0298894524098077, \"min\": 1.0298894524098077}}, \"EndTime\": 1579888117.204804, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.204785}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.109144326119861, \"sum\": 1.109144326119861, \"min\": 1.109144326119861}}, \"EndTime\": 1579888117.204869, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.204856}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8972023629951731, \"sum\": 0.8972023629951731, \"min\": 0.8972023629951731}}, \"EndTime\": 1579888117.204929, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.204911}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0989351071943139, \"sum\": 1.0989351071943139, \"min\": 1.0989351071943139}}, \"EndTime\": 1579888117.204996, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.204978}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.927118639177711, \"sum\": 0.927118639177711, \"min\": 0.927118639177711}}, \"EndTime\": 1579888117.205067, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205049}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8921722777832681, \"sum\": 0.8921722777832681, \"min\": 0.8921722777832681}}, \"EndTime\": 1579888117.205126, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205112}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9295967599840838, \"sum\": 0.9295967599840838, \"min\": 0.9295967599840838}}, \"EndTime\": 1579888117.205171, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205156}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9352699168036368, \"sum\": 0.9352699168036368, \"min\": 0.9352699168036368}}, \"EndTime\": 1579888117.205237, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.20522}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.8587998391467626, \"sum\": 3.8587998391467626, \"min\": 3.8587998391467626}}, \"EndTime\": 1579888117.205299, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205282}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9298162386674538, \"sum\": 0.9298162386674538, \"min\": 0.9298162386674538}}, \"EndTime\": 1579888117.20537, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205352}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9706891015429947, \"sum\": 0.9706891015429947, \"min\": 0.9706891015429947}}, \"EndTime\": 1579888117.205432, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205415}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.915934369408497, \"sum\": 0.915934369408497, \"min\": 0.915934369408497}}, \"EndTime\": 1579888117.205494, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205477}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.109283268467564, \"sum\": 1.109283268467564, \"min\": 1.109283268467564}}, \"EndTime\": 1579888117.205557, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205539}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.022709921927014, \"sum\": 2.022709921927014, \"min\": 2.022709921927014}}, \"EndTime\": 1579888117.205618, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.2056}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8958464072642727, \"sum\": 0.8958464072642727, \"min\": 0.8958464072642727}}, \"EndTime\": 1579888117.205677, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.20566}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8941752915058567, \"sum\": 0.8941752915058567, \"min\": 0.8941752915058567}}, \"EndTime\": 1579888117.205737, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.20572}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8967077913043026, \"sum\": 0.8967077913043026, \"min\": 0.8967077913043026}}, \"EndTime\": 1579888117.205798, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.20578}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2312550469816286, \"sum\": 1.2312550469816286, \"min\": 1.2312550469816286}}, \"EndTime\": 1579888117.205867, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.20585}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.6913655653139563, \"sum\": 1.6913655653139563, \"min\": 1.6913655653139563}}, \"EndTime\": 1579888117.205931, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205914}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.4641395294555175, \"sum\": 2.4641395294555175, \"min\": 2.4641395294555175}}, \"EndTime\": 1579888117.205989, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.205976}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.120329924683755, \"sum\": 2.120329924683755, \"min\": 2.120329924683755}}, \"EndTime\": 1579888117.206039, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206022}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9539250684959117, \"sum\": 0.9539250684959117, \"min\": 0.9539250684959117}}, \"EndTime\": 1579888117.206108, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.20609}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9127786770959033, \"sum\": 0.9127786770959033, \"min\": 0.9127786770959033}}, \"EndTime\": 1579888117.206183, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206164}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0533034811007198, \"sum\": 1.0533034811007198, \"min\": 1.0533034811007198}}, \"EndTime\": 1579888117.206254, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206236}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9025312604027963, \"sum\": 0.9025312604027963, \"min\": 0.9025312604027963}}, \"EndTime\": 1579888117.206327, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206308}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0891866102358314, \"sum\": 1.0891866102358314, \"min\": 1.0891866102358314}}, \"EndTime\": 1579888117.206395, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206379}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9054056139665342, \"sum\": 0.9054056139665342, \"min\": 0.9054056139665342}}, \"EndTime\": 1579888117.206456, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206439}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9891236855091649, \"sum\": 0.9891236855091649, \"min\": 0.9891236855091649}}, \"EndTime\": 1579888117.206517, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.2065}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9875093015627601, \"sum\": 0.9875093015627601, \"min\": 0.9875093015627601}}, \"EndTime\": 1579888117.206579, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206563}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9539594993134154, \"sum\": 0.9539594993134154, \"min\": 0.9539594993134154}}, \"EndTime\": 1579888117.206643, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206626}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9092573386851386, \"sum\": 0.9092573386851386, \"min\": 0.9092573386851386}}, \"EndTime\": 1579888117.206713, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888117.206696}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:37 INFO 140276589545280] #quality_metric: host=algo-1, epoch=6, validation multiclass_cross_entropy_objective <loss>=0.899998520852\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:37 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=6, criteria=multiclass_cross_entropy_objective, value=0.892172277783\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:37 INFO 140276589545280] #progress_metric: host=algo-1, completed 46 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 439, \"sum\": 439.0, \"min\": 439}, \"Total Records Seen\": {\"count\": 1, \"max\": 432553, \"sum\": 432553.0, \"min\": 432553}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 9, \"sum\": 9.0, \"min\": 9}}, \"EndTime\": 1579888117.207687, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 6}, \"StartTime\": 1579888114.426166}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:37 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=21598.2610638 records/second\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:39.690] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 17, \"duration\": 2482, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8969505920410156, \"sum\": 0.8969505920410156, \"min\": 0.8969505920410156}}, \"EndTime\": 1579888119.690755, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.690621}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0535011006673176, \"sum\": 1.0535011006673176, \"min\": 1.0535011006673176}}, \"EndTime\": 1579888119.690878, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.690858}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1470117299397786, \"sum\": 1.1470117299397786, \"min\": 1.1470117299397786}}, \"EndTime\": 1579888119.690963, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.690947}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8931239166259766, \"sum\": 0.8931239166259766, \"min\": 0.8931239166259766}}, \"EndTime\": 1579888119.691014, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691001}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0280403238932292, \"sum\": 1.0280403238932292, \"min\": 1.0280403238932292}}, \"EndTime\": 1579888119.691089, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691051}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9210873504638671, \"sum\": 0.9210873504638671, \"min\": 0.9210873504638671}}, \"EndTime\": 1579888119.691151, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691137}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8956805033365886, \"sum\": 0.8956805033365886, \"min\": 0.8956805033365886}}, \"EndTime\": 1579888119.691199, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691187}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9237013397216797, \"sum\": 0.9237013397216797, \"min\": 0.9237013397216797}}, \"EndTime\": 1579888119.691275, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691259}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9303130818684896, \"sum\": 0.9303130818684896, \"min\": 0.9303130818684896}}, \"EndTime\": 1579888119.691331, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691318}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.8870770751953123, \"sum\": 2.8870770751953123, \"min\": 2.8870770751953123}}, \"EndTime\": 1579888119.691384, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691372}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.933703022257487, \"sum\": 0.933703022257487, \"min\": 0.933703022257487}}, \"EndTime\": 1579888119.691465, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691449}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9613868011474609, \"sum\": 0.9613868011474609, \"min\": 0.9613868011474609}}, \"EndTime\": 1579888119.691512, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.6915}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9060152364095052, \"sum\": 0.9060152364095052, \"min\": 0.9060152364095052}}, \"EndTime\": 1579888119.691621, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691605}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0618822692871093, \"sum\": 1.0618822692871093, \"min\": 1.0618822692871093}}, \"EndTime\": 1579888119.691673, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.69166}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.2878060770670574, \"sum\": 2.2878060770670574, \"min\": 2.2878060770670574}}, \"EndTime\": 1579888119.691758, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691723}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8892079528808594, \"sum\": 0.8892079528808594, \"min\": 0.8892079528808594}}, \"EndTime\": 1579888119.691811, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691798}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8883934265136719, \"sum\": 0.8883934265136719, \"min\": 0.8883934265136719}}, \"EndTime\": 1579888119.691856, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691845}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.890681948852539, \"sum\": 0.890681948852539, \"min\": 0.890681948852539}}, \"EndTime\": 1579888119.691901, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691889}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7266198079427084, \"sum\": 1.7266198079427084, \"min\": 1.7266198079427084}}, \"EndTime\": 1579888119.691954, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691941}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5199883728027344, \"sum\": 1.5199883728027344, \"min\": 1.5199883728027344}}, \"EndTime\": 1579888119.692003, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.691991}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.8959397115071615, \"sum\": 1.8959397115071615, \"min\": 1.8959397115071615}}, \"EndTime\": 1579888119.69205, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692038}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7369619801839193, \"sum\": 1.7369619801839193, \"min\": 1.7369619801839193}}, \"EndTime\": 1579888119.692109, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692084}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9312787089029948, \"sum\": 0.9312787089029948, \"min\": 0.9312787089029948}}, \"EndTime\": 1579888119.692157, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692145}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9046386464436849, \"sum\": 0.9046386464436849, \"min\": 0.9046386464436849}}, \"EndTime\": 1579888119.692202, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.69219}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5288019022623698, \"sum\": 1.5288019022623698, \"min\": 1.5288019022623698}}, \"EndTime\": 1579888119.692247, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692235}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8962170928955078, \"sum\": 0.8962170928955078, \"min\": 0.8962170928955078}}, \"EndTime\": 1579888119.692292, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.69228}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1544090260823567, \"sum\": 1.1544090260823567, \"min\": 1.1544090260823567}}, \"EndTime\": 1579888119.692341, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692326}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8993243001302084, \"sum\": 0.8993243001302084, \"min\": 0.8993243001302084}}, \"EndTime\": 1579888119.6924, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692384}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9812665120442708, \"sum\": 0.9812665120442708, \"min\": 0.9812665120442708}}, \"EndTime\": 1579888119.692485, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692444}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9983911387125651, \"sum\": 0.9983911387125651, \"min\": 0.9983911387125651}}, \"EndTime\": 1579888119.692557, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692534}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9596092732747395, \"sum\": 0.9596092732747395, \"min\": 0.9596092732747395}}, \"EndTime\": 1579888119.692625, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692606}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9003222615559896, \"sum\": 0.9003222615559896, \"min\": 0.9003222615559896}}, \"EndTime\": 1579888119.692701, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.692681}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:39 INFO 140276589545280] #quality_metric: host=algo-1, epoch=7, train multiclass_cross_entropy_objective <loss>=0.896950592041\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:39.944] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 23, \"duration\": 230, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9007713795343189, \"sum\": 0.9007713795343189, \"min\": 0.9007713795343189}}, \"EndTime\": 1579888119.951822, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.951759}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0310306554151756, \"sum\": 1.0310306554151756, \"min\": 1.0310306554151756}}, \"EndTime\": 1579888119.951908, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.951889}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0906794115008114, \"sum\": 1.0906794115008114, \"min\": 1.0906794115008114}}, \"EndTime\": 1579888119.951973, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.951954}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8971301732780771, \"sum\": 0.8971301732780771, \"min\": 0.8971301732780771}}, \"EndTime\": 1579888119.952038, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.95202}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9814842701593188, \"sum\": 0.9814842701593188, \"min\": 0.9814842701593188}}, \"EndTime\": 1579888119.952103, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952085}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.92639312540961, \"sum\": 0.92639312540961, \"min\": 0.92639312540961}}, \"EndTime\": 1579888119.952176, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952157}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8912863212006388, \"sum\": 0.8912863212006388, \"min\": 0.8912863212006388}}, \"EndTime\": 1579888119.952248, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952231}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9283519069300828, \"sum\": 0.9283519069300828, \"min\": 0.9283519069300828}}, \"EndTime\": 1579888119.952317, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.9523}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9383183993607481, \"sum\": 0.9383183993607481, \"min\": 0.9383183993607481}}, \"EndTime\": 1579888119.952382, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952364}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.381268627507074, \"sum\": 2.381268627507074, \"min\": 2.381268627507074}}, \"EndTime\": 1579888119.952445, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952427}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9295983244671168, \"sum\": 0.9295983244671168, \"min\": 0.9295983244671168}}, \"EndTime\": 1579888119.952531, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952513}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.970362352150258, \"sum\": 0.970362352150258, \"min\": 0.970362352150258}}, \"EndTime\": 1579888119.952589, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952573}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9146364011396263, \"sum\": 0.9146364011396263, \"min\": 0.9146364011396263}}, \"EndTime\": 1579888119.952651, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952633}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1489994216695447, \"sum\": 1.1489994216695447, \"min\": 1.1489994216695447}}, \"EndTime\": 1579888119.952714, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952697}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.8809207352436335, \"sum\": 1.8809207352436335, \"min\": 1.8809207352436335}}, \"EndTime\": 1579888119.952776, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952758}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8948586333131345, \"sum\": 0.8948586333131345, \"min\": 0.8948586333131345}}, \"EndTime\": 1579888119.952846, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952827}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8931415054991781, \"sum\": 0.8931415054991781, \"min\": 0.8931415054991781}}, \"EndTime\": 1579888119.952918, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952899}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.895919848567795, \"sum\": 0.895919848567795, \"min\": 0.895919848567795}}, \"EndTime\": 1579888119.952978, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.952962}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2675557752423852, \"sum\": 1.2675557752423852, \"min\": 1.2675557752423852}}, \"EndTime\": 1579888119.953039, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953021}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5182158001570822, \"sum\": 1.5182158001570822, \"min\": 1.5182158001570822}}, \"EndTime\": 1579888119.953108, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953089}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.3041213603216546, \"sum\": 2.3041213603216546, \"min\": 2.3041213603216546}}, \"EndTime\": 1579888119.95318, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953161}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1970394942160452, \"sum\": 1.1970394942160452, \"min\": 1.1970394942160452}}, \"EndTime\": 1579888119.953236, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953222}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9491746031333222, \"sum\": 0.9491746031333222, \"min\": 0.9491746031333222}}, \"EndTime\": 1579888119.953298, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.95328}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.913295326283705, \"sum\": 0.913295326283705, \"min\": 0.913295326283705}}, \"EndTime\": 1579888119.953363, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953345}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.66664527049871, \"sum\": 1.66664527049871, \"min\": 1.66664527049871}}, \"EndTime\": 1579888119.953435, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953417}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9023993927692764, \"sum\": 0.9023993927692764, \"min\": 0.9023993927692764}}, \"EndTime\": 1579888119.953497, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.95348}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.105033059761464, \"sum\": 1.105033059761464, \"min\": 1.105033059761464}}, \"EndTime\": 1579888119.953554, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.95354}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9053776727059234, \"sum\": 0.9053776727059234, \"min\": 0.9053776727059234}}, \"EndTime\": 1579888119.95362, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953603}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9961277622674976, \"sum\": 0.9961277622674976, \"min\": 0.9961277622674976}}, \"EndTime\": 1579888119.953692, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953675}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9684379608113661, \"sum\": 0.9684379608113661, \"min\": 0.9684379608113661}}, \"EndTime\": 1579888119.953756, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953738}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9539347682907166, \"sum\": 0.9539347682907166, \"min\": 0.9539347682907166}}, \"EndTime\": 1579888119.953802, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953791}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9084200476838175, \"sum\": 0.9084200476838175, \"min\": 0.9084200476838175}}, \"EndTime\": 1579888119.953838, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888119.953825}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:39 INFO 140276589545280] #quality_metric: host=algo-1, epoch=7, validation multiclass_cross_entropy_objective <loss>=0.900771379534\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:39 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=7, criteria=multiclass_cross_entropy_objective, value=0.891286321201\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:39 INFO 140276589545280] #progress_metric: host=algo-1, completed 53 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 500, \"sum\": 500.0, \"min\": 500}, \"Total Records Seen\": {\"count\": 1, \"max\": 492632, \"sum\": 492632.0, \"min\": 492632}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 10, \"sum\": 10.0, \"min\": 10}}, \"EndTime\": 1579888119.954848, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 7}, \"StartTime\": 1579888117.208398}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:39 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=21874.0488858 records/second\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:42.509] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 19, \"duration\": 2553, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8964414031982422, \"sum\": 0.8964414031982422, \"min\": 0.8964414031982422}}, \"EndTime\": 1579888122.509153, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509057}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0533740010579427, \"sum\": 1.0533740010579427, \"min\": 1.0533740010579427}}, \"EndTime\": 1579888122.509243, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509227}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1314473876953124, \"sum\": 1.1314473876953124, \"min\": 1.1314473876953124}}, \"EndTime\": 1579888122.509296, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509283}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8928306671142578, \"sum\": 0.8928306671142578, \"min\": 0.8928306671142578}}, \"EndTime\": 1579888122.509343, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509331}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0199511647542319, \"sum\": 1.0199511647542319, \"min\": 1.0199511647542319}}, \"EndTime\": 1579888122.509389, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509377}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9198939442952474, \"sum\": 0.9198939442952474, \"min\": 0.9198939442952474}}, \"EndTime\": 1579888122.509434, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509422}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8952477925618489, \"sum\": 0.8952477925618489, \"min\": 0.8952477925618489}}, \"EndTime\": 1579888122.509479, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509467}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9223466532389323, \"sum\": 0.9223466532389323, \"min\": 0.9223466532389323}}, \"EndTime\": 1579888122.509523, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509512}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9285824228922526, \"sum\": 0.9285824228922526, \"min\": 0.9285824228922526}}, \"EndTime\": 1579888122.509568, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509556}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 3.008125891113281, \"sum\": 3.008125891113281, \"min\": 3.008125891113281}}, \"EndTime\": 1579888122.509615, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509601}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9335202067057292, \"sum\": 0.9335202067057292, \"min\": 0.9335202067057292}}, \"EndTime\": 1579888122.509662, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.50965}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.961828271484375, \"sum\": 0.961828271484375, \"min\": 0.961828271484375}}, \"EndTime\": 1579888122.509706, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509695}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9054424794514974, \"sum\": 0.9054424794514974, \"min\": 0.9054424794514974}}, \"EndTime\": 1579888122.50975, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509739}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0646300333658854, \"sum\": 1.0646300333658854, \"min\": 1.0646300333658854}}, \"EndTime\": 1579888122.509795, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509783}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.2843345052083333, \"sum\": 2.2843345052083333, \"min\": 2.2843345052083333}}, \"EndTime\": 1579888122.509839, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509828}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8884547790527344, \"sum\": 0.8884547790527344, \"min\": 0.8884547790527344}}, \"EndTime\": 1579888122.509883, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509872}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8874913686116537, \"sum\": 0.8874913686116537, \"min\": 0.8874913686116537}}, \"EndTime\": 1579888122.50993, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509919}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8900276397705078, \"sum\": 0.8900276397705078, \"min\": 0.8900276397705078}}, \"EndTime\": 1579888122.509978, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.509966}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7189918924967449, \"sum\": 1.7189918924967449, \"min\": 1.7189918924967449}}, \"EndTime\": 1579888122.510022, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510011}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.505040771484375, \"sum\": 1.505040771484375, \"min\": 1.505040771484375}}, \"EndTime\": 1579888122.51007, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510058}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.855529065958659, \"sum\": 1.855529065958659, \"min\": 1.855529065958659}}, \"EndTime\": 1579888122.510116, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510104}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.647180561319987, \"sum\": 1.647180561319987, \"min\": 1.647180561319987}}, \"EndTime\": 1579888122.51016, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510149}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9316526102701823, \"sum\": 0.9316526102701823, \"min\": 0.9316526102701823}}, \"EndTime\": 1579888122.510204, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510193}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9045942250569662, \"sum\": 0.9045942250569662, \"min\": 0.9045942250569662}}, \"EndTime\": 1579888122.51025, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510238}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.5321817270914713, \"sum\": 1.5321817270914713, \"min\": 1.5321817270914713}}, \"EndTime\": 1579888122.510294, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510283}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8960892669677735, \"sum\": 0.8960892669677735, \"min\": 0.8960892669677735}}, \"EndTime\": 1579888122.510339, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510327}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1508296885172526, \"sum\": 1.1508296885172526, \"min\": 1.1508296885172526}}, \"EndTime\": 1579888122.510385, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510373}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8992943939208984, \"sum\": 0.8992943939208984, \"min\": 0.8992943939208984}}, \"EndTime\": 1579888122.510429, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510418}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9835898244222006, \"sum\": 0.9835898244222006, \"min\": 0.9835898244222006}}, \"EndTime\": 1579888122.510475, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510464}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9920156280517578, \"sum\": 0.9920156280517578, \"min\": 0.9920156280517578}}, \"EndTime\": 1579888122.51052, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510508}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9596033233642578, \"sum\": 0.9596033233642578, \"min\": 0.9596033233642578}}, \"EndTime\": 1579888122.510564, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510553}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8995603454589843, \"sum\": 0.8995603454589843, \"min\": 0.8995603454589843}}, \"EndTime\": 1579888122.510608, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.510597}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:42 INFO 140276589545280] #quality_metric: host=algo-1, epoch=8, train multiclass_cross_entropy_objective <loss>=0.896441403198\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:42.721] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 26, \"duration\": 190, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.90274923962061, \"sum\": 0.90274923962061, \"min\": 0.90274923962061}}, \"EndTime\": 1579888122.727464, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727398}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.029420497604756, \"sum\": 1.029420497604756, \"min\": 1.029420497604756}}, \"EndTime\": 1579888122.727555, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727532}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.074189353973348, \"sum\": 1.074189353973348, \"min\": 1.074189353973348}}, \"EndTime\": 1579888122.727628, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727611}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8968012417362787, \"sum\": 0.8968012417362787, \"min\": 0.8968012417362787}}, \"EndTime\": 1579888122.72769, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727672}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1067774380253412, \"sum\": 1.1067774380253412, \"min\": 1.1067774380253412}}, \"EndTime\": 1579888122.727759, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727741}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9257040295556446, \"sum\": 0.9257040295556446, \"min\": 0.9257040295556446}}, \"EndTime\": 1579888122.727817, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727804}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8905384536113308, \"sum\": 0.8905384536113308, \"min\": 0.8905384536113308}}, \"EndTime\": 1579888122.727868, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727851}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9276477879754078, \"sum\": 0.9276477879754078, \"min\": 0.9276477879754078}}, \"EndTime\": 1579888122.727916, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727905}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9341701258673332, \"sum\": 0.9341701258673332, \"min\": 0.9341701258673332}}, \"EndTime\": 1579888122.727964, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.727949}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.4384314891342793, \"sum\": 2.4384314891342793, \"min\": 2.4384314891342793}}, \"EndTime\": 1579888122.728029, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.72801}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9294378247622961, \"sum\": 0.9294378247622961, \"min\": 0.9294378247622961}}, \"EndTime\": 1579888122.728066, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728057}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9705196416171666, \"sum\": 0.9705196416171666, \"min\": 0.9705196416171666}}, \"EndTime\": 1579888122.728107, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728093}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.91359661320713, \"sum\": 0.91359661320713, \"min\": 0.91359661320713}}, \"EndTime\": 1579888122.728167, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.72815}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1487753551904751, \"sum\": 1.1487753551904751, \"min\": 1.1487753551904751}}, \"EndTime\": 1579888122.728234, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728216}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7140743575623127, \"sum\": 1.7140743575623127, \"min\": 1.7140743575623127}}, \"EndTime\": 1579888122.728276, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728266}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8941318070047228, \"sum\": 0.8941318070047228, \"min\": 0.8941318070047228}}, \"EndTime\": 1579888122.72833, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728314}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8922949535710198, \"sum\": 0.8922949535710198, \"min\": 0.8922949535710198}}, \"EndTime\": 1579888122.728391, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728375}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8952585680030792, \"sum\": 0.8952585680030792, \"min\": 0.8952585680030792}}, \"EndTime\": 1579888122.728431, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728421}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7320855395930426, \"sum\": 1.7320855395930426, \"min\": 1.7320855395930426}}, \"EndTime\": 1579888122.728506, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728488}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.3237491043842584, \"sum\": 1.3237491043842584, \"min\": 1.3237491043842584}}, \"EndTime\": 1579888122.728564, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.72855}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.0281343950253827, \"sum\": 2.0281343950253827, \"min\": 2.0281343950253827}}, \"EndTime\": 1579888122.728615, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728605}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.739429499274405, \"sum\": 1.739429499274405, \"min\": 1.739429499274405}}, \"EndTime\": 1579888122.72867, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728651}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9508941203395472, \"sum\": 0.9508941203395472, \"min\": 0.9508941203395472}}, \"EndTime\": 1579888122.728732, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728718}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9135285195910661, \"sum\": 0.9135285195910661, \"min\": 0.9135285195910661}}, \"EndTime\": 1579888122.728792, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728774}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.3004273111112268, \"sum\": 1.3004273111112268, \"min\": 1.3004273111112268}}, \"EndTime\": 1579888122.728854, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728836}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9022864858573985, \"sum\": 0.9022864858573985, \"min\": 0.9022864858573985}}, \"EndTime\": 1579888122.728916, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.728899}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1807880995911701, \"sum\": 1.1807880995911701, \"min\": 1.1807880995911701}}, \"EndTime\": 1579888122.729035, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.729014}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9053205020259446, \"sum\": 0.9053205020259446, \"min\": 0.9053205020259446}}, \"EndTime\": 1579888122.729078, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.729068}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9961844818887317, \"sum\": 0.9961844818887317, \"min\": 0.9961844818887317}}, \"EndTime\": 1579888122.729125, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.72911}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9887291838421167, \"sum\": 0.9887291838421167, \"min\": 0.9887291838421167}}, \"EndTime\": 1579888122.729186, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.729169}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9539195704555384, \"sum\": 0.9539195704555384, \"min\": 0.9539195704555384}}, \"EndTime\": 1579888122.729253, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.729235}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9076902793345851, \"sum\": 0.9076902793345851, \"min\": 0.9076902793345851}}, \"EndTime\": 1579888122.729326, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888122.729306}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:42 INFO 140276589545280] #quality_metric: host=algo-1, epoch=8, validation multiclass_cross_entropy_objective <loss>=0.902749239621\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:42 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=8, criteria=multiclass_cross_entropy_objective, value=0.890538453611\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:42 INFO 140276589545280] #progress_metric: host=algo-1, completed 60 % of epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 561, \"sum\": 561.0, \"min\": 561}, \"Total Records Seen\": {\"count\": 1, \"max\": 552711, \"sum\": 552711.0, \"min\": 552711}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 11, \"sum\": 11.0, \"min\": 11}}, \"EndTime\": 1579888122.730302, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 8}, \"StartTime\": 1579888119.955593}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:42 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=21651.2277884 records/second\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:45.332] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/train\", \"epoch\": 21, \"duration\": 2601, \"num_examples\": 61, \"num_bytes\": 24752548}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8963564676920573, \"sum\": 0.8963564676920573, \"min\": 0.8963564676920573}}, \"EndTime\": 1579888125.333136, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333038}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0492856028238933, \"sum\": 1.0492856028238933, \"min\": 1.0492856028238933}}, \"EndTime\": 1579888125.333244, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333221}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1091911987304688, \"sum\": 1.1091911987304688, \"min\": 1.1091911987304688}}, \"EndTime\": 1579888125.333317, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333298}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8925723276774089, \"sum\": 0.8925723276774089, \"min\": 0.8925723276774089}}, \"EndTime\": 1579888125.333371, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333354}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0259234842936198, \"sum\": 1.0259234842936198, \"min\": 1.0259234842936198}}, \"EndTime\": 1579888125.333435, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333417}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9189919545491536, \"sum\": 0.9189919545491536, \"min\": 0.9189919545491536}}, \"EndTime\": 1579888125.33351, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.33349}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8950826426188151, \"sum\": 0.8950826426188151, \"min\": 0.8950826426188151}}, \"EndTime\": 1579888125.333585, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333565}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9213744344075521, \"sum\": 0.9213744344075521, \"min\": 0.9213744344075521}}, \"EndTime\": 1579888125.333659, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.33364}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.929374863688151, \"sum\": 0.929374863688151, \"min\": 0.929374863688151}}, \"EndTime\": 1579888125.333732, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333713}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.888092962646484, \"sum\": 2.888092962646484, \"min\": 2.888092962646484}}, \"EndTime\": 1579888125.333793, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333775}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9331543792724609, \"sum\": 0.9331543792724609, \"min\": 0.9331543792724609}}, \"EndTime\": 1579888125.333851, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333834}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9621699178059896, \"sum\": 0.9621699178059896, \"min\": 0.9621699178059896}}, \"EndTime\": 1579888125.333909, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333892}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9049572692871094, \"sum\": 0.9049572692871094, \"min\": 0.9049572692871094}}, \"EndTime\": 1579888125.333974, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.333954}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0624990743001301, \"sum\": 1.0624990743001301, \"min\": 1.0624990743001301}}, \"EndTime\": 1579888125.334036, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334018}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.275364577229818, \"sum\": 2.275364577229818, \"min\": 2.275364577229818}}, \"EndTime\": 1579888125.334097, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.33408}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8878055155436197, \"sum\": 0.8878055155436197, \"min\": 0.8878055155436197}}, \"EndTime\": 1579888125.334169, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.33415}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8866959940592448, \"sum\": 0.8866959940592448, \"min\": 0.8866959940592448}}, \"EndTime\": 1579888125.334232, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334214}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8894726369222005, \"sum\": 0.8894726369222005, \"min\": 0.8894726369222005}}, \"EndTime\": 1579888125.334295, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334277}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7024817036946616, \"sum\": 1.7024817036946616, \"min\": 1.7024817036946616}}, \"EndTime\": 1579888125.334366, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334347}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4848847971598307, \"sum\": 1.4848847971598307, \"min\": 1.4848847971598307}}, \"EndTime\": 1579888125.334427, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.33441}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.7980990498860676, \"sum\": 1.7980990498860676, \"min\": 1.7980990498860676}}, \"EndTime\": 1579888125.334491, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334472}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.573785312906901, \"sum\": 1.573785312906901, \"min\": 1.573785312906901}}, \"EndTime\": 1579888125.334563, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334535}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9320350880940755, \"sum\": 0.9320350880940755, \"min\": 0.9320350880940755}}, \"EndTime\": 1579888125.334634, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334615}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9045122029622396, \"sum\": 0.9045122029622396, \"min\": 0.9045122029622396}}, \"EndTime\": 1579888125.334696, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334679}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4859294291178387, \"sum\": 1.4859294291178387, \"min\": 1.4859294291178387}}, \"EndTime\": 1579888125.334759, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334741}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8960080556233724, \"sum\": 0.8960080556233724, \"min\": 0.8960080556233724}}, \"EndTime\": 1579888125.33483, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334812}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.1335443725585939, \"sum\": 1.1335443725585939, \"min\": 1.1335443725585939}}, \"EndTime\": 1579888125.334901, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334883}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.899276870727539, \"sum\": 0.899276870727539, \"min\": 0.899276870727539}}, \"EndTime\": 1579888125.334972, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.334955}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9787449015299479, \"sum\": 0.9787449015299479, \"min\": 0.9787449015299479}}, \"EndTime\": 1579888125.335047, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.335028}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9962810740152995, \"sum\": 0.9962810740152995, \"min\": 0.9962810740152995}}, \"EndTime\": 1579888125.335118, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.335101}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9596000040690105, \"sum\": 0.9596000040690105, \"min\": 0.9596000040690105}}, \"EndTime\": 1579888125.335188, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.33517}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"train_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8988157511393229, \"sum\": 0.8988157511393229, \"min\": 0.8988157511393229}}, \"EndTime\": 1579888125.335236, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.335225}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, epoch=9, train multiclass_cross_entropy_objective <loss>=0.896356467692\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:45.543] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 29, \"duration\": 185, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.90343017578125, \"sum\": 0.90343017578125, \"min\": 0.90343017578125}}, \"EndTime\": 1579888125.550419, \"Dimensions\": {\"model\": 0, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.550353}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0318534759643392, \"sum\": 1.0318534759643392, \"min\": 1.0318534759643392}}, \"EndTime\": 1579888125.550502, \"Dimensions\": {\"model\": 1, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.550487}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0906681390957889, \"sum\": 1.0906681390957889, \"min\": 1.0906681390957889}}, \"EndTime\": 1579888125.550573, \"Dimensions\": {\"model\": 2, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.550554}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8964231041553969, \"sum\": 0.8964231041553969, \"min\": 0.8964231041553969}}, \"EndTime\": 1579888125.550652, \"Dimensions\": {\"model\": 3, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.55063}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9773358544401736, \"sum\": 0.9773358544401736, \"min\": 0.9773358544401736}}, \"EndTime\": 1579888125.550722, \"Dimensions\": {\"model\": 4, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.550703}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9253167692100637, \"sum\": 0.9253167692100637, \"min\": 0.9253167692100637}}, \"EndTime\": 1579888125.550788, \"Dimensions\": {\"model\": 5, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.550775}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8902079210141686, \"sum\": 0.8902079210141686, \"min\": 0.8902079210141686}}, \"EndTime\": 1579888125.55085, \"Dimensions\": {\"model\": 6, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.550832}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9271750052664156, \"sum\": 0.9271750052664156, \"min\": 0.9271750052664156}}, \"EndTime\": 1579888125.550913, \"Dimensions\": {\"model\": 7, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.550901}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9361957737989972, \"sum\": 0.9361957737989972, \"min\": 0.9361957737989972}}, \"EndTime\": 1579888125.55097, \"Dimensions\": {\"model\": 8, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.550952}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.2513808980603986, \"sum\": 2.2513808980603986, \"min\": 2.2513808980603986}}, \"EndTime\": 1579888125.551033, \"Dimensions\": {\"model\": 9, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551022}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9288241329269308, \"sum\": 0.9288241329269308, \"min\": 0.9288241329269308}}, \"EndTime\": 1579888125.551092, \"Dimensions\": {\"model\": 10, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551074}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9708807324283768, \"sum\": 0.9708807324283768, \"min\": 0.9708807324283768}}, \"EndTime\": 1579888125.551157, \"Dimensions\": {\"model\": 11, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551138}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9126264589921771, \"sum\": 0.9126264589921771, \"min\": 0.9126264589921771}}, \"EndTime\": 1579888125.551226, \"Dimensions\": {\"model\": 12, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551207}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.0334452663375915, \"sum\": 1.0334452663375915, \"min\": 1.0334452663375915}}, \"EndTime\": 1579888125.551292, \"Dimensions\": {\"model\": 13, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551273}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.4436272307496254, \"sum\": 2.4436272307496254, \"min\": 2.4436272307496254}}, \"EndTime\": 1579888125.551354, \"Dimensions\": {\"model\": 14, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551337}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8933699525307403, \"sum\": 0.8933699525307403, \"min\": 0.8933699525307403}}, \"EndTime\": 1579888125.551406, \"Dimensions\": {\"model\": 15, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551393}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8915363784160182, \"sum\": 0.8915363784160182, \"min\": 0.8915363784160182}}, \"EndTime\": 1579888125.551441, \"Dimensions\": {\"model\": 16, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551431}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.8946924730242489, \"sum\": 0.8946924730242489, \"min\": 0.8946924730242489}}, \"EndTime\": 1579888125.551499, \"Dimensions\": {\"model\": 17, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551482}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.4467132283907915, \"sum\": 1.4467132283907915, \"min\": 1.4467132283907915}}, \"EndTime\": 1579888125.551568, \"Dimensions\": {\"model\": 18, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.55155}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.3794935366126415, \"sum\": 1.3794935366126415, \"min\": 1.3794935366126415}}, \"EndTime\": 1579888125.551633, \"Dimensions\": {\"model\": 19, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551614}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 2.0156841333950566, \"sum\": 2.0156841333950566, \"min\": 2.0156841333950566}}, \"EndTime\": 1579888125.551695, \"Dimensions\": {\"model\": 20, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551678}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.3408575918957015, \"sum\": 1.3408575918957015, \"min\": 1.3408575918957015}}, \"EndTime\": 1579888125.551758, \"Dimensions\": {\"model\": 21, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551741}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9510605366347474, \"sum\": 0.9510605366347474, \"min\": 0.9510605366347474}}, \"EndTime\": 1579888125.55182, \"Dimensions\": {\"model\": 22, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551802}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.913547008936002, \"sum\": 0.913547008936002, \"min\": 0.913547008936002}}, \"EndTime\": 1579888125.55188, \"Dimensions\": {\"model\": 23, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551863}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.2125794919924475, \"sum\": 1.2125794919924475, \"min\": 1.2125794919924475}}, \"EndTime\": 1579888125.551949, \"Dimensions\": {\"model\": 24, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551931}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9021295336686501, \"sum\": 0.9021295336686501, \"min\": 0.9021295336686501}}, \"EndTime\": 1579888125.552016, \"Dimensions\": {\"model\": 25, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.551998}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 1.288657246067743, \"sum\": 1.288657246067743, \"min\": 1.288657246067743}}, \"EndTime\": 1579888125.552085, \"Dimensions\": {\"model\": 26, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.552066}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.905316288080736, \"sum\": 0.905316288080736, \"min\": 0.905316288080736}}, \"EndTime\": 1579888125.552149, \"Dimensions\": {\"model\": 27, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.552131}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9736106943671459, \"sum\": 0.9736106943671459, \"min\": 0.9736106943671459}}, \"EndTime\": 1579888125.552214, \"Dimensions\": {\"model\": 28, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.552196}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9678176900200774, \"sum\": 0.9678176900200774, \"min\": 0.9678176900200774}}, \"EndTime\": 1579888125.552277, \"Dimensions\": {\"model\": 29, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.552259}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9539102404476323, \"sum\": 0.9539102404476323, \"min\": 0.9539102404476323}}, \"EndTime\": 1579888125.552339, \"Dimensions\": {\"model\": 30, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.552322}\n",
      "\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"validation_multiclass_cross_entropy_objective\": {\"count\": 1, \"max\": 0.9067518333302992, \"sum\": 0.9067518333302992, \"min\": 0.9067518333302992}}, \"EndTime\": 1579888125.552411, \"Dimensions\": {\"model\": 31, \"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888125.552393}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, epoch=9, validation multiclass_cross_entropy_objective <loss>=0.903430175781\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=9, criteria=multiclass_cross_entropy_objective, value=0.890207921014\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] Early stop condition met. Stopping training.\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #progress_metric: host=algo-1, completed 100 % epochs\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 61, \"sum\": 61.0, \"min\": 61}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Total Batches Seen\": {\"count\": 1, \"max\": 622, \"sum\": 622.0, \"min\": 622}, \"Total Records Seen\": {\"count\": 1, \"max\": 612790, \"sum\": 612790.0, \"min\": 612790}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 60079, \"sum\": 60079.0, \"min\": 60079}, \"Reset Count\": {\"count\": 1, \"max\": 12, \"sum\": 12.0, \"min\": 12}}, \"EndTime\": 1579888125.553458, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"training_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\", \"epoch\": 9}, \"StartTime\": 1579888122.731017}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #throughput_metric: host=algo-1, train throughput=21285.1966758 records/second\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 WARNING 140276589545280] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 WARNING 140276589545280] wait_for_all_workers will not sync workers since the kv store is not running distributed\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:45.773] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 32, \"duration\": 200, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #early_stopping_criteria_metric: host=algo-1, epoch=9, criteria=multiclass_cross_entropy_objective, value=0.890207921014\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:45.841] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/validation\", \"epoch\": 34, \"duration\": 57, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #validation_score (algo-1) : ('multiclass_cross_entropy_objective', 0.8930183908434587)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #validation_score (algo-1) : ('multiclass_accuracy', 0.5541944074567243)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #validation_score (algo-1) : ('multiclass_top_k_accuracy_3', 1.0)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #validation_score (algo-1) : ('dcg', 0.8181544074047103)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #validation_score (algo-1) : ('macro_recall', 0.54473513)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #validation_score (algo-1) : ('macro_precision', 0.57366365)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #validation_score (algo-1) : ('macro_f_1.000', 0.5493278)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, validation multiclass_cross_entropy_objective <loss>=0.893018390843\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, validation multiclass_accuracy <score>=0.554194407457\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, validation multiclass_top_k_accuracy_3 <score>=1.0\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, validation dcg <score>=0.818154407405\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, validation macro_recall <score>=0.544735133648\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, validation macro_precision <score>=0.573663651943\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, validation macro_f_1.000 <score>=0.549327790737\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] Best model found for hyperparameters: {\"lr_scheduler_step\": 455, \"wd\": 0.0017892183551686785, \"optimizer\": \"adam\", \"lr_scheduler_factor\": 0.9898241409192494, \"l1\": 0.03914801904802198, \"learning_rate\": 0.03668943475753457, \"lr_scheduler_minimum_lr\": 2.294182951069527e-05}\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] Saved checkpoint to \"/tmp/tmppiN0KF/mx-mod-0000.params\"\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:45.863] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/test\", \"epoch\": 0, \"duration\": 29198, \"num_examples\": 1, \"num_bytes\": 412000}\u001b[0m\n",
      "\u001b[34m[2020-01-24 17:48:45.926] [tensorio] [info] epoch_stats={\"data_pipeline\": \"/opt/ml/input/data/test\", \"epoch\": 1, \"duration\": 63, \"num_examples\": 8, \"num_bytes\": 3094120}\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"Max Batches Seen Between Resets\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Batches Since Last Reset\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Number of Records Since Last Reset\": {\"count\": 1, \"max\": 7510, \"sum\": 7510.0, \"min\": 7510}, \"Total Batches Seen\": {\"count\": 1, \"max\": 8, \"sum\": 8.0, \"min\": 8}, \"Total Records Seen\": {\"count\": 1, \"max\": 7510, \"sum\": 7510.0, \"min\": 7510}, \"Max Records Seen Between Resets\": {\"count\": 1, \"max\": 7510, \"sum\": 7510.0, \"min\": 7510}, \"Reset Count\": {\"count\": 1, \"max\": 1, \"sum\": 1.0, \"min\": 1}}, \"EndTime\": 1579888125.943314, \"Dimensions\": {\"Host\": \"algo-1\", \"Meta\": \"test_data_iter\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1579888125.863112}\n",
      "\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #test_score (algo-1) : ('multiclass_cross_entropy_objective', 0.8869041432712113)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #test_score (algo-1) : ('multiclass_accuracy', 0.5596537949400799)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #test_score (algo-1) : ('multiclass_top_k_accuracy_3', 1.0)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #test_score (algo-1) : ('dcg', 0.8206574542862122)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #test_score (algo-1) : ('macro_recall', 0.5506639)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #test_score (algo-1) : ('macro_precision', 0.5803919)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #test_score (algo-1) : ('macro_f_1.000', 0.5559134)\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, test multiclass_cross_entropy_objective <loss>=0.886904143271\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, test multiclass_accuracy <score>=0.55965379494\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, test multiclass_top_k_accuracy_3 <score>=1.0\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, test dcg <score>=0.820657454286\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, test macro_recall <score>=0.550663888454\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, test macro_precision <score>=0.58039188385\u001b[0m\n",
      "\u001b[34m[01/24/2020 17:48:45 INFO 140276589545280] #quality_metric: host=algo-1, test macro_f_1.000 <score>=0.555913388729\u001b[0m\n",
      "\u001b[34m#metrics {\"Metrics\": {\"totaltime\": {\"count\": 1, \"max\": 29390.14196395874, \"sum\": 29390.14196395874, \"min\": 29390.14196395874}, \"finalize.time\": {\"count\": 1, \"max\": 306.02192878723145, \"sum\": 306.02192878723145, \"min\": 306.02192878723145}, \"initialize.time\": {\"count\": 1, \"max\": 310.3640079498291, \"sum\": 310.3640079498291, \"min\": 310.3640079498291}, \"check_early_stopping.time\": {\"count\": 11, \"max\": 1.0781288146972656, \"sum\": 6.229639053344727, \"min\": 0.2079010009765625}, \"setuptime\": {\"count\": 1, \"max\": 29.234886169433594, \"sum\": 29.234886169433594, \"min\": 29.234886169433594}, \"update.time\": {\"count\": 10, \"max\": 3197.6890563964844, \"sum\": 28558.29405784607, \"min\": 2746.2689876556396}, \"epochs\": {\"count\": 1, \"max\": 15, \"sum\": 15.0, \"min\": 15}}, \"EndTime\": 1579888125.95494, \"Dimensions\": {\"Host\": \"algo-1\", \"Operation\": \"training\", \"Algorithm\": \"Linear Learner\"}, \"StartTime\": 1579888096.651691}\n",
      "\u001b[0m\n",
      "\n",
      "2020-01-24 17:48:57 Uploading - Uploading generated training model\n",
      "2020-01-24 17:48:57 Completed - Training job completed\n",
      "Training seconds: 79\n",
      "Billable seconds: 79\n"
     ]
    }
   ],
   "source": [
    "linear_learner_estimator.fit([train_records, val_records, test_records])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------------------------------------------------------------!"
     ]
    }
   ],
   "source": [
    "linear_learner_predictor = linear_learner_estimator.deploy(initial_instance_count=1, instance_type='ml.m4.xlarge')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_metrics(predictor, test_features, test_labels):\n",
    "    \"\"\"\n",
    "    Evaluate a model on a test set using the given prediction endpoint. Display classification metrics.\n",
    "    \"\"\"\n",
    "    # split the test dataset into 100 batches and evaluate using prediction endpoint\n",
    "    prediction_batches = [predictor.predict(batch) for batch in np.array_split(xTest, 100)]\n",
    "\n",
    "    # parse protobuf responses to extract predicted labels\n",
    "    extract_label = lambda x: x.label['predicted_label'].float32_tensor.values\n",
    "    test_preds = np.concatenate([np.array([extract_label(x) for x in batch]) for batch in prediction_batches])\n",
    "    test_preds = test_preds.reshape((-1,))\n",
    "\n",
    "    # calculate accuracy\n",
    "    accuracy = (test_preds == test_labels).sum() / test_labels.shape[0]\n",
    "\n",
    "    # calculate confusion matrix\n",
    "    label_mapper = np.vectorize(lambda x: activty_map[x])\n",
    "    confusion_matrix = pd.crosstab(label_mapper(test_labels), label_mapper(test_preds), \n",
    "                                   rownames=['Actuals'], colnames=['Predictions'], normalize='index')\n",
    "    # display results\n",
    "    plt.figure(figsize=(16, 6))\n",
    "    sns.heatmap(confusion_matrix, annot=True, fmt='.2f', cmap=\"YlGnBu\").set_title('Confusion Matrix')  \n",
    "    \n",
    "    print('Accuracy: {:.3f}'.format(accuracy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.560\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA80AAAGHCAYAAACK6rDWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3XmcVmXd+PHPPYOgggKyyCKKAl6a+76GuKBlbj1p5ZKapY9LPaWp1VMZWfrT9Ekr9XF5zH1LK3MrNZU00xRFQYyvuCDIqoICAiIz8/vjPoMzw9wwcGa9+bxfr/s19znnOud8r2GZ+d7f67pOoaamBkmSJEmStLyKtg5AkiRJkqT2yqRZkiRJkqQSTJolSZIkSSrBpFmSJEmSpBJMmiVJkiRJKsGkWZIkSZKkEjq1dQCSJGl5KaUhwG3ADsBfIuKInNebDIyKiBtzB9eOpJQGA28Bm0bE5LaNRpJUjkyaJUlaRSmlzYBRwEigB/A28ABwSUTMaqbb/ACYDwwD5jXD9XYBFjTDdRqVUhoBPAG8FhGpwbHrgG8CP4uIUU283mSaluRPBfoD765SwJIkNZFJsyRJqyCltCXwNPAP4ChgCrAx8A3gVOBnzXSrTYGnI2JKc1wsIlorqeyaUto1Ip4DSCl1Ab4ETGvuG6WUukTEx8DM5r62JEm1TJolSVo1VwKTgMMjoibbNwX4R0qpR22jlNJZwHeBvsCLwLci4sXs2IkUK9XfBy4GugN/AE6PiCVZlXUTYP+U0nnA17PLjoqIwXXuMQoYEREjsu2jgZ8Cg4EPgPsi4pTs2GTqVG5TSjsDvwF2BGYDl0XEZXWuXQOcAJwI7A5MAL4eEa+s5PtzO3Ac8Fy2fUj2/VpUt1FK6XLgMKAfxeHV50fEXdmx0Vn/b0gp3QD8PSJGpJRupLgey+zse/JISumH1BmenVL6PbB+RHwuu1YCxgJHRMQjK4ldkqTluBCYJElNlFLqDYwAflUnYV4mIj7I2n2FYsX5+xTnJE8AHkopda3TvC9wDHAoxUrslyhWq6E4lPoZ4H8oDj2+qwmx9QduBM4HEsVk9YUSbbsBDwHjs/h+APwipXRkg6Y/BS4Dtqc4/Pn6lcUB3Ap8JaVU+8H8cdm+ht4HvgpsDfwauCWltE127D+Adyh+6NA/26bOsQKwB/CTRq57OrBDSunklFIlcBNwqwmzJGl1WWmWJKnphlBM2GIl7b4L/DYi7gBIKZ0GfA44Frg2a9MF+EZEvJe1uRvYB/jfiHg3pbQEWBARM7PjK4ttALAY+HNEfERxnvWYEm2PzdqeHhFVwL9TStsCZwH31Gl3ZUTcn93/F8DTKaV1ImLRclfMRMS4lNIs4MCU0jMU532fQvFDgbrtfl5n89qU0mHAF4HxETEnpVQFfFjb/zqmA2fXfmiRLQRW97rvZd/vG4FtKVayR5aKV5KklbHSLElS89sCeLZ2IyKWUkxgt6jTZmZtwpyZQbH6vLpeBsYBb6aUbkwpHZVSWmsF8Y3JEuZazzSID4qV6LrxAfRpQiy3Uqwwf5ni0Orl5lOnlL6WUno+pfReSmkBcBAwqAnXfqmxKn9dEfFH4GHgW8A3I2J+E64rSVKjTJolSWq6N7KvKy37NsEnDbZrWPHP5WqKVe66liXFWWI+guKQ51nAJRTnWTeWODe8TlNirE1Um/K7w20Uh52fQiNDs1NKewH/B9xMsQq8PcUkt1SSX9fClTVIKa2bXbMKGNqEa0qSVJJJsyRJTZRVhkcD300pLZd4ppS61zaluHhW7f5OwM7AxBy3fxfonc3TrbVN3QYRURURT0TE94Fds1e9NpmJwM4NrrVHzvjqxjENeB7YHPhzI032oDgM+7cRMZbihxFDGrT5BKhc7symuZji47qOAS5pOIRbkqRV4ZxmSZJWzbcoPm7qkZTSxcDrFOcTn0RxFe3zKS5sdW1KaRzwEvA9inOYb89x3+cpVojPSyndAhwMDM+uT0ppN4qV5kcpLrJ1FMV5y409suo24BfAVSmlyyguBvZtPl2luzkcAqwVEY1Vht8AtkopfYHi9+9bwMAGbd4G9k4pPQgsiogPm3LT7HnRJwO7ZvOrjwCuTykdsLJh3ZIkNcZKsyRJqyAiJlBc3XoWxeRzInADMAe4OmtzB8Xk+VKKc423Bg6OiAU57vsexaT2eIqJ8nbA/9ZpMg/YH3gE+DdwNPAfDeZN115rPsWke7ssvl8C50XE71c3vkbusXAFie69wHUUh27/k+KQ6z81aPMzitX6qTRerV5Otjr574ALImJctvtbwGeA01apA5IkZQo1NX7oKkmSJElSY6w0S5IkSZJUgkmzJEmSJEklmDRLkiRJklSCSbMkSZIkSSWYNEuSJEmSVILPaVabGnLCXS7fLnUgNbv1a+sQJK2iN0/v39YhSFotmxfaOoJVsc7GR+f+vX7RlDvaZZ9NmiVJkiRJuRQK5TuIuXx7JkmSJElSTlaaJUmSJEm5FMq4HmvSLEmSJEnKpZyHZ5s0S5IkSZJyKeekuXx7JkmSJElSTlaaJUmSJEm5FArt8mlRzcKkWZIkSZKUU/kOYjZpliRJkiTlUs5zmk2aJUmSJEm5lHPSXL49kyRJkiQpJyvNkiRJkqRcCmVcjzVpliRJkiTlUs7Ds02aJUmSJEm5mDRLkiRJklRCOSfN5dszSZIkSZJystIsSZIkScqlQKGtQ2gxJs2SJEmSpFzKeXi2SbMkSZIkKZdyTprLt2eSJEmSJOVkpVmSJEmSlEs5V5pNmiVJkiRJOZk0S5IkSZLUKCvNkiRJkiSVUM5Jc/n2TJIkSZKknKw0S5IkSZJyKZRxPdakWZIkSZKUSzkPzzZpliRJkiTlUigU2jqEFmPSLEmSJEnKxUqzJEmSJEltLKW0OXAT0At4Hzg+IiY10u7LwE+AAlADHBARs1JKlcBvgM9l+y+KiP9b0T3L9+MASZIkSVKrKFCR+9VEVwNXRsTmwJXANQ0bpJR2BkYBIyNia2Bv4MPs8LHAUGAYsAcwKqU0eEU3NGmWJEmSJOVSKFTkfq1MSqkvsCNwR7brDmDHlFKfBk3PBC6NiJkAEfFhRCzOjn0FuC4iqiPiXeBe4KgV3dfh2ZIkSZKkXFppTvMgYFpEVAFERFVKaXq2/9067T4DvJVSehLoBvwRuCAiaoCNgbfrtJ2SnV+SSbMkSZIkqZx0ArYFRgKdgb9STI5vXp2LOTxbkiRJkpRLK81pngoMzBbzIvs6INtf19vAPRHxcUTMB/4M7JodmwJsUqftxo2cX49JsyRJkiQpn0JF/tdKRMRs4CXg6GzX0cDYbG5yXbcDB6aUCimltYD9gZezY3cDJ6eUKrK50EcAf1jRfU2aJUmSJEm5tMZCYJlTgW+nlF4Dvp1tk1J6KFs1G+BOYDbwKsUkewJwfXbsFuBNYBLwLHB+RLy5wr7V1NQ0/TshNbMhJ9zlX0CpA6nZrV9bhyBpFb15ev+2DkHSatm80NYRrIqhO/869+/1r4/5Trvss5VmSZIkSZJKcPVsSZIkSVIuTVzIq0MyaZYkSZIk5dJKz2luEybNkiRJkqR8Cu1yOnKzMGmWJEmSJOVTvoXmcu6aJEmSJEn5WGmWJEmSJOXj8GxJkiRJkkowaZYkSZIkqYQynvhbxl2TJEmSJCkfK82SJEmSpFxqHJ4tSUXDt+nHT47dgcqKAnf9/U2ueXBiveNf2nsw3//KdsyauwiAWx57nd///U0AbvjecLYf0osxk97j5MueavXYpTXV8EE9OW/voVRUFPj9qzO4euzURtt9frPeXPm5rTj87hcY/+4Ctu27HheO2ByAAvDr5yfzyFvvt2Lk0prrySdf4IILrqO6upqjjhrJKaccVe/4kiWfcO65v2LChDfo0WM9LrvsXDbaaEMAJk58i5/+9EoWLFhIRUUF99zzK7p06dwW3dCapHxz5vaTNKeUJgOHRMQrdfaNAc6OiNEteZ86x7YGLgWGUhy6/iJwZkQ0+ttFSmkAcFtE7Jsjnh7AKRHxy9U8fzKN9CelNAJ4CHitzu5xEXH8Sq53BDA9Ip7Ltnem+D04dnXiU3mpKBQYdfxOnPDL0cycs4g/jRrJY2On8/r0efXaPfjcVH52y4vLnX/dXyaydudOHL3vkNYKWVrjVRTgZ8OHcfz945i54GPuPXJH/jb5fV6fu7Beu65rVXLCtgMZO/PTf8+vzfmIw+9+gaoa6LNuZx788k48NvkZqmpauxfSmqWqqorzz7+aG274ORtu2IsjjzyL/fbbjaFDN17W5u67H2H99bvx6KPX8uCDT3LppTdy+eXfZ+nSKs4551dccslZbLHFpsydO49OnSrbsDdaY1SUb9bsnOZMSqkn8Dfg+ogYGhGbAU8Dj6SU1mqkfaeImJ4nYc70AM7NeY1SXo2I7eu8VpgwZ44Adq3diIgxJsyqtd1mG/D2rPlMffcjPqmq5oF/TeGAHQc2+fx/vjqbjxZ/0oIRSmpou77r8/aHi5g6bzGfVNfwwOuzGblpr+XanbXrYK4dO5WPq6qX7Vu8tHpZgtyl0l8ZpNYybtwkNtmkP4MG9aNz57X4wheG89hj/6rX5vHH/8UXv7g/AAcdtBfPPPMyNTU1PP30WFIazBZbbApAz57rU1lp0qxWUCjkf7VT7abSvDIppWOA7wC1Y0vOjojHsmOTgZuBkUB/4NKIuCI79lngKmAR8CylBw58GxgdEXfX7oiIy1JKXwG+CtySUhoN/BPYDVicUjoDGBMRvbN77QZcBKyfXeK8iHgwpTQYGANcAxwMrAt8IyL+AVwJ9EgpvQQsjIg9U0r9gd8CGwPrAHdExIWr2J+SUkrbZNfoCqwNXBsRl6eUDgIOAw5IKX0T+BUwJft+7rySfpBS+hbFP6MPKFa5z6j93qg8bNhzHWbMWbRse+achWw3ZPlfvj+380bsmvrw1sz5XHD72HrnSGpd/bp2ZsaCj5dtz1jwMdtvuH69Np/p3Y3+3brw+Ntz+Ob2g+od267vely8X2Lgemvzvb/92yqz1ApmzXqffv0+/RVqww17MW7ca8u16d+/2KZTp0rWW68rc+fO4623plEowDe+cR5z5nzIwQcP5+STv9Sq8Uvlpr19bHxPSuml2hfwmTrHHgZ2j4gdKCaxNzU4d92I2AMYAVyUUuqWUuoC3Al8OyJ2Bf5BMRFtzLYUk9CG/gVsV2d7a+CgiPhC3UbZMOurgWMiYifgEOCabD9AL+CZLP7zgYuz/WcAH2SV4D2zfTcDv8li3gn4fEpp5Cr2B+Azdb+fKaXzsv2TgQMiYkeKVeVTUkpbRsTDwH3ARVk8NzdyzUb7kVLaFvghsGdE7AJ0X0Fc6qAa/QCwwS/Qj42dzj7fe4Av/Phhnp4wi0tO3q1VYpNUQiP/bmtq6h/+8V5DuOCfbzR6+suz5/O5O8dwxD0vctqOG9O5sv1WAqRyUVOz/KdThQY/hEu1qaqq4oUXXuWSS77H7bdfzN/+9gzPPPNyi8UqLVNohlc71d6S5iPrDicGXq1zbAjwcEppAnAX0C+l1K/O8TsBImIyMBfYCEgUq7ejs2O/Bz4sce+m/jHdHhFLG9m/J7Ap8Jcs4f8LxXRiaHZ8QUQ8kL1/NuvPclJKXSkm/r/JrvMcMADYchX7A8sPzz4/278ucH1KaTzFIegDqP/BwIqU6scI4KGIeDfbvqGJ11MHMnPOIvpvsM6y7X4brMusD+pXkT/4aAlLlhaHd941+k22HtyzVWOUVN/MBUvo363Lsu3+3bowe+GnledunSvZfIOu3HH49jx53G7ssOH6XHvw1mzTp1u967wxdyELl1aTNujaarFLa6p+/Xozc+Z7y7ZnzXqfvn03WK7NjBnFNkuXVjF//kf06LEe/fr1Ztddt2aDDbqzzjprM3z4zkyY0PiHYlKzqijkf7VT7S1pXpE7gKsiYitgR2ApxaHFtRbXeV9Fcej5qnznXwZ2b2T/bsC4OtsLSpxfoLjQVt0kdVBEjMmOf1ynbW18jamgmGzvUuc6QyLiNzTf5y8XAjOBHSJiO4qJ+dorPmWZUv0osFzNUeVm3FtzGLzhemzUuytrVVZwyG4b89jYafXa9On+6V+lA3YcwOvT57d2mJLqGDd7HoO7r8NG663NWhUFDhnal7/VWQF7/pIqdr7hnwy/9V8Mv/VfjJ01j1MeeoXx7y5go/XWprawPKBbFzbrsQ7vzF9c4k6Smss22wxj8uTpTJ06kyVLPuHBB59kv/12rddmv/12409/egyAhx9+mt1335ZCocDee+9IxGQWLVrM0qVVPP/8KwwdOqix20jNyznN7UIP4K3s/TeALitoW2sisE5KaXhEPJlSOpLSw4avAP6dUjqqdl5zSulMoCfFhH1l/gkMSyntGxFPZOfvQnEO8IrMA9bNFhZbGhHzU0pPAT8Afp5dZxDwySr2Z0V6UEzwl2Yrhn8WuL1OPKtzzdHAOSml3hHxHnDCalxD7VxVdQ0/u+VFbjxnHyoqCtzz5JtMmjaP735xa8ZPnsNjY6dzwoHD2H+HgVRV1fDhRx9z7v99unDJnf+9H5v1X4+ua3fiH5cdyg+vf56nXpnZhj2Syl9VDYx66nVuOnQbKgoF7p44k0lzF/LdXQYz/t35PDa59COkdu6/PqfuuDVLq2uorqnhvCcnMXdxY4OtJDWnTp0qOe+8U/nmN39KVVU1X/rSAQwbtgm//vWtbL31MPbffzeOPHIk55zzK0aOPIXu3btx2WXFdWW7d+/GiScewZFHnkWhUGD48J0ZMWKXNu6R1gjtN+fNrSMlzd8F7k0pTQP+Dqz0QZER8XFK6WjgqpTSIuBxigtbNdZ2TkppJHBpSukiin/sLwEHRsRKl/uNiLkppcOAS1JKl1NcsOxN4NCVnDcnpXQbMD6lNDeb13wscFk2fBpgPnBSRMxsan8yn8mGeNeaHhEHA7+guLDZccAbwJN12twC3JhSOopPFwJbqYh4OaX0S+CZlNJMiiuRr2jouDqo0eNmMHrcjHr7Lv/Tp088u/Tu8Vx69/iGpwHw1Qsfb9HYJDVu9JQ5jL59Tr19lz8/udG2x/z507mP9742m3tfm92SoUkqYZ99dmaffXaut+873zlu2fsuXTrzm9/8oNFzDz98Xw4/PO8DXiTVKjS2iIC0OlJK60XE/Oz9KGBoRBy3onOGnHCXfwGlDqRmt34rbySpXXnz9P5tHYKk1bJ5h6rdDvvc73L/Xj/prye1yz53pEqz2r+LUkp78WmV/ZQ2jkeSJElSa2iX6W7zMGlWs4mIM9o6BkmSJEmtr6YdL+SVV0daPVuSJEmSpFZlpVmSJEmSlE87fs5yXibNkiRJkqR8yjdnNmmWJEmSJOVUxnOaTZolSZIkSfmU8fBsFwKTJEmSJKkEK82SJEmSpHzKt9Bs0ixJkiRJysk5zZIkSZIklWDSLEmSJElSCWW8WlYZd02SJEmSpHysNEuSJEmS8nF4tiRJkiRJJZRvzmzSLEmSJEnKp6aifLNm5zRLkiRJklSClWZJkiRJUj7OaZYkSZIkqYTyzZlNmiVJkiRJOZXxnGaTZkmSJElSPmU8PNuFwCRJkiRJKsFKsyRJkiQpn/ItNJs0S5IkSZJyck6zJEmSJEkllHHS7JxmSZIkSZJKsNIsSZIkScqlpnwLzSbNkiRJkqScynh4tkmzJEmSJCmfMn5Os0mzJEmSJCmfMq40uxCYJEmSJEklWGmWJEmSJOVTxuVYk2ZJkiRJUj7OaZYkSZIkqYQyntNs0ixJkiRJyqWmjCvNZTzyXJIkSZKkfKw0S5IkSZLyKeNyrEmzJEmSJCkf5zRLkiRJklSCc5olSZIkSVrzWGmWJEmSJOXj8GxJkiRJkkoo35zZpFmSJEmSlE+NlWZJkiRJkkoo46TZhcAkSZIkSSrBSrMkSZIkKZ8yfuSUSbMkSZIkKZ8yHsNs0ixJkiRJysdKs9Qy7v9117YOQdIq2Gnbq9s6BEmraMaJx7R1CJJWQ/91N2/rEFaNC4FJkiRJkrTmsdIsSZIkScqnjCvNJs2SJEmSpFxqWmlOc0ppc+AmoBfwPnB8REwq0TYBY4GrIuLsbN+NwAHAe1mzuyPighXd06RZkiRJkpRP6038vRq4MiJuTSkdB1wD7NewUUqpMjt2byPXuCgirmjqDZ3TLEmSJElq91JKfYEdgTuyXXcAO6aU+jTS/AfAA8Bree9r0ixJkiRJyqdQyP9auUHAtIioAsi+Ts/2L5NS2hY4CLisxHXOSimNTyndm1LacmU3NWmWJEmSJOVTUcj/agYppbWA64BTa5PrBn4EDI2IbYA/An/NhnKX7lqzRCZJkiRJWnO1TtI8FRhYm+RmXwdk+2v1B4YAD6WUJgPfBU5OKV0LEBHTIqI6e38z0A3YaEU3dSEwSZIkSVI+rbB4dkTMTim9BBwN3Jp9HRsR79ZpMwXoXbudUhoFdKuzevbAiJiWvT8IqAKmrei+Js2SJEmSpI7iVOCmlNJ5wFzgeICU0kPAeRExZiXn35RS2hCoBuYBh0XE0hWdYNIsSZIkScqlppnmJK9MREwEdmtk/8El2o9qsH3Aqt7TpFmSJEmSlE/TVr/ukEyaJUmSJEn5tFKluS2YNEuSJEmS8infnNlHTkmSJEmSVIqVZkmSJElSLhVlXI41aZYkSZIk5VLG64CZNEuSJEmS8innpLmMi+iSJEmSJOVjpVmSJEmSlEuhjEvNJs2SJEmSpFzKOGc2aZYkSZIk5WPSLEmSJElSCYUyXi2rjLsmSZIkSVI+VpolSZIkSbk4PFuSJEmSpBIqTJolSZIkSWqclWZJkiRJkkoo56TZhcAkSZIkSSrBSrMkSZIkKZdCGZeaTZolSZIkSbmU83OaTZolSZIkSbmUcaHZOc2SJEmSJJVipVmSJEmSlEs5V5pNmiVJkiRJuazxSXNK6WvACxHxakppS+B6YClwakS82pIBSpIkSZLat4oyTpqbOqf5F8C72ftfAU8BfwOuaomgJEmSJEkdR6GQ/9VeNXV49gYR8W5KaR1gL+AI4BPguy0WmSRJkiRJbayplebZKaWhwMHAcxHxMbA20I4/D5AkSZIktQYrzXA+MAaoAb6c7dsfeLklgpIkSZIkdRyFMp7U3KRKc0TcBAwABkbEo9nufwFfbanAJEmSJEkdwxpZaU4pNUyoFzfY/15LBSVJkiRJ6jjac9Kb14qGZy+lOBy7lEJ2vLJZI5IkSZIkqZ1YUdK8aatFIUmSJEnqsNbISnNEvN2agUiSJEmSOqYyXgesyatnk1L6PDAC6E2dR01FxEnNH5YkSZIkqaMo50pzk1bPTin9N3ATsC5wNPARcCiwsOVCkyRJkiSpbTUpaQZOAfaPiG8DS7KvhwIbtVhkkiRJkqQOoVCR/9VeNTW0nhExPnu/JKW0VkQ8C+zbQnFJkiRJkjqINfI5zQ28kVLaKiImABOA01JKc4EPWi40Se3Ri89M5Ppf3Ut1dTUHHLYbXzph/3rHJ4x9g99d9mcmvz6D7/38OPbcf7tlx26+4gHGPP1vAL580gHsPXKHVo1dWlON3Gc7Lh11PJWVFdx45xNcetV9y7X50iG786Mzv0RNDYx/9W1O/K8rABg0oBdX/fIUNurfixpqOOKEi5nyznut3QVpjfOvpydyxSV/pqq6mi8csRvHnrRfveMvv/AGV1x6H29MmsF5/+9YRoz89Oftfjudw6ZD+wOwYb8eXPhrlyBSyyu056w3p6Ymzf8N9Kzz/lagK3B6U05OKU0GDomIV+rsGwOcHRGjmxjDat2nzrGtgUuBoRQr7C8CZ0bE1BLXGgDcFhGrXU1PKfUATomIX67m+ZMp3Z9hwP8DdgLmUnxe9kPAjyOianVjbi4ppdHApRHxQFvHouZTVVXNtZf8kVG//U969e3OuSdezq6f3YpBm/Vb1qbPhj359k++yp9vG13v3DH/eJU34x0uu+UsPvlkKT8+9Sp23GNL1u22div3QlqzVFQUuPwXX+cLx17ItBnv84/7L+CBR19g4qRpy9oMGdyPs08/nP3+YxQffPgRfXqtv+zY/112OhdfcS+PPzWerut2obq6pi26Ia1Rqqqq+fVFf+LS/z2FPht259Rjf81e+3yGwUM+/Xnbt39PfvCzr3DXzX9f7vzOXdbi+rvOas2QpXZdKc6rScOzI+KvEfGP7P0zETEkIvpFxB9bNrzmkVLqCfwNuD4ihkbEZsDTwCMppbUaad8pIqbnSZgzPYBzc15jOSml/sBTwAMRsWlE7Ah8luJCbV2a+34lYmjyyusqH5NenUL/jXrRb2Av1lqrE3uP3IHnnpxQr03fARsweNgACg2eOzD1rVlstcMQKjtVsvY6XRg8bABjn53YmuFLa6Rdth/KG5NnMnnKbD75pIq773+GQw7cuV6bk47Zj2tufoQPPvwIgHffnwfAFsMG0qlTBY8/VZyh9dHCj1m0eEnrdkBaA018ZQoDB/ViwEbFn7f7HbQ9T4+u//O2/4ANGLL58j9vJTW/JiU+KaWNSx2LiCl5g0gpHQN8B+ic7To7Ih7Ljk0GbgZGAv0pVi+vyI59FrgKWAQ8S51HYTXwbWB0RNxdJ+7LUkpfAb4K3JJVRv8J7AYsTimdAYyJiN7ZvXYDLgJqP34/LyIeTCkNBsYA1wAHU0xcv5F9yHAl0COl9BKwMCL2zBLe3wIbA+sAd0TEhavYnzOAJyLixjr9mZd9D2u/pwm4nOIjwjoDl0fEDdmxGuBHwBeBXsA5EfGHJvbzCuAA4NaU0iTgF8DaFP8uXRARd5aIWWVgzuwP6b1hj2Xbvfp257UJTfsvYNNhA7jr+kc47Jh9+HjxEl554XUGbbphS4UqKTOgX0/emf7+su1pM95n1+2H1mszbNNi9erxP46isqKCX1z2Bx79+8sM27Q/H8xbyJ3XnMkmg/ryxD/G8+OL7rDaLLWwd2d/SJ86P2/7bNiDV195u8nnL1mylFOOuZzKThUc8/X9+Oy+W7dEmFI95Vxpbmq1cDJQw6dJXN2flpVNvMY9KaXFdbY3r/P+YYrJY02W7D1G/ZW5142IPbLE7ZWU0o3AJ8CdwLERMTql9GWKyXFjtgUNB7vQAAAgAElEQVT+0cj+fwHbAbdk21sDB0XE0uxewLJh1lcDB0fEjCzxfT4b8g3FxPOZiPhRSulY4GJgL4rJ7ZiI2L7OPW8Gfh4RT6aUOgOPpZSeB55chf7sCDxS4lhtFfj27FoTU0rrAWNSSs9ERG1pb15E7JJS2gv4PfCHJvbz3xExKrtPT2DviKhKKW0IvJBSejgi5paKTR1bY78mN/U/yO13T0z691R+8M3f0r1nV9I2m1BR2Y6XSZTKRGNzzGoa/GOu7FTJ0MH9OPDLP2dg/w147J6fstPIc+nUqYK9dtmC3Q/+IVOnvcetV/4XXztqH266a3TrBC9pmULJWsryfv/Qj+jdtzvT33mfM0+5ms2G9mPgoN4tGJ1k0gzQcAjzAOAnwPKTKEo7spE5zbWGAHeklAZSTIb7pZT6RcTM7PidABExOVuAbCOK1dOFtXOiI+L3KaVrS9y7qX+Et0fE0kb27wlsCvylmNMDxfxhKPAesKDO3N1ngf9p7OIppa7ACKBPneusB2wJzFqF/jS87g8oVsx7AV+huEDblsCdde7TJdtXmzTXVoSfBQaklNZuQj8XU0ywa/UBfpfNr14KbACk7JoqQ736due9WZ+u//f+7A/ZoHf3Jp9/1NcP4KivHwDAr35yKwMG9Wn2GCXVN23GHDYa0GvZ9sD+vZg+e+5ybZ57cRJLl1bx9tR3ee3NGQwd3I9pM+bw8oTJTJ4yG4D7HhnDrjsMM2mWWlifvt15t87P23dnfUDvPuuv4Iz6evct/mwesFEvtt95CJMmTjNpVosr55kCTUqaG1lYampK6b+AAG5rhjjuAL4XEfemlCqAhRSH/NaqW6Guohj3qvyxvAzs3sj+3SgOh661oMT5BWBcRAxveCCrSH/cSHyNqaCYhO4SEZ80uM52jZ/SqLHALrUbEXERcFH2QUTnLN73GlS4G1qcnVuVJci139MV9fOjiKhbn/hf4D7gP7JRAq9R/89NZWbYloOYMfU9Zk1/nw36dOcfj47lzJ8f16Rzq6qq+WjBItbv3pXJk6Yz+fUZfGe3zVd+oqRcxrz8BkM37ccmg/owfeYcjjp0j2UrY9e6/+ExfPnwPbn1nifp1XM9hm3an7emzOaDeR/Ro3tXem+wHu/Nmc+IPbfixXFvtlFPpDVH2moQ70x5jxnT3qd33+48/vBL/Pj/Hdukc+fPW0iXtTvTuXMnPpj7Ea+8NJmjTxjRsgFLmDSXsjnF+bvNoQfwVvb+GzRtMauJwDoppeHZUOcjgVIlryuAf6eUjqqd15xSOpPiiuB3NOFe/wSGpZT2jYgnsvN3oTjHd0XmAetmC4stjYj5KaWngB8AP8+uM4hidX1V+nMlMDal9LWIuCW7TiWfft8CWNjg+BbA9Gzuc3P1swcwOUuYR1KsSKuMVXaq5OSz/4Of/de1VFfXsP+hu7LxZv24/Zq/MnTLjdh1+NZMenUKF597IwvmL+L5p17lzuse5jd3nkvV0ip+dMqVAKzbtQtn/uwYKjs1dXaHpNVVVVXNmT+5kftv+SGVlRXcdNdo/v3aO/zkrCN5cfxbPPjoCzz695c5YPg2vPjYJVRVVfPfF9zGnA+KnyP/8ILbeOiOH1MowNjxb/G7Ox5v4x5J5a9Tp0q+8/0vcs7p11FdXcPnD9+FTYf043dX/ZX0mUHsNWIrJk6Ywo/PuokF8xbyzJOvcuPVj3DjH87h7Tdn8z8X3ENFoUB1TQ3HfH3feqtuS1p1hZqGE5sakSV6dRuuC2wFjIqIi5tw/mRW8MiplNLXgPOBaRSHfP8nsHM2HLveuXW3Gyyc9ThwNPCFEo9o2pbiI6eGUKyovkTxkVNvZ8dHU+cRSbULX9VZCGwX4BKKiXZn4E3gUIoLetVt1/C864C9gbnZQmD9gMsozp8GmA+clM09XpX+JIqPnNqR4tDpJRRXCL80IuZlQ6YvBwZRnHc+C/hyRLyXLQS2XkQsyK61bLup/czOG5nF+x4wjmLl/rvZn2m972cpr37wgKvJSB3ITts2x+AiSa3pzYnHtHUIklZD/3UP7VC124Me/kfu3+sfPmjvdtnnpibNJzTYtQB4OSJeb5GotMYwaZY6FpNmqeMxaZY6po6WNH/+kfxJ818ObJ9Jc1OHZ1fXDvOtK6V0bET4G5QkSZIkrcHK+ZkoTe3blSX2/7a5ApEkSZIkqb1ZYaU5pbRZ9rYipbQp9VesHkL9Va0lSZIkSWugikL5zrpc2fDs1ykuAFYA3mhwbCYwqgVikiRJkiR1IGvsI6ciogIgpfT3iNindUKSJEmSJHUkzmmGL6aUutbdkVLqmlLq0QIxSZIkSZI6kIpC/ld71dSk+QEgNdiXsv2SJEmSJJWlpj5yaquIeLHBvrHAVs0cjyRJkiSpgymU8UJgTa00z0kpDWywbwAwv5njkSRJkiR1MOU8PLuplea7gdtSSt+muIr2UODXwD0tFZgkSZIkqWNwITD4CTAeeBZYADwDvAL8qIXikiRJkiSpzTUpaY6IjyPi20A3YEPgc8DawPQWjE2SJEmS1AFUFGpyv9qrpg7PJqU0CDgeOAEYAvwF+EoLxSVJkiRJ6iDa85zkvFaYNKeU1gGOBE4ERlBcMftqisOyT4qI2S0cnyRJkiSpnSvnOc0rqzTPAj6mmCifERETAVJK57R0YJIkSZKkjqGcK80r+0DgGaAHcBBwYEqpT8uHJEmSJElS+7DCpDkiDgI2Af4InAZMSyn9FegKdG758CRJkiRJ7V05LwS20qHnETE9Ii6KiC2BfYDJwFJgXErpVy0cnyRJkiSpnaso5H+1V6s0XzsinomIU4H+wBnAli0SlSRJkiSpw6hohld71eRHTtUVER8Dd2QvSZIkSdIarD0Pr86rPSf0kiRJkiS1qdWqNEuSJEmSVKs9z0nOy6RZkiRJkpSLSbMkSZIkSSWU87zfcu6bJEmSJEm5WGmWJEmSJOXSWqtnp5Q2B24CegHvA8dHxKQGbb4OnAlUA5XAdRHxm+xYJfAb4HNADXBRRPzfiu5ppVmSJEmSlEtFIf+ria4GroyIzYErgWsaafMHYLuI2B7YE/heSmnb7NixwFBgGLAHMCqlNHiFfWtyaJIkSZIkNaKiGV4rk1LqC+wI3JHtugPYMaXUp267iJgXEbWl73WBtShWlQG+QrHyXB0R7wL3AketrG+SJEmSJK22Vqo0DwKmRUQVQPZ1era/npTSYSmlCcDbwCURMT47tHG2r9aUxs6v17cmhSZJkiRJUgcREfdFxFbA5sDXUkppda9l0ixJkiRJyqVQqMn9aoKpwMBsMa/aRb0GZPsbFRFTgOeAQ7JdU4BN6jTZeEXng0mzJEmSJCmn1hieHRGzgZeAo7NdRwNjs7nJy6SUtqjzvjewL1A7PPtu4OSUUkU2F/oIiguHleQjpyRJkiRJubRiNfZU4KaU0nnAXOB4gJTSQ8B5ETEG+M+U0oHAJ0ABuCIiHsnOvwXYDah9TNX5EfHmim5o0ixJkiRJyqW1ntMcERMpJr0N9x9c5/2ZKzi/CjhtVe7p8GxJkiRJkkqw0ixJkiRJyqWJj4zqkEyaJUmSJEm5mDRLkiRJklRCZVsH0IKc0yxJkiRJUglWmiVJkiRJubTW6tltwaRZkiRJkpSLc5olSZIkSSrBpFmSJEmSpBIqyzhpdiEwSZIkSZJKsNIsSZIkScrF4dmSJEmSJJXg6tmSJEmSJJVgpVmSJEmSpBIq2zqAFuRCYJIkSZIklWClWW1qk27rtXUIklbBpqNOa+sQJK2iPc/7sK1DkLQa3rq0rSNYNQ7PliRJkiSpBBcCkyRJkiSphMoyrjQ7p1mSJEmSpBKsNEuSJEmScnFOsyRJkiRJJZg0S5IkSZJUgkmzJEmSJEklVJbx6tkuBCZJkiRJUglWmiVJkiRJuZRzNdakWZIkSZKUi3OaJUmSJEkqwaRZkiRJkqQSXAhMkiRJkqQ1kJVmSZIkSVIuDs+WJEmSJKkEk2ZJkiRJkkoo56TZOc2SJEmSJJVgpVmSJEmSlEtlGVeaTZolSZIkSblUlPEjp0yaJUmSJEm5lPO8X5NmSZIkSVIuLgQmSZIkSdIayEqzJEmSJCkXFwKTJEmSJKkEFwKTJEmSJKmEcp7TbNIsSZIkScqlnJNmFwKTJEmSJKkEK82SJEmSpFzKuRpr0ixJkiRJyqVQxsOzTZolSZIkSbmUcc5c1lV0SZIkSZJysdIsSZIkScrF4dmSJEmSJJVQzkOYTZolSZIkSbkUCjVtHUKLMWmWJEmSJOVSxqOzy7qKLkmSJElSLlaaJUmSJEm5uBCYJEmSJEkllHHObNIsSZIkScqnooyzZpNmSZIkSVIuZZwzuxCYJEmSJEmlWGmWJEmSJOXiQmCSJEmSJJVQxjmzSbMkSZIkKZ9yTpqd0yxJkiRJUglWmiVJkiRJufjIKUmSJEmSSijjnNmkWZIkSZKUT6FQ09YhtBiTZkmSJElSLuVcaXYhMEmr5OmnXuGLX/gJh33uR9xw3V+WO/7CmNc45sifs8u2p/K3h1+od+yMU37N8N2/w3+d/tvWClcSsPfAnjz4pZ3565G78M1tB5Vsd+Dg3rx60nC26tUNgO5dOnHD57dlzNf24ke7D2mtcCUBw1MfHjt3X574wX6cuu/Qku0+v21/3rr0ULbZqDsAnSoKXPrV7fnL9/bh0XNGcNp+pc+V1DQdutKcUpoMLAC2jYjqOvsOiYhXmukeJ2bXO7LOvkOAsyNiRHPco9R96hyrAH4AnAhUATXAZRFx3Qqudz4wISLuyhHTEcD0iHhuNc49kRL9UcdVVVXNxRfczlXXncmGG/bkuK9cyD77bsdmQwcsa9O//waMuuDr3HLjI8udf/xJB7J40RL+cPeTrRm2tEarKMCP9xjKNx8ez6yPPuauw3bgiSnv88YHC+u1W7dTJcd9ZiAvz563bN+Sqmp+++JkhvXoytCe67Z26NIaq6IA539xG7527bPM/HARf/7OZ/nbqzN5fdaCeu26dqnkxL03Zezbc5ftO3i7AXSurODz//N31l6rkkfPGcF9Y6cxbe6i1u6G1jCFMi41l0OluRvwtbYOooX9CDgI2D0itszefzeldFxjjVNKlRFxXp6EOXMEsGvOa6iMvDL+LTYa1JeNBvVhrc6dOOjgXRj9xMv12gwY2JvN00ZUNPI/5267b0nXrmu3VriSgG16r8eUeYt4Z/5iPqmu4S9vvst+G/dart1/7bQJ14+fysdV1cv2LVpazYuz5tXbJ6nlbbdxT95+/yOmzlnIJ1U13P/SdEZu1W+5dmcdtAXXPPE6Hy+tWravpqaGdbtUUllRYO21KvikqpoFi5e2ZvhaQ1U0w6u96tCV5swoYFRK6Y6IWFL3QEppKHAN0AdYCvx3RPw1O1ZDMRn9ItALOCci/rCqN08pdQIezK6xDvAc8J8RsSSrth4DzAW2Bj4AvhQRM1NKnYHfAiOAacDEEtfvAvwQ2DEi5gBExNSU0tnAFcCtKaURwOXAU8AuwC9SSkcCYyLiiuxeFwD7AJ2B8cBpEbEgpXQjsBjYHBgEPAOcABwIHAYckFL6JvCriLg5pXQCcDrFvzsfZteJpvZHHdu7sz6gX/8Nlm333bAHr4x7qw0jkrQyG3btwsyPPl62PfOjj9m2z3r12my5QVf6de3C36fO4etbb9TaIUpqoF/3tZnxwaeV4ZkfLGb7TXrUa/OZAevTv8c6PP7v2Zw84tPpE38ZN4ORW/XjX+eNZJ3OlfzizxP4cNEnrRa71lxWmtu3MdnrtEaO3QbcHhHbAsdRTDD71Dk+LyJ2oVip/s0K7nFASuml2leDtlXAMRGxM8XEuBI4qc7xXSgO5d4KeBX4drb/P4FNs3MOoXRFdxiwJCIaJqHPApullGp/89km6+vuEfFAg7bnAh9GxK4RsT0wnWIiXmtr4GBgK2An4ICIeBi4D7goIrbPEubPAl8GhkfETsAlwO9WsT/qwGpYflXEQjn/DymVgZX9Cy0A399tCL987s3WCEdSEzT277amzo/gQgF+cvhWXHD/hOXabbdxD6pqatj9/EcZfuFjfHOfIQzawOkVUh7lUGkG+DHwRErp+todWTK5PXADQES8miW8uwP3Z83uzL4+CwxIKa0dEYsbuf7fGpvTnG1WAGenlD5PMWHuCdSdKPZ0REytc5+R2ft9gZsi4hPgk5TSrcDejdy71O87NQ2+ToqIZ0q0PQxYP6s+A3QB6o6pvbe23ymlF4EhwKONXOdQYDvgXyml2th6rmJ/1IH13bAnM2fMWbY9e9YH9OnbYwVnSGprMz/6mH5duyzb7te1C7MXfjowq+talQzr2ZWbPr8dAL3X6cyVI7fijEcnMOH9BctdT1LLm/HhYvr3WGfZdr8eazNr3qe/onbr0onN+63PnaftCUCf9bpw3dd35eQbnuPwHQby5MR3WVpdw/sLljBm8hy2HdSdqXMWLncfqTmVcxmlLJLmbHjwQ8BZdXavLNmE4rBkIqIqSwJX5/txDMXk8LMRMT+l9N8UhzrXu0emqs49mvr3ahLQOaW0RYNq8x7AW9kQayguiFZKATg9Ih4vcbxUjI1d53cRcV6JYypzW209mKlTZjPtnffo27cHDz/0PBde8s22DkvSCrzy3nw26b4OA7utzeyFH/P5zfpw7uhPf5ws+KSKvW7/9DPXGz+/LZc896YJs9SGxk39gMG9u7LRBusw68PFHLr9AL5z24vLjs9fvJSdfvrwsu07TtuDC+9/lfHvfMiew3qzx7Be/OnFd1incyU7bNKTG55yJIlaXjkPPiyLpDkzCniBrE8RMS+rLJ8A3JBS2oKsStrM9+0BvJclzN0pJtFjmnDeY8DXUkp3AWtl501p2CgiFqeULgauTil9MSLmppQGURwaPaqJMd4HnJVSeiYiFmVV+I0i4t8rOW8e0L3O9v3AzSmlayPinZRSJbB9RLzQ1P6oY+vUqZLv/+hozjjlcqqrqznsi3sxZOgA/ve3f+YzW23CPvttz4Txk/ned65i3ryFPDl6HFdfeR/33PczAE762i+Z/NZMFi38mM/tdy7nnX8Ce+69VRv3SipvVTVwwTOvc91BW1NRKPCnSTN5/YOFfGuHTZjw3nyemDpnhec/etSudOtcyVoVFey/SW9Ofnj8citvS2peVdU1/PRPr3DzybtTUShw9/NTmTRrAWcelBg/9QP+9uqskufe8vRkLvnK9jx89ggKBbjn+alMnDG/FaPXmqq1cuaU0ubATRTXlHofOD4iJjVocyBwIcUprL+NiLPrHBtFcY2m6dmupyPijBXds2yS5iyJuwX4Xp3dxwLXpJTOpLgQ2Nci4t1mvvXNwOEppQkUF8B6iuKCYCtzLbAtMAF4B/g7xTnBjfk5xQrwcyml2uUPfx0RNzcxxosoJtjPp5SqKVbbfwasLGm+BbgxpXQUny4E9iPgvixh7gzcTfHDilXpjzqwvYdvw97Dt6m377RvH77s/VbbDOavj/+y0XN/d8u5LRqbpMY9+c5cnnyn/ue5V4x9u9G2J/5lXL3tkXev8lMHJTWD0RNnM3ri7Hr7Lns4Gm179P9+Olpk4ZIqzrjlhRaNTWpMRetVmq8GroyIW7OnCV0D7NegzZvAycCXgMYe3XJz3UR6ZQo1Ncsv7CO1lo+W/t2/gFIHssvNZTz2SipTi179sK1DkLQa3rr00A71Q3f6wvtz/14/YN0V9zml1Bd4DeiVTbGtpFhtHtZYcTSrKndrpNLcbVWS5nJYPVuSJEmS1IYKzfBqgkHAtIioguLaVBSHWQ9axXC/mlIal1J6JKW0x8oamzRLkiRJknIpFGpyv1rJ1cCm2WOJLwH+nFLqtaITTJolSZIkSbm0UqV5KjAwG5ZN9nVAtr9JImJm9phcIuLR7NytV3SOSbMkSZIkKZdCIf9rZSJiNvAScHS262hg7Kos9pxSGljn/fbAYKDxVfYyZbN6tiRJkiSp7J0K3JRSOg+YCxwPkFJ6CDgvIsaklPYG7gTWBwoppa8C34iIh4ELU0o7UXw60RKKT1iauaIbunq22pSrZ0sdi6tnSx2Pq2dLHVNHWz373cX35f69vs/ah7XLPltpliRJkiTlUs7zfk2aJUmSJEm5NGVOckdVzh8ISJIkSZKUi5VmSZIkSVJO5VtqNmmWJEmSJOVSMGmWJEmSJKlxhUL5zvw1aZYkSZIk5VS+leby/ThAkiRJkqScrDRLkiRJknJxTrMkSZIkSSWZNEuSJEmS1CgXAvv/7d171GbTfcDx7+t+G4aYuiRSlPwM41KMW9Gx1qBSUmmkxBBaKRVJKyKqqM5C0iFSs1JKZLlkghHSVkVokFC3mTAVxi0/VmNayZRVcc9kXMbbP/Z+eLx5zjvvO8+YeeeZ72etWebZ55y9zz6e/ZzzO/sykiRJkiQ16t2e5t59HSBJkiRJUpfsaZYkSZIkdcWFwCRJkiRJamDQLEmSJElSo96d+du7NZMkSZIkqUv2NEuSJEmSutLX5/BsSZIkSZIaGDRLkiRJktSRC4FJkiRJktSod5fL6t2aSZIkSZLUJXuaJUmSJEldcXi2JEmSJEkNXD1bkiRJkqRGBs2SJEmSJHXU18PLZfVuzSRJkiRJ6pI9zZIkSZKkLjk8W5IkSZKkjlwITJIkSZKkRr0bNDunWZIkSZKkBvY0S5IkSZK60surZxs0S5IkSZK61LvDsw2aJUmSJEld6TNoliRJkiSps15ePbt3B55LkiRJktQle5olSZIkSV3q3f5Yg2ZJkiRJUlec0yxJkiRJUiODZkmSJEmSOnIhMEmSJEmSlkP2NEuSJEmSutS7/bEGzZIkSZKkrvTyQmB9/f39S/scJEmSJEkakXq3D12SJEmSpC4ZNEuSJEmS1MCgWZIkSZKkBgbNkiRJkiQ1MGiWJEmSJKmBQbMkSZIkSQ38d5qlJSgi5gAHZuajbWmzgJMz8873s5y2beOA84EtKC/OHgS+kJnPNOS1MXB1Zu7TxfmMBo7NzPMW8fg5dKhPREwAbgaebEuenZmfXkh+BwNzM/P++nlnyjWYtCjnp95iO1287bRu2xL4e2An4EVgRUrbPSMzFyzqOS8uEXEncH5m3rS0z0Xvr/o9fQ3YLjPfbkvr+N1dxDKOrvkd0pZ2IOU3ZMLiKKOpnLZtKwCnAkcDC4B+4ILM/OYg+Z0FPJaZ3+ninN5zfx3msUfTUB9pabOnWVqORMS6wO3AZZm5RWZuDtwL3BoRK3fYf6XMnNvNg3g1GjilyzyaPJ6ZO7T9GTRgrg4Gdml9yMxZBswaKXqtnUbERsDdwE2ZuVlm7gjsBawBrLq4y2s4BzsJ1G4t4MilfRLvs9OB/YHdMnNs/fuJEXFEp50jYsXMPLObgLl6z/1V6hXeRKQRJCIOB/4KWKUmnZyZP6zb5gDTgH2BjSi9IhfWbXsB/wT8GpgJ9DUU8Xngzsy8vpWQmRdExKHAYcC3a4/LfcCuwPyIOAGYlZnr17J2BaYAa9cszszM70fEpsAs4BvARykPxMdk5j3ARcDoiHgImJeZe9QH6X8EPgysDkzPzK8Msz6NImLbmseawGrApZk5NSL2Bz4GTIyIzwD/APxPvZ47L6QeRMTnKP+PXqL0lJ3QujZaPthOh91OTwDuyMwr2+rzSr2GrWsawFRg/Xpdp2bmFXVbPyUA+DjwAeBLmfnPQ6znhcBE4KqIeAo4h/J7sBLw5cy8tuGc1dsmA5MjYnpmvtG+ISK2oLSPMcBbwGmZ+e91W+N3cTjqS5zv1zxWB+4HjsvMN2pv6+GUERnjKPeaT2TmsxGxCqU9TgB+Afy0If9Vgb8BdszMFwAy85mIOJnSJq6qI7WmUl5ojQfOiYhDKL8jF9ayvgz8PqVNPgIcn5mvRcSVwHzgI8AmwAzgKGA/BtxfM3NaRBwFfJbS7l6u+eRQ6yONBPY0S0vedyPiodYfYOu2bT+gvBX+XcrD8bcGHLtGZu5OucFMiYi16s3xWuDzmbkLcA/lAbeT7SgPtwP9GNi+7fM4YP/M/MP2nerwzUuAwzNzJ+BA4Bs1HcoDwIx6/mcB59b0E4CXak/wHjVtGvD1es47AQdExL7DrA/A1u3XMyLOrOlzgIm1V2sX4NiIGJuZPwBuBKbU85nWIc+O9YiI7SgPIntk5nhgnUHOS8s222mxONrpjvXcO6oBxDWU4efjgT2BUyNiq7bdXqnbjgS+Pox6PpGZe2bmJZQh7nvWek8Ezq+9+lr+zKp/ju+w7WrgmszcDjiCEmCOadv+G9/FBhMH/Ia077uA8r3dmdKOVwT+rG37eMrLuG2Axykv0gCOAzarxxxIc4/ulsAbmTkwCJ0JbB4Ro+rnbWtdd+swNeEU4OXM3CUzdwDmUu5/LeMoL962ofw2TOx0f60v1/4E2Lu2068Clw+zPtJSZ0+ztOQd0mGuZMvvANMj4oPAm8CGEbFhZj5bt18LkJlzIuJF4EOUN8DzWnMtM/O6iLi0oeyh9thek5lvdUjfg3KDu6V0DAFlntQWwPPAa2033pnA1zplHhFrUgKKMW35jALGAs8Noz5Qhmfv3CF9DeDiiNgeeBvYmBJwPDFIXi1N9ZgA3JyZ/1c/X0F5qFLvsZ0u3nbanu+plJcNHwAOpfSkjQWubStn1ZrWeuhv9QjPBDaOiNWGUM/5wHVtRY8BLo8yv/otYD0g6PyCQr3vDOCOiLislVCDyR0ov+1k5uM14N0N+F7d7Te+i5k5v0P+t2eHOc314wrAyRFxACVgXheY13bsvfnu+gUzKSNXAPYBvpWZbwJvRsRVlJdMAzX9hvQP+O9TmTmjYd+PAWvX3mcobfLhtu03tOodEQ9Sfhdv65DPQZR7749rO+2j1Hc49ZGWOoNmaWSZDnwxM2+IsojHPMpQwpb2G/MCShseztDlhyk3/4F2pQyzbHmt4fg+ykJbewUAOnwAAAa6SURBVA/cUIdDvt7h/DpZgXLTHl9vlu35bN/5kGH7CvAscHRmvhURt/LeazmYpnr08e7DhpZfttPhtdOfUHrOAMjMKZQe+FmUlwl9wPO1N6vJ/Hrsgvrg3bqmg9XzV5nZ3l4vpvSC/XFm9kfEkwz9N0E9pg4Pvhk4qS15YcEmdP4uDtfhlOBwr8x8NSJOowx1fk8Z1cB70FA8BawSEVsN6G3eHXi6DrGG5t+QVlmfzcwfNWxvOsdO+VyemWc2bJOWCQ7PlkaW0cDT9e/HMLRFcn4KrB4RewPUt8JNw4YvBPaJiE+2EiLiC5S3vtOHUNZ9wJYR8c6CQxExPiIWduN7BVijDsMkM1+lzKM6tS2fTSJiw2HWZzCjgWdqwDyOsvBQ+/ksSp53Ah+NiNYc5qMWIQ8t+2ynw6vPRZShqu8svBQRK/LudUtg3oDtW0XE2gxuuPUcDcypAfO+lB5pLd8mU6YljIJ35to/RP1tr1MEtmeQ6QWLaDTlRdGrEbEOJYgeih8CR0bEShGxetNxtQf4XOCS1hSEiNiEMjR68hDLuhE4qZZDRIyKiLFDOG7g/fV7wKcj4kM1nxUjYqfh1EcaCQyapZHlROCGiLgH2BT45cIOyMzXgU8BF0XE/cDOlIWtOu37AmWY159HxH9FxM8oweR+A3uSGo5/kTJk6+8i4uGIeIJyAx70YbyWezXwSETcV5MnUeYjPxIRjwDfAUYPpz7VwDnNN9f0c2o9H6AMw7ur7ZhvA4fX/Yey2narHg8D5wEzIuJuyoImLw/1ePUM2+nw6jMX2Bv4eETMqT3MdwP/CjxYh5gfBBwWEbMj4jFKj/oqnfLrop6nUuYxzwAOAWYPlr96X2b+nHI/WK8teRJwRETMpsy1P7JtSs7iMg0YVb/r11Paw1BcSmlnjwE3Af8xyL5nU1bhv7+2jVuBCxvW8ehkCmXUywP1WtxDmTKxMO+5v2bmXZTF026MiIeBR4E/WoT6SEtVX3+/Iw0laagiYlTtgSMiJgNbZKbzmiVJknqUc5olaXimRMTvUXrBfgYcu5TPR5IkSe8je5olSZIkSWrgnGZJkiRJkhoYNEuSJEmS1MCgWZIkSZKkBgbNkiT1uPrvoPZHxKb18yUR8beLmNdjETFhcZ6fJEkjmQuBSZI0QkTEHGADYAEwD7gD+MvMfK7LfFcC3gQ2y8w5wzjuSuDnmXlGN+VLkrQss6dZkqSR5aDMXAvYmhJAXzBwhxoES5KkJcCbriRJI1BmPh8R1wPH1x7fXwObABOAwyLiNuBs4FBgNeDfgBMzcx5ARHwJOAl4GzizPe+BPcgRcSBwFrAF8AJwMrAeMAnoj4gTgfsyc7/aG/6ZzLw9IlYFzgU+CfQB3wVOycz5dQj3VcB5wGlAP/DXmTmtlnkA8LVap18BUzNzymK6fJIkLTb2NEuSNAJFxG9RgtGf1KRJlAB0FHA7MAXYBtiJEuxuSAl8iYj9KYHvvsBHgP0HKWc8cA1wOrAusCuQmXkpcDVwXmaulZn7dTj8dGB3YEdge2AXoH2u9IY1z02A44CLI2Ltuu0y4LjMHAWMBW4bynWRJGlJs6dZkqSR5YaIeBN4DfgR8EXgfODGzLwLICJeB44FdsjM52va2cC/UILlQ4ErMvPRum0yJQDv5Bjg8sy8pX5+rv4Zikm0zbmu5VxCCaahzKM+JzMXADfW894S+E/gLWDbiJidmS/WNEmSRhyDZkmSRpaDM/P29oSIAPjvtqQxwBrAA3UblOHRK0fECsDGvDcIbT92oA8DNy/iuW4MzGn7PKemtfyyBswt8yg95QCfoAwbnxIRj1CGdd+7iOchSdL7xqBZkqRlQ/s/d/E8ZY7zNpn5i4E7RsT/UoZEt3x4kHyfoQzvXliZncwFfht4rK2cuQs5BoDMfAA4KCJWBv6CMh96o6EcK0nSkuScZkmSljGZ+TbwTeCCiNgAICI+GBF/UHe5Djg6IraOiDUZsBDYAJcBfxoR+0XEChGxQUSMq9ueAzYf5NjpwBkRMSYi1q/lTF/Y+UfEKhExKSLWyczWUPS3F3acJElLg0GzJEnLplOAp4AZEfEKZXGwsQB1fvJUypzoJ+u2jjLzfuAo4KvAy8BMoDXm+zJg64h4KSJu6XD4OcAsYDbwCPAwdTGyITgSeLqe++eATw3xOEmSlqi+/v6FjbySJEmSJGn5ZE+zJEmSJEkNDJolSZIkSWpg0CxJkiRJUgODZkmSJEmSGhg0S5IkSZLUwKBZkiRJkqQGBs2SJEmSJDUwaJYkSZIkqYFBsyRJkiRJDf4fZ13SNbyRjTYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "evaluate_metrics(linear_learner_predictor, xTest, yTest)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# should be the name of directory you created to save your features data\n",
    "data_dir = 'activity_data'\n",
    "\n",
    "# set prefix, a descriptive name for a directory  \n",
    "prefix = 'activity_data'\n",
    "\n",
    "output_path = 's3://{}/{}'.format(bucket, prefix)\n",
    "\n",
    "input_data = sagemaker_session.upload_data(path=data_dir, bucket=bucket, key_prefix=prefix)\n",
    "print(input_data)\n",
    "\n",
    "# upload all data to S3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        Specs        Score\n",
      "42       XVAR  1208.759578\n",
      "43       YVAR  1139.774652\n",
      "39  XSTANDDEV  1083.781498\n",
      "36  XABSOLDEV  1070.791139\n",
      "7          X7   970.931499\n",
      "40  YSTANDDEV   906.095967\n",
      "37  YABSOLDEV   902.136731\n",
      "44       ZVAR   881.935620\n",
      "6          X6   760.882796\n",
      "24         Z4   713.046560\n",
      "selected_features ['XVAR', 'YVAR', 'XSTANDDEV', 'XABSOLDEV', 'X7', 'YSTANDDEV', 'YABSOLDEV', 'ZVAR', 'X6', 'Z4']\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.feature_selection import SelectKBest\n",
    "from sklearn.feature_selection import chi2\n",
    "\n",
    "\n",
    "\n",
    "X = df  #independent columns\n",
    "y = activity   #target column i.e activity\n",
    "\n",
    "#apply SelectKBest class to extract top 10 best features\n",
    "bestfeatures = SelectKBest(score_func=chi2, k=10)\n",
    "fit = bestfeatures.fit(X,y)\n",
    "dfscores = pd.DataFrame(fit.scores_)\n",
    "dfcolumns = pd.DataFrame(X.columns)\n",
    "#concat two dataframes for better visualization \n",
    "featureScores = pd.concat([dfcolumns,dfscores],axis=1)\n",
    "featureScores.columns = ['Specs','Score']  #naming the dataframe columns\n",
    "tenlarget = featureScores.nlargest(10,'Score')  #print 10 best features\n",
    "print(tenlarget)\n",
    "selected_features = tenlarget['Specs'].tolist()\n",
    "print('selected_features', selected_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = df.filter(selected_features).to_numpy(dtype=np.float32)\n",
    "#filter the dataframme to get only the 10 best columns values\n",
    "activity_labels = activity.to_numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_csv(x, y, filename, data_dir):\n",
    "    '''Merges features and labels and converts them into one csv file with labels in the first column.\n",
    "       :param x: Data features\n",
    "       :param y: Data labels\n",
    "       :param file_name: Name of csv file, ex. 'train.csv'\n",
    "       :param data_dir: The directory where files will be saved\n",
    "       '''\n",
    "    # make data dir, if it does not exist\n",
    "    if not os.path.exists(data_dir):\n",
    "        os.makedirs(data_dir)\n",
    "    \n",
    "    \n",
    "    # your code here\n",
    "    pd.concat([pd.DataFrame(y), pd.DataFrame(x)], axis=1)\\\n",
    "             .to_csv(os.path.join(data_dir, filename), header=False, index=False)\n",
    "    \n",
    "    # nothing is returned, but a print statement indicates that the function has run\n",
    "    print('Path created: '+str(data_dir)+'/'+str(filename))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "xTrain, xTest, yTrain, yTest = train_test_split(features, activity_labels, test_size = 0.2, random_state = 0)\n",
    "#get the train and test split for the SVM models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training size:  60079\n",
      "Test size:  15020\n",
      "\n",
      "Training df sample: \n",
      " [[0.02416954 0.02283454 0.00084125 0.00090211 0.         0.00071396\n",
      "  0.00070293 0.0115161  0.         0.        ]\n",
      " [0.11350783 0.09438273 0.0139809  0.01551688 0.         0.00964575\n",
      "  0.00980561 0.1113731  0.         0.        ]\n",
      " [0.02822756 0.03661668 0.0010958  0.00109826 0.         0.00164519\n",
      "  0.00164558 0.03683381 0.         0.        ]\n",
      " [0.52763885 0.4936828  0.2811205  0.27412927 0.01052632 0.24587977\n",
      "  0.23138976 0.4623356  0.025      0.06      ]\n",
      " [0.02220703 0.02011587 0.00072984 0.00079059 0.         0.00057474\n",
      "  0.00058996 0.01154429 0.         0.        ]\n",
      " [0.09020344 0.02507647 0.00903122 0.00374016 0.         0.0008398\n",
      "  0.00075177 0.04858614 0.         0.        ]\n",
      " [0.24410373 0.2724036  0.06159811 0.05826021 0.         0.07591432\n",
      "  0.07669707 0.26938123 0.         0.        ]\n",
      " [0.4515078  0.36770114 0.2065596  0.24495876 0.         0.13721101\n",
      "  0.14603937 0.3466096  0.         0.57      ]\n",
      " [0.37805337 0.21404313 0.14548798 0.12382603 0.         0.04726629\n",
      "  0.04250152 0.3368904  0.01       0.045     ]\n",
      " [0.09833475 0.0765732  0.01063616 0.00848946 0.         0.00647367\n",
      "  0.00418777 0.11958249 0.         0.        ]]\n"
     ]
    }
   ],
   "source": [
    "# check that division of samples seems correct\n",
    "# these should add up to 95 (100 - 5 original files)\n",
    "print('Training size: ', len(xTrain))\n",
    "print('Test size: ', len(xTest))\n",
    "print()\n",
    "print('Training df sample: \\n', xTrain[:10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Path created: activity_data/train.csv\n",
      "Path created: activity_data/test.csv\n"
     ]
    }
   ],
   "source": [
    "data_dir = 'activity_data'\n",
    "\n",
    "make_csv(xTrain, yTrain, filename='train.csv', data_dir=data_dir)\n",
    "make_csv(xTest, yTest, filename='test.csv', data_dir=data_dir)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Upload Data to S3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "s3://sagemaker-ap-south-1-724515668840/activity_data\n"
     ]
    }
   ],
   "source": [
    "# should be the name of directory you created to save your features data\n",
    "data_dir = 'activity_data'\n",
    "\n",
    "# set prefix, a descriptive name for a directory  \n",
    "prefix = 'activity_data'\n",
    "\n",
    "input_data = sagemaker_session.upload_data(path=data_dir, bucket=bucket, key_prefix=prefix)\n",
    "print(input_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LinearSVC Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sagemaker.sklearn.estimator import SKLearn\n",
    "\n",
    "output_path = 's3://{}/{}'.format(bucket, prefix)\n",
    "\n",
    "linear_SVC_estimator = SKLearn(entry_point='train.py',\n",
    "                            role=role,\n",
    "                            train_instance_type='ml.m4.4xlarge',\n",
    "                            train_instance_count=1,\n",
    "                            output_path=output_path,\n",
    "                            framework_version='0.20.0',\n",
    "                            sagemaker_session=sagemaker_session,\n",
    "                            hyperparameters = { 'tol': 0.0001 })\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2020-01-24 18:29:01 Starting - Starting the training job...\n",
      "2020-01-24 18:29:03 Starting - Launching requested ML instances...\n",
      "2020-01-24 18:29:57 Starting - Preparing the instances for training...\n",
      "2020-01-24 18:30:31 Downloading - Downloading input data...\n",
      "2020-01-24 18:30:47 Training - Downloading the training image..\u001b[34m2020-01-24 18:31:07,942 sagemaker-containers INFO     Imported framework sagemaker_sklearn_container.training\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:07,945 sagemaker-containers INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:07,958 sagemaker_sklearn_container.training INFO     Invoking user training script.\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:08,197 sagemaker-containers INFO     Module train does not provide a setup.py. \u001b[0m\n",
      "\u001b[34mGenerating setup.py\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:08,198 sagemaker-containers INFO     Generating setup.cfg\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:08,198 sagemaker-containers INFO     Generating MANIFEST.in\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:08,198 sagemaker-containers INFO     Installing module with the following command:\u001b[0m\n",
      "\u001b[34m/miniconda3/bin/python -m pip install . \u001b[0m\n",
      "\u001b[34mProcessing /opt/ml/code\u001b[0m\n",
      "\u001b[34mBuilding wheels for collected packages: train\n",
      "  Building wheel for train (setup.py): started\n",
      "  Building wheel for train (setup.py): finished with status 'done'\n",
      "  Created wheel for train: filename=train-1.0.0-py2.py3-none-any.whl size=6312 sha256=773744abf32e082b09c294888418080a7c72ac10698a21660d6e077d3d1cdf95\n",
      "  Stored in directory: /tmp/pip-ephem-wheel-cache-0osbv1bo/wheels/35/24/16/37574d11bf9bde50616c67372a334f94fa8356bc7164af8ca3\u001b[0m\n",
      "\u001b[34mSuccessfully built train\u001b[0m\n",
      "\u001b[34mInstalling collected packages: train\u001b[0m\n",
      "\u001b[34mSuccessfully installed train-1.0.0\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:09,577 sagemaker-containers INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:09,589 sagemaker-containers INFO     Invoking user script\n",
      "\u001b[0m\n",
      "\u001b[34mTraining Env:\n",
      "\u001b[0m\n",
      "\u001b[34m{\n",
      "    \"additional_framework_parameters\": {},\n",
      "    \"channel_input_dirs\": {\n",
      "        \"train\": \"/opt/ml/input/data/train\"\n",
      "    },\n",
      "    \"current_host\": \"algo-1\",\n",
      "    \"framework_module\": \"sagemaker_sklearn_container.training:main\",\n",
      "    \"hosts\": [\n",
      "        \"algo-1\"\n",
      "    ],\n",
      "    \"hyperparameters\": {\n",
      "        \"tol\": 0.0001\n",
      "    },\n",
      "    \"input_config_dir\": \"/opt/ml/input/config\",\n",
      "    \"input_data_config\": {\n",
      "        \"train\": {\n",
      "            \"TrainingInputMode\": \"File\",\n",
      "            \"S3DistributionType\": \"FullyReplicated\",\n",
      "            \"RecordWrapperType\": \"None\"\n",
      "        }\n",
      "    },\n",
      "    \"input_dir\": \"/opt/ml/input\",\n",
      "    \"is_master\": true,\n",
      "    \"job_name\": \"sagemaker-scikit-learn-2020-01-24-18-29-01-252\",\n",
      "    \"log_level\": 20,\n",
      "    \"master_hostname\": \"algo-1\",\n",
      "    \"model_dir\": \"/opt/ml/model\",\n",
      "    \"module_dir\": \"s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-18-29-01-252/source/sourcedir.tar.gz\",\n",
      "    \"module_name\": \"train\",\n",
      "    \"network_interface_name\": \"eth0\",\n",
      "    \"num_cpus\": 16,\n",
      "    \"num_gpus\": 0,\n",
      "    \"output_data_dir\": \"/opt/ml/output/data\",\n",
      "    \"output_dir\": \"/opt/ml/output\",\n",
      "    \"output_intermediate_dir\": \"/opt/ml/output/intermediate\",\n",
      "    \"resource_config\": {\n",
      "        \"current_host\": \"algo-1\",\n",
      "        \"hosts\": [\n",
      "            \"algo-1\"\n",
      "        ],\n",
      "        \"network_interface_name\": \"eth0\"\n",
      "    },\n",
      "    \"user_entry_point\": \"train.py\"\u001b[0m\n",
      "\u001b[34m}\n",
      "\u001b[0m\n",
      "\u001b[34mEnvironment variables:\n",
      "\u001b[0m\n",
      "\u001b[34mSM_HOSTS=[\"algo-1\"]\u001b[0m\n",
      "\u001b[34mSM_NETWORK_INTERFACE_NAME=eth0\u001b[0m\n",
      "\u001b[34mSM_HPS={\"tol\":0.0001}\u001b[0m\n",
      "\u001b[34mSM_USER_ENTRY_POINT=train.py\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_PARAMS={}\u001b[0m\n",
      "\u001b[34mSM_RESOURCE_CONFIG={\"current_host\":\"algo-1\",\"hosts\":[\"algo-1\"],\"network_interface_name\":\"eth0\"}\u001b[0m\n",
      "\u001b[34mSM_INPUT_DATA_CONFIG={\"train\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}}\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DATA_DIR=/opt/ml/output/data\u001b[0m\n",
      "\u001b[34mSM_CHANNELS=[\"train\"]\u001b[0m\n",
      "\u001b[34mSM_CURRENT_HOST=algo-1\u001b[0m\n",
      "\u001b[34mSM_MODULE_NAME=train\u001b[0m\n",
      "\u001b[34mSM_LOG_LEVEL=20\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_MODULE=sagemaker_sklearn_container.training:main\u001b[0m\n",
      "\u001b[34mSM_INPUT_DIR=/opt/ml/input\u001b[0m\n",
      "\u001b[34mSM_INPUT_CONFIG_DIR=/opt/ml/input/config\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DIR=/opt/ml/output\u001b[0m\n",
      "\u001b[34mSM_NUM_CPUS=16\u001b[0m\n",
      "\u001b[34mSM_NUM_GPUS=0\u001b[0m\n",
      "\u001b[34mSM_MODEL_DIR=/opt/ml/model\u001b[0m\n",
      "\u001b[34mSM_MODULE_DIR=s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-18-29-01-252/source/sourcedir.tar.gz\u001b[0m\n",
      "\u001b[34mSM_TRAINING_ENV={\"additional_framework_parameters\":{},\"channel_input_dirs\":{\"train\":\"/opt/ml/input/data/train\"},\"current_host\":\"algo-1\",\"framework_module\":\"sagemaker_sklearn_container.training:main\",\"hosts\":[\"algo-1\"],\"hyperparameters\":{\"tol\":0.0001},\"input_config_dir\":\"/opt/ml/input/config\",\"input_data_config\":{\"train\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}},\"input_dir\":\"/opt/ml/input\",\"is_master\":true,\"job_name\":\"sagemaker-scikit-learn-2020-01-24-18-29-01-252\",\"log_level\":20,\"master_hostname\":\"algo-1\",\"model_dir\":\"/opt/ml/model\",\"module_dir\":\"s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-18-29-01-252/source/sourcedir.tar.gz\",\"module_name\":\"train\",\"network_interface_name\":\"eth0\",\"num_cpus\":16,\"num_gpus\":0,\"output_data_dir\":\"/opt/ml/output/data\",\"output_dir\":\"/opt/ml/output\",\"output_intermediate_dir\":\"/opt/ml/output/intermediate\",\"resource_config\":{\"current_host\":\"algo-1\",\"hosts\":[\"algo-1\"],\"network_interface_name\":\"eth0\"},\"user_entry_point\":\"train.py\"}\u001b[0m\n",
      "\u001b[34mSM_USER_ARGS=[\"--tol\",\"0.0001\"]\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_INTERMEDIATE_DIR=/opt/ml/output/intermediate\u001b[0m\n",
      "\u001b[34mSM_CHANNEL_TRAIN=/opt/ml/input/data/train\u001b[0m\n",
      "\u001b[34mSM_HP_TOL=0.0001\u001b[0m\n",
      "\u001b[34mPYTHONPATH=/miniconda3/bin:/miniconda3/lib/python37.zip:/miniconda3/lib/python3.7:/miniconda3/lib/python3.7/lib-dynload:/miniconda3/lib/python3.7/site-packages\n",
      "\u001b[0m\n",
      "\u001b[34mInvoking script with the following command:\n",
      "\u001b[0m\n",
      "\u001b[34m/miniconda3/bin/python -m train --tol 0.0001\n",
      "\n",
      "\u001b[0m\n",
      "\u001b[34m/miniconda3/lib/python3.7/site-packages/sklearn/externals/joblib/externals/cloudpickle/cloudpickle.py:47: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  import imp\u001b[0m\n",
      "\u001b[34m2020-01-24 18:31:13,263 sagemaker-containers INFO     Reporting training SUCCESS\u001b[0m\n",
      "\n",
      "2020-01-24 18:31:24 Uploading - Uploading generated training model\n",
      "2020-01-24 18:31:24 Completed - Training job completed\n",
      "Training seconds: 53\n",
      "Billable seconds: 53\n",
      "CPU times: user 485 ms, sys: 28.3 ms, total: 513 ms\n",
      "Wall time: 2min 41s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# Train your estimator on S3 training data\n",
    "linear_SVC_estimator.fit({'train': input_data})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------------------------!CPU times: user 336 ms, sys: 18.6 ms, total: 355 ms\n",
      "Wall time: 5min 23s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# deploy your model to create a predictor\n",
    "linear_SVC_predictor = linear_SVC_estimator.deploy(instance_type='ml.m4.xlarge',\n",
    "                                     initial_instance_count=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read in test data, assuming it is stored locally\n",
    "test_data = pd.read_csv(os.path.join(data_dir, \"test.csv\"), header=None, names=None)\n",
    "\n",
    "# labels are in the first column\n",
    "test_y = test_data.iloc[:,0]\n",
    "test_x = test_data.iloc[:,1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_y_preds = linear_SVC_predictor.predict(test_x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_metrics(test_labels, test_preds):\n",
    "    \"\"\"\n",
    "    Evaluate a model on a test set using the given prediction endpoint. Display classification metrics.\n",
    "    \"\"\"\n",
    "    # calculate accuracy\n",
    "    accuracy = (test_labels == test_preds).sum() / test_labels.shape[0]\n",
    "    \n",
    "    # calculate confusion matrix\n",
    "    label_mapper = np.vectorize(lambda x: activty_map[x])\n",
    "    confusion_matrix = pd.crosstab(label_mapper(test_labels), label_mapper(test_preds), \n",
    "                                   rownames=['Actuals'], colnames=['Predictions'], normalize='index')\n",
    "\n",
    "    # display results\n",
    "    plt.figure(figsize=(16, 6))\n",
    "    sns.heatmap(confusion_matrix, annot=True, fmt='.2f', cmap=\"YlGnBu\").set_title('Confusion Matrix')  \n",
    "    \n",
    "    print('Accuracy: {:.3f}'.format(accuracy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.467\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "evaluate_metrics(test_y, test_y_preds)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sagemaker.sklearn.estimator import SKLearn\n",
    "\n",
    "output_path = 's3://{}/{}'.format(bucket, prefix)\n",
    "\n",
    "KNN_estimator = SKLearn(entry_point='train_knn.py',\n",
    "                            role=role,\n",
    "                            train_instance_type='ml.m4.4xlarge',\n",
    "                            train_instance_count=1,\n",
    "                            output_path=output_path,\n",
    "                            framework_version='0.20.0',\n",
    "                            sagemaker_session=sagemaker_session)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2020-01-24 18:53:38 Starting - Starting the training job...\n",
      "2020-01-24 18:53:41 Starting - Launching requested ML instances...\n",
      "2020-01-24 18:54:35 Starting - Preparing the instances for training......\n",
      "2020-01-24 18:55:35 Downloading - Downloading input data\n",
      "2020-01-24 18:55:35 Training - Downloading the training image..\u001b[34m2020-01-24 18:55:48,899 sagemaker-containers INFO     Imported framework sagemaker_sklearn_container.training\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:48,902 sagemaker-containers INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:48,914 sagemaker_sklearn_container.training INFO     Invoking user training script.\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:49,150 sagemaker-containers INFO     Module train_knn does not provide a setup.py. \u001b[0m\n",
      "\u001b[34mGenerating setup.py\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:49,150 sagemaker-containers INFO     Generating setup.cfg\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:49,150 sagemaker-containers INFO     Generating MANIFEST.in\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:49,151 sagemaker-containers INFO     Installing module with the following command:\u001b[0m\n",
      "\u001b[34m/miniconda3/bin/python -m pip install . \u001b[0m\n",
      "\u001b[34mProcessing /opt/ml/code\u001b[0m\n",
      "\u001b[34mBuilding wheels for collected packages: train-knn\n",
      "  Building wheel for train-knn (setup.py): started\n",
      "  Building wheel for train-knn (setup.py): finished with status 'done'\n",
      "  Created wheel for train-knn: filename=train_knn-1.0.0-py2.py3-none-any.whl size=6120 sha256=ebd8d3cb3a488f27d7acfdbe5a2fd1c98c410b75fb70d7f67e1339cc6dd63791\n",
      "  Stored in directory: /tmp/pip-ephem-wheel-cache-re0r5sls/wheels/35/24/16/37574d11bf9bde50616c67372a334f94fa8356bc7164af8ca3\u001b[0m\n",
      "\u001b[34mSuccessfully built train-knn\u001b[0m\n",
      "\u001b[34mInstalling collected packages: train-knn\u001b[0m\n",
      "\u001b[34mSuccessfully installed train-knn-1.0.0\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:50,479 sagemaker-containers INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:50,491 sagemaker-containers INFO     Invoking user script\n",
      "\u001b[0m\n",
      "\u001b[34mTraining Env:\n",
      "\u001b[0m\n",
      "\u001b[34m{\n",
      "    \"additional_framework_parameters\": {},\n",
      "    \"channel_input_dirs\": {\n",
      "        \"train\": \"/opt/ml/input/data/train\"\n",
      "    },\n",
      "    \"current_host\": \"algo-1\",\n",
      "    \"framework_module\": \"sagemaker_sklearn_container.training:main\",\n",
      "    \"hosts\": [\n",
      "        \"algo-1\"\n",
      "    ],\n",
      "    \"hyperparameters\": {},\n",
      "    \"input_config_dir\": \"/opt/ml/input/config\",\n",
      "    \"input_data_config\": {\n",
      "        \"train\": {\n",
      "            \"TrainingInputMode\": \"File\",\n",
      "            \"S3DistributionType\": \"FullyReplicated\",\n",
      "            \"RecordWrapperType\": \"None\"\n",
      "        }\n",
      "    },\n",
      "    \"input_dir\": \"/opt/ml/input\",\n",
      "    \"is_master\": true,\n",
      "    \"job_name\": \"sagemaker-scikit-learn-2020-01-24-18-53-38-114\",\n",
      "    \"log_level\": 20,\n",
      "    \"master_hostname\": \"algo-1\",\n",
      "    \"model_dir\": \"/opt/ml/model\",\n",
      "    \"module_dir\": \"s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-18-53-38-114/source/sourcedir.tar.gz\",\n",
      "    \"module_name\": \"train_knn\",\n",
      "    \"network_interface_name\": \"eth0\",\n",
      "    \"num_cpus\": 16,\n",
      "    \"num_gpus\": 0,\n",
      "    \"output_data_dir\": \"/opt/ml/output/data\",\n",
      "    \"output_dir\": \"/opt/ml/output\",\n",
      "    \"output_intermediate_dir\": \"/opt/ml/output/intermediate\",\n",
      "    \"resource_config\": {\n",
      "        \"current_host\": \"algo-1\",\n",
      "        \"hosts\": [\n",
      "            \"algo-1\"\n",
      "        ],\n",
      "        \"network_interface_name\": \"eth0\"\n",
      "    },\n",
      "    \"user_entry_point\": \"train_knn.py\"\u001b[0m\n",
      "\u001b[34m}\n",
      "\u001b[0m\n",
      "\u001b[34mEnvironment variables:\n",
      "\u001b[0m\n",
      "\u001b[34mSM_HOSTS=[\"algo-1\"]\u001b[0m\n",
      "\u001b[34mSM_NETWORK_INTERFACE_NAME=eth0\u001b[0m\n",
      "\u001b[34mSM_HPS={}\u001b[0m\n",
      "\u001b[34mSM_USER_ENTRY_POINT=train_knn.py\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_PARAMS={}\u001b[0m\n",
      "\u001b[34mSM_RESOURCE_CONFIG={\"current_host\":\"algo-1\",\"hosts\":[\"algo-1\"],\"network_interface_name\":\"eth0\"}\u001b[0m\n",
      "\u001b[34mSM_INPUT_DATA_CONFIG={\"train\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}}\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DATA_DIR=/opt/ml/output/data\u001b[0m\n",
      "\u001b[34mSM_CHANNELS=[\"train\"]\u001b[0m\n",
      "\u001b[34mSM_CURRENT_HOST=algo-1\u001b[0m\n",
      "\u001b[34mSM_MODULE_NAME=train_knn\u001b[0m\n",
      "\u001b[34mSM_LOG_LEVEL=20\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_MODULE=sagemaker_sklearn_container.training:main\u001b[0m\n",
      "\u001b[34mSM_INPUT_DIR=/opt/ml/input\u001b[0m\n",
      "\u001b[34mSM_INPUT_CONFIG_DIR=/opt/ml/input/config\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DIR=/opt/ml/output\u001b[0m\n",
      "\u001b[34mSM_NUM_CPUS=16\u001b[0m\n",
      "\u001b[34mSM_NUM_GPUS=0\u001b[0m\n",
      "\u001b[34mSM_MODEL_DIR=/opt/ml/model\u001b[0m\n",
      "\u001b[34mSM_MODULE_DIR=s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-18-53-38-114/source/sourcedir.tar.gz\u001b[0m\n",
      "\u001b[34mSM_TRAINING_ENV={\"additional_framework_parameters\":{},\"channel_input_dirs\":{\"train\":\"/opt/ml/input/data/train\"},\"current_host\":\"algo-1\",\"framework_module\":\"sagemaker_sklearn_container.training:main\",\"hosts\":[\"algo-1\"],\"hyperparameters\":{},\"input_config_dir\":\"/opt/ml/input/config\",\"input_data_config\":{\"train\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}},\"input_dir\":\"/opt/ml/input\",\"is_master\":true,\"job_name\":\"sagemaker-scikit-learn-2020-01-24-18-53-38-114\",\"log_level\":20,\"master_hostname\":\"algo-1\",\"model_dir\":\"/opt/ml/model\",\"module_dir\":\"s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-18-53-38-114/source/sourcedir.tar.gz\",\"module_name\":\"train_knn\",\"network_interface_name\":\"eth0\",\"num_cpus\":16,\"num_gpus\":0,\"output_data_dir\":\"/opt/ml/output/data\",\"output_dir\":\"/opt/ml/output\",\"output_intermediate_dir\":\"/opt/ml/output/intermediate\",\"resource_config\":{\"current_host\":\"algo-1\",\"hosts\":[\"algo-1\"],\"network_interface_name\":\"eth0\"},\"user_entry_point\":\"train_knn.py\"}\u001b[0m\n",
      "\u001b[34mSM_USER_ARGS=[]\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_INTERMEDIATE_DIR=/opt/ml/output/intermediate\u001b[0m\n",
      "\u001b[34mSM_CHANNEL_TRAIN=/opt/ml/input/data/train\u001b[0m\n",
      "\u001b[34mPYTHONPATH=/miniconda3/bin:/miniconda3/lib/python37.zip:/miniconda3/lib/python3.7:/miniconda3/lib/python3.7/lib-dynload:/miniconda3/lib/python3.7/site-packages\n",
      "\u001b[0m\n",
      "\u001b[34mInvoking script with the following command:\n",
      "\u001b[0m\n",
      "\u001b[34m/miniconda3/bin/python -m train_knn\n",
      "\n",
      "\u001b[0m\n",
      "\u001b[34m/miniconda3/lib/python3.7/site-packages/sklearn/externals/joblib/externals/cloudpickle/cloudpickle.py:47: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  import imp\u001b[0m\n",
      "\u001b[34m2020-01-24 18:55:53,708 sagemaker-containers INFO     Reporting training SUCCESS\u001b[0m\n",
      "\n",
      "2020-01-24 18:56:05 Uploading - Uploading generated training model\n",
      "2020-01-24 18:56:05 Completed - Training job completed\n",
      "Training seconds: 46\n",
      "Billable seconds: 46\n",
      "CPU times: user 325 ms, sys: 24.9 ms, total: 350 ms\n",
      "Wall time: 2min 41s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# Train your estimator on S3 training data\n",
    "KNN_estimator.fit({'train': input_data})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-------------------------------------------------------------!CPU times: user 320 ms, sys: 11.3 ms, total: 331 ms\n",
      "Wall time: 5min 7s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# deploy your model to create a predictor\n",
    "KNN_predictor = KNN_estimator.deploy(instance_type='ml.m4.xlarge',\n",
    "                                     initial_instance_count=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_y_preds = KNN_predictor.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.613\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "evaluate_metrics(test_y, test_y_preds)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KNeighborsClassifier - Hyperparameter Tuning(n_neighbors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "KNN_estimator = SKLearn(entry_point='train_knn.py',\n",
    "                            role=role,\n",
    "                            train_instance_type='ml.m4.4xlarge',\n",
    "                            train_instance_count=1,\n",
    "                            output_path=output_path,\n",
    "                            framework_version='0.20.0',\n",
    "                            sagemaker_session=sagemaker_session,\n",
    "                            hyperparameters = { 'n_neighbors': 3 })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2020-01-24 19:08:18 Starting - Starting the training job...\n",
      "2020-01-24 19:08:21 Starting - Launching requested ML instances......\n",
      "2020-01-24 19:09:42 Starting - Preparing the instances for training......\n",
      "2020-01-24 19:10:43 Downloading - Downloading input data\n",
      "2020-01-24 19:10:43 Training - Downloading the training image..\u001b[34m2020-01-24 19:10:56,336 sagemaker-containers INFO     Imported framework sagemaker_sklearn_container.training\u001b[0m\n",
      "\u001b[34m2020-01-24 19:10:56,338 sagemaker-containers INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2020-01-24 19:10:56,350 sagemaker_sklearn_container.training INFO     Invoking user training script.\u001b[0m\n",
      "\u001b[34m2020-01-24 19:10:56,599 sagemaker-containers INFO     Module train_knn does not provide a setup.py. \u001b[0m\n",
      "\u001b[34mGenerating setup.py\u001b[0m\n",
      "\u001b[34m2020-01-24 19:10:56,599 sagemaker-containers INFO     Generating setup.cfg\u001b[0m\n",
      "\u001b[34m2020-01-24 19:10:56,599 sagemaker-containers INFO     Generating MANIFEST.in\u001b[0m\n",
      "\u001b[34m2020-01-24 19:10:56,599 sagemaker-containers INFO     Installing module with the following command:\u001b[0m\n",
      "\u001b[34m/miniconda3/bin/python -m pip install . \u001b[0m\n",
      "\u001b[34mProcessing /opt/ml/code\u001b[0m\n",
      "\u001b[34mBuilding wheels for collected packages: train-knn\n",
      "  Building wheel for train-knn (setup.py): started\u001b[0m\n",
      "\u001b[34m  Building wheel for train-knn (setup.py): finished with status 'done'\n",
      "  Created wheel for train-knn: filename=train_knn-1.0.0-py2.py3-none-any.whl size=6118 sha256=08a371dded8d404b2c420bd91921a2dabd3d78e723face827d13c96d3a5aac8b\n",
      "  Stored in directory: /tmp/pip-ephem-wheel-cache-hy0tugq1/wheels/35/24/16/37574d11bf9bde50616c67372a334f94fa8356bc7164af8ca3\u001b[0m\n",
      "\u001b[34mSuccessfully built train-knn\u001b[0m\n",
      "\u001b[34mInstalling collected packages: train-knn\u001b[0m\n",
      "\u001b[34mSuccessfully installed train-knn-1.0.0\u001b[0m\n",
      "\u001b[34m2020-01-24 19:10:57,945 sagemaker-containers INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2020-01-24 19:10:57,957 sagemaker-containers INFO     Invoking user script\n",
      "\u001b[0m\n",
      "\u001b[34mTraining Env:\n",
      "\u001b[0m\n",
      "\u001b[34m{\n",
      "    \"additional_framework_parameters\": {},\n",
      "    \"channel_input_dirs\": {\n",
      "        \"train\": \"/opt/ml/input/data/train\"\n",
      "    },\n",
      "    \"current_host\": \"algo-1\",\n",
      "    \"framework_module\": \"sagemaker_sklearn_container.training:main\",\n",
      "    \"hosts\": [\n",
      "        \"algo-1\"\n",
      "    ],\n",
      "    \"hyperparameters\": {\n",
      "        \"n_neighbors\": 3\n",
      "    },\n",
      "    \"input_config_dir\": \"/opt/ml/input/config\",\n",
      "    \"input_data_config\": {\n",
      "        \"train\": {\n",
      "            \"TrainingInputMode\": \"File\",\n",
      "            \"S3DistributionType\": \"FullyReplicated\",\n",
      "            \"RecordWrapperType\": \"None\"\n",
      "        }\n",
      "    },\n",
      "    \"input_dir\": \"/opt/ml/input\",\n",
      "    \"is_master\": true,\n",
      "    \"job_name\": \"sagemaker-scikit-learn-2020-01-24-19-08-18-062\",\n",
      "    \"log_level\": 20,\n",
      "    \"master_hostname\": \"algo-1\",\n",
      "    \"model_dir\": \"/opt/ml/model\",\n",
      "    \"module_dir\": \"s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-19-08-18-062/source/sourcedir.tar.gz\",\n",
      "    \"module_name\": \"train_knn\",\n",
      "    \"network_interface_name\": \"eth0\",\n",
      "    \"num_cpus\": 16,\n",
      "    \"num_gpus\": 0,\n",
      "    \"output_data_dir\": \"/opt/ml/output/data\",\n",
      "    \"output_dir\": \"/opt/ml/output\",\n",
      "    \"output_intermediate_dir\": \"/opt/ml/output/intermediate\",\n",
      "    \"resource_config\": {\n",
      "        \"current_host\": \"algo-1\",\n",
      "        \"hosts\": [\n",
      "            \"algo-1\"\n",
      "        ],\n",
      "        \"network_interface_name\": \"eth0\"\n",
      "    },\n",
      "    \"user_entry_point\": \"train_knn.py\"\u001b[0m\n",
      "\u001b[34m}\n",
      "\u001b[0m\n",
      "\u001b[34mEnvironment variables:\n",
      "\u001b[0m\n",
      "\u001b[34mSM_HOSTS=[\"algo-1\"]\u001b[0m\n",
      "\u001b[34mSM_NETWORK_INTERFACE_NAME=eth0\u001b[0m\n",
      "\u001b[34mSM_HPS={\"n_neighbors\":3}\u001b[0m\n",
      "\u001b[34mSM_USER_ENTRY_POINT=train_knn.py\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_PARAMS={}\u001b[0m\n",
      "\u001b[34mSM_RESOURCE_CONFIG={\"current_host\":\"algo-1\",\"hosts\":[\"algo-1\"],\"network_interface_name\":\"eth0\"}\u001b[0m\n",
      "\u001b[34mSM_INPUT_DATA_CONFIG={\"train\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}}\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DATA_DIR=/opt/ml/output/data\u001b[0m\n",
      "\u001b[34mSM_CHANNELS=[\"train\"]\u001b[0m\n",
      "\u001b[34mSM_CURRENT_HOST=algo-1\u001b[0m\n",
      "\u001b[34mSM_MODULE_NAME=train_knn\u001b[0m\n",
      "\u001b[34mSM_LOG_LEVEL=20\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_MODULE=sagemaker_sklearn_container.training:main\u001b[0m\n",
      "\u001b[34mSM_INPUT_DIR=/opt/ml/input\u001b[0m\n",
      "\u001b[34mSM_INPUT_CONFIG_DIR=/opt/ml/input/config\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DIR=/opt/ml/output\u001b[0m\n",
      "\u001b[34mSM_NUM_CPUS=16\u001b[0m\n",
      "\u001b[34mSM_NUM_GPUS=0\u001b[0m\n",
      "\u001b[34mSM_MODEL_DIR=/opt/ml/model\u001b[0m\n",
      "\u001b[34mSM_MODULE_DIR=s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-19-08-18-062/source/sourcedir.tar.gz\u001b[0m\n",
      "\u001b[34mSM_TRAINING_ENV={\"additional_framework_parameters\":{},\"channel_input_dirs\":{\"train\":\"/opt/ml/input/data/train\"},\"current_host\":\"algo-1\",\"framework_module\":\"sagemaker_sklearn_container.training:main\",\"hosts\":[\"algo-1\"],\"hyperparameters\":{\"n_neighbors\":3},\"input_config_dir\":\"/opt/ml/input/config\",\"input_data_config\":{\"train\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}},\"input_dir\":\"/opt/ml/input\",\"is_master\":true,\"job_name\":\"sagemaker-scikit-learn-2020-01-24-19-08-18-062\",\"log_level\":20,\"master_hostname\":\"algo-1\",\"model_dir\":\"/opt/ml/model\",\"module_dir\":\"s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-19-08-18-062/source/sourcedir.tar.gz\",\"module_name\":\"train_knn\",\"network_interface_name\":\"eth0\",\"num_cpus\":16,\"num_gpus\":0,\"output_data_dir\":\"/opt/ml/output/data\",\"output_dir\":\"/opt/ml/output\",\"output_intermediate_dir\":\"/opt/ml/output/intermediate\",\"resource_config\":{\"current_host\":\"algo-1\",\"hosts\":[\"algo-1\"],\"network_interface_name\":\"eth0\"},\"user_entry_point\":\"train_knn.py\"}\u001b[0m\n",
      "\u001b[34mSM_USER_ARGS=[\"--n_neighbors\",\"3\"]\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_INTERMEDIATE_DIR=/opt/ml/output/intermediate\u001b[0m\n",
      "\u001b[34mSM_CHANNEL_TRAIN=/opt/ml/input/data/train\u001b[0m\n",
      "\u001b[34mSM_HP_N_NEIGHBORS=3\u001b[0m\n",
      "\u001b[34mPYTHONPATH=/miniconda3/bin:/miniconda3/lib/python37.zip:/miniconda3/lib/python3.7:/miniconda3/lib/python3.7/lib-dynload:/miniconda3/lib/python3.7/site-packages\n",
      "\u001b[0m\n",
      "\u001b[34mInvoking script with the following command:\n",
      "\u001b[0m\n",
      "\u001b[34m/miniconda3/bin/python -m train_knn --n_neighbors 3\n",
      "\n",
      "\u001b[0m\n",
      "\u001b[34m/miniconda3/lib/python3.7/site-packages/sklearn/externals/joblib/externals/cloudpickle/cloudpickle.py:47: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  import imp\u001b[0m\n",
      "\u001b[34m2020-01-24 19:11:01,316 sagemaker-containers INFO     Reporting training SUCCESS\u001b[0m\n",
      "\n",
      "2020-01-24 19:11:12 Uploading - Uploading generated training model\n",
      "2020-01-24 19:11:12 Completed - Training job completed\n",
      "Training seconds: 44\n",
      "Billable seconds: 44\n",
      "CPU times: user 387 ms, sys: 13.2 ms, total: 400 ms\n",
      "Wall time: 3min 11s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# Train your estimator on S3 training data\n",
    "KNN_estimator.fit({'train': input_data})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-------------------------------------------------------------!CPU times: user 314 ms, sys: 17.9 ms, total: 332 ms\n",
      "Wall time: 5min 7s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# deploy your model to create a predictor\n",
    "KNN_predictor = KNN_estimator.deploy(instance_type='ml.m4.xlarge',\n",
    "                                     initial_instance_count=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_y_preds = KNN_predictor.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.598\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "evaluate_metrics(test_y, test_y_preds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "KNN_estimator = SKLearn(entry_point='train_knn.py',\n",
    "                            role=role,\n",
    "                            train_instance_type='ml.m4.4xlarge',\n",
    "                            train_instance_count=1,\n",
    "                            output_path=output_path,\n",
    "                            framework_version='0.20.0',\n",
    "                            sagemaker_session=sagemaker_session,\n",
    "                            hyperparameters = { 'n_neighbors': 13 })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2020-01-24 19:36:48 Starting - Starting the training job...\n",
      "2020-01-24 19:36:49 Starting - Launching requested ML instances...\n",
      "2020-01-24 19:37:44 Starting - Preparing the instances for training...\n",
      "2020-01-24 19:38:18 Downloading - Downloading input data......\n",
      "2020-01-24 19:39:12 Training - Training image download completed. Training in progress..\u001b[34m2020-01-24 19:39:12,786 sagemaker-containers INFO     Imported framework sagemaker_sklearn_container.training\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:12,789 sagemaker-containers INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:12,802 sagemaker_sklearn_container.training INFO     Invoking user training script.\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:13,013 sagemaker-containers INFO     Module train_knn does not provide a setup.py. \u001b[0m\n",
      "\u001b[34mGenerating setup.py\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:13,014 sagemaker-containers INFO     Generating setup.cfg\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:13,014 sagemaker-containers INFO     Generating MANIFEST.in\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:13,014 sagemaker-containers INFO     Installing module with the following command:\u001b[0m\n",
      "\u001b[34m/miniconda3/bin/python -m pip install . \u001b[0m\n",
      "\u001b[34mProcessing /opt/ml/code\u001b[0m\n",
      "\u001b[34mBuilding wheels for collected packages: train-knn\n",
      "  Building wheel for train-knn (setup.py): started\n",
      "  Building wheel for train-knn (setup.py): finished with status 'done'\n",
      "  Created wheel for train-knn: filename=train_knn-1.0.0-py2.py3-none-any.whl size=5976 sha256=fb73a6cd1150da3273acc2abf46aa865fb4fdf8d9931b75539864fe080e8b7e5\n",
      "  Stored in directory: /tmp/pip-ephem-wheel-cache-mv45nip3/wheels/35/24/16/37574d11bf9bde50616c67372a334f94fa8356bc7164af8ca3\u001b[0m\n",
      "\u001b[34mSuccessfully built train-knn\u001b[0m\n",
      "\u001b[34mInstalling collected packages: train-knn\u001b[0m\n",
      "\u001b[34mSuccessfully installed train-knn-1.0.0\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:14,337 sagemaker-containers INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:14,350 sagemaker-containers INFO     Invoking user script\n",
      "\u001b[0m\n",
      "\u001b[34mTraining Env:\n",
      "\u001b[0m\n",
      "\u001b[34m{\n",
      "    \"additional_framework_parameters\": {},\n",
      "    \"channel_input_dirs\": {\n",
      "        \"train\": \"/opt/ml/input/data/train\"\n",
      "    },\n",
      "    \"current_host\": \"algo-1\",\n",
      "    \"framework_module\": \"sagemaker_sklearn_container.training:main\",\n",
      "    \"hosts\": [\n",
      "        \"algo-1\"\n",
      "    ],\n",
      "    \"hyperparameters\": {\n",
      "        \"n_neighbors\": 13\n",
      "    },\n",
      "    \"input_config_dir\": \"/opt/ml/input/config\",\n",
      "    \"input_data_config\": {\n",
      "        \"train\": {\n",
      "            \"TrainingInputMode\": \"File\",\n",
      "            \"S3DistributionType\": \"FullyReplicated\",\n",
      "            \"RecordWrapperType\": \"None\"\n",
      "        }\n",
      "    },\n",
      "    \"input_dir\": \"/opt/ml/input\",\n",
      "    \"is_master\": true,\n",
      "    \"job_name\": \"sagemaker-scikit-learn-2020-01-24-19-36-47-954\",\n",
      "    \"log_level\": 20,\n",
      "    \"master_hostname\": \"algo-1\",\n",
      "    \"model_dir\": \"/opt/ml/model\",\n",
      "    \"module_dir\": \"s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-19-36-47-954/source/sourcedir.tar.gz\",\n",
      "    \"module_name\": \"train_knn\",\n",
      "    \"network_interface_name\": \"eth0\",\n",
      "    \"num_cpus\": 16,\n",
      "    \"num_gpus\": 0,\n",
      "    \"output_data_dir\": \"/opt/ml/output/data\",\n",
      "    \"output_dir\": \"/opt/ml/output\",\n",
      "    \"output_intermediate_dir\": \"/opt/ml/output/intermediate\",\n",
      "    \"resource_config\": {\n",
      "        \"current_host\": \"algo-1\",\n",
      "        \"hosts\": [\n",
      "            \"algo-1\"\n",
      "        ],\n",
      "        \"network_interface_name\": \"eth0\"\n",
      "    },\n",
      "    \"user_entry_point\": \"train_knn.py\"\u001b[0m\n",
      "\u001b[34m}\n",
      "\u001b[0m\n",
      "\u001b[34mEnvironment variables:\n",
      "\u001b[0m\n",
      "\u001b[34mSM_HOSTS=[\"algo-1\"]\u001b[0m\n",
      "\u001b[34mSM_NETWORK_INTERFACE_NAME=eth0\u001b[0m\n",
      "\u001b[34mSM_HPS={\"n_neighbors\":13}\u001b[0m\n",
      "\u001b[34mSM_USER_ENTRY_POINT=train_knn.py\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_PARAMS={}\u001b[0m\n",
      "\u001b[34mSM_RESOURCE_CONFIG={\"current_host\":\"algo-1\",\"hosts\":[\"algo-1\"],\"network_interface_name\":\"eth0\"}\u001b[0m\n",
      "\u001b[34mSM_INPUT_DATA_CONFIG={\"train\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}}\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DATA_DIR=/opt/ml/output/data\u001b[0m\n",
      "\u001b[34mSM_CHANNELS=[\"train\"]\u001b[0m\n",
      "\u001b[34mSM_CURRENT_HOST=algo-1\u001b[0m\n",
      "\u001b[34mSM_MODULE_NAME=train_knn\u001b[0m\n",
      "\u001b[34mSM_LOG_LEVEL=20\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_MODULE=sagemaker_sklearn_container.training:main\u001b[0m\n",
      "\u001b[34mSM_INPUT_DIR=/opt/ml/input\u001b[0m\n",
      "\u001b[34mSM_INPUT_CONFIG_DIR=/opt/ml/input/config\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DIR=/opt/ml/output\u001b[0m\n",
      "\u001b[34mSM_NUM_CPUS=16\u001b[0m\n",
      "\u001b[34mSM_NUM_GPUS=0\u001b[0m\n",
      "\u001b[34mSM_MODEL_DIR=/opt/ml/model\u001b[0m\n",
      "\u001b[34mSM_MODULE_DIR=s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-19-36-47-954/source/sourcedir.tar.gz\u001b[0m\n",
      "\u001b[34mSM_TRAINING_ENV={\"additional_framework_parameters\":{},\"channel_input_dirs\":{\"train\":\"/opt/ml/input/data/train\"},\"current_host\":\"algo-1\",\"framework_module\":\"sagemaker_sklearn_container.training:main\",\"hosts\":[\"algo-1\"],\"hyperparameters\":{\"n_neighbors\":13},\"input_config_dir\":\"/opt/ml/input/config\",\"input_data_config\":{\"train\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}},\"input_dir\":\"/opt/ml/input\",\"is_master\":true,\"job_name\":\"sagemaker-scikit-learn-2020-01-24-19-36-47-954\",\"log_level\":20,\"master_hostname\":\"algo-1\",\"model_dir\":\"/opt/ml/model\",\"module_dir\":\"s3://sagemaker-ap-south-1-724515668840/sagemaker-scikit-learn-2020-01-24-19-36-47-954/source/sourcedir.tar.gz\",\"module_name\":\"train_knn\",\"network_interface_name\":\"eth0\",\"num_cpus\":16,\"num_gpus\":0,\"output_data_dir\":\"/opt/ml/output/data\",\"output_dir\":\"/opt/ml/output\",\"output_intermediate_dir\":\"/opt/ml/output/intermediate\",\"resource_config\":{\"current_host\":\"algo-1\",\"hosts\":[\"algo-1\"],\"network_interface_name\":\"eth0\"},\"user_entry_point\":\"train_knn.py\"}\u001b[0m\n",
      "\u001b[34mSM_USER_ARGS=[\"--n_neighbors\",\"13\"]\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_INTERMEDIATE_DIR=/opt/ml/output/intermediate\u001b[0m\n",
      "\u001b[34mSM_CHANNEL_TRAIN=/opt/ml/input/data/train\u001b[0m\n",
      "\u001b[34mSM_HP_N_NEIGHBORS=13\u001b[0m\n",
      "\u001b[34mPYTHONPATH=/miniconda3/bin:/miniconda3/lib/python37.zip:/miniconda3/lib/python3.7:/miniconda3/lib/python3.7/lib-dynload:/miniconda3/lib/python3.7/site-packages\n",
      "\u001b[0m\n",
      "\u001b[34mInvoking script with the following command:\n",
      "\u001b[0m\n",
      "\u001b[34m/miniconda3/bin/python -m train_knn --n_neighbors 13\n",
      "\n",
      "\u001b[0m\n",
      "\u001b[34m/miniconda3/lib/python3.7/site-packages/sklearn/externals/joblib/externals/cloudpickle/cloudpickle.py:47: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  import imp\u001b[0m\n",
      "\u001b[34m2020-01-24 19:39:17,513 sagemaker-containers INFO     Reporting training SUCCESS\u001b[0m\n",
      "\n",
      "2020-01-24 19:39:48 Uploading - Uploading generated training model\n",
      "2020-01-24 19:39:48 Completed - Training job completed\n",
      "Training seconds: 90\n",
      "Billable seconds: 90\n",
      "CPU times: user 401 ms, sys: 17.2 ms, total: 418 ms\n",
      "Wall time: 3min 11s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# Train your estimator on S3 training data\n",
    "KNN_estimator.fit({'train': input_data})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------------------------------------------------!CPU times: user 333 ms, sys: 5.62 ms, total: 338 ms\n",
      "Wall time: 5min 12s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# deploy your model to create a predictor\n",
    "KNN_predictor = KNN_estimator.deploy(instance_type='ml.m4.xlarge',\n",
    "                                     initial_instance_count=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.622\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_y_preds = KNN_predictor.predict(test_x)\n",
    "evaluate_metrics(test_y, test_y_preds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "KNN_predictor.delete_endpoint()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "conda_pytorch_p36",
   "language": "python",
   "name": "conda_pytorch_p36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}