Kaggle/Instant Gratification/Instant Gratification - Deep Learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_kg_hide-input": true,
    "_kg_hide-output": true,
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import warnings\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from keras import optimizers\n",
    "from keras.models import Sequential\n",
    "from keras.callbacks import ReduceLROnPlateau, EarlyStopping\n",
    "from keras.layers import Dense, Dropout, BatchNormalization, Activation\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler, StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix, roc_auc_score\n",
    "\n",
    "# Set seeds to make the experiment more reproducible.\n",
    "from tensorflow import set_random_seed\n",
    "from numpy.random import seed\n",
    "set_random_seed(0)\n",
    "seed(0)\n",
    "\n",
    "%matplotlib inline\n",
    "sns.set(style=\"whitegrid\")\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_kg_hide-input": true,
    "_kg_hide-output": true
   },
   "outputs": [],
   "source": [
    "train = pd.read_csv('../input/train.csv')\n",
    "test = pd.read_csv('../input/test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train set shape: (262144, 258)\n",
      "Test set shape: (131073, 257)\n",
      "Train set overview:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>muggy-smalt-axolotl-pembus</th>\n",
       "      <th>dorky-peach-sheepdog-ordinal</th>\n",
       "      <th>slimy-seashell-cassowary-goose</th>\n",
       "      <th>snazzy-harlequin-chicken-distraction</th>\n",
       "      <th>frumpy-smalt-mau-ordinal</th>\n",
       "      <th>stealthy-beige-pinscher-golden</th>\n",
       "      <th>chummy-cream-tarantula-entropy</th>\n",
       "      <th>hazy-emerald-cuttlefish-unsorted</th>\n",
       "      <th>nerdy-indigo-wolfhound-sorted</th>\n",
       "      <th>leaky-amaranth-lizard-sorted</th>\n",
       "      <th>ugly-tangerine-chihuahua-important</th>\n",
       "      <th>shaggy-silver-indri-fimbus</th>\n",
       "      <th>flaky-chocolate-beetle-grandmaster</th>\n",
       "      <th>squirrely-harlequin-sheep-sumble</th>\n",
       "      <th>freaky-tan-angelfish-noise</th>\n",
       "      <th>lousy-plum-penguin-sumble</th>\n",
       "      <th>bluesy-rose-wallaby-discard</th>\n",
       "      <th>baggy-copper-oriole-dummy</th>\n",
       "      <th>stealthy-scarlet-hound-fepid</th>\n",
       "      <th>greasy-cinnamon-bonobo-contributor</th>\n",
       "      <th>cranky-cardinal-dogfish-ordinal</th>\n",
       "      <th>snippy-auburn-vole-learn</th>\n",
       "      <th>greasy-sepia-coral-dataset</th>\n",
       "      <th>flabby-tangerine-fowl-entropy</th>\n",
       "      <th>lousy-smalt-pinscher-dummy</th>\n",
       "      <th>bluesy-brass-chihuahua-distraction</th>\n",
       "      <th>goopy-eggplant-indri-entropy</th>\n",
       "      <th>homey-sepia-bombay-sorted</th>\n",
       "      <th>homely-ruby-bulldog-entropy</th>\n",
       "      <th>hasty-blue-sheep-contributor</th>\n",
       "      <th>blurry-wisteria-oyster-master</th>\n",
       "      <th>snoopy-auburn-dogfish-expert</th>\n",
       "      <th>stinky-maroon-blue-kernel</th>\n",
       "      <th>bumpy-amaranth-armadillo-important</th>\n",
       "      <th>slaphappy-peach-oyster-master</th>\n",
       "      <th>dorky-tomato-ragdoll-dataset</th>\n",
       "      <th>messy-mauve-wolverine-ordinal</th>\n",
       "      <th>geeky-pumpkin-moorhen-important</th>\n",
       "      <th>crabby-teal-otter-unsorted</th>\n",
       "      <th>...</th>\n",
       "      <th>beady-mauve-frog-distraction</th>\n",
       "      <th>surly-brass-maltese-ordinal</th>\n",
       "      <th>beady-asparagus-opossum-expert</th>\n",
       "      <th>beady-rust-impala-dummy</th>\n",
       "      <th>droopy-amethyst-dachshund-hint</th>\n",
       "      <th>homey-crimson-budgerigar-grandmaster</th>\n",
       "      <th>droopy-cardinal-impala-important</th>\n",
       "      <th>woozy-apricot-moose-hint</th>\n",
       "      <th>paltry-sapphire-labradoodle-dummy</th>\n",
       "      <th>crappy-carmine-eagle-entropy</th>\n",
       "      <th>greasy-magnolia-spider-grandmaster</th>\n",
       "      <th>crabby-carmine-flounder-sorted</th>\n",
       "      <th>skimpy-copper-fowl-grandmaster</th>\n",
       "      <th>hasty-seashell-woodpecker-hint</th>\n",
       "      <th>snappy-purple-bobcat-important</th>\n",
       "      <th>thirsty-carmine-corgi-ordinal</th>\n",
       "      <th>homely-auburn-reindeer-unsorted</th>\n",
       "      <th>crappy-beige-tiger-fepid</th>\n",
       "      <th>cranky-auburn-swan-novice</th>\n",
       "      <th>chewy-bistre-buzzard-expert</th>\n",
       "      <th>skinny-cyan-macaque-pembus</th>\n",
       "      <th>slimy-periwinkle-otter-expert</th>\n",
       "      <th>snazzy-burgundy-clam-novice</th>\n",
       "      <th>cozy-ochre-gorilla-gaussian</th>\n",
       "      <th>homey-sangria-wolfhound-dummy</th>\n",
       "      <th>snazzy-asparagus-hippopotamus-contributor</th>\n",
       "      <th>paltry-red-hamster-sorted</th>\n",
       "      <th>zippy-dandelion-insect-golden</th>\n",
       "      <th>baggy-coral-bandicoot-unsorted</th>\n",
       "      <th>goopy-lavender-wolverine-fimbus</th>\n",
       "      <th>wheezy-myrtle-mandrill-entropy</th>\n",
       "      <th>wiggy-lilac-lemming-sorted</th>\n",
       "      <th>gloppy-cerise-snail-contributor</th>\n",
       "      <th>woozy-silver-havanese-gaussian</th>\n",
       "      <th>jumpy-thistle-discus-sorted</th>\n",
       "      <th>muggy-turquoise-donkey-important</th>\n",
       "      <th>blurry-buff-hyena-entropy</th>\n",
       "      <th>bluesy-chocolate-kudu-fepid</th>\n",
       "      <th>gamy-white-monster-expert</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>707b395ecdcbb4dc2eabea00e4d1b179</td>\n",
       "      <td>-2.070654</td>\n",
       "      <td>1.018160</td>\n",
       "      <td>0.228643</td>\n",
       "      <td>0.857221</td>\n",
       "      <td>0.052271</td>\n",
       "      <td>0.230303</td>\n",
       "      <td>-6.385090</td>\n",
       "      <td>0.439369</td>\n",
       "      <td>-0.721946</td>\n",
       "      <td>-0.227027</td>\n",
       "      <td>0.575964</td>\n",
       "      <td>1.541908</td>\n",
       "      <td>1.745286</td>\n",
       "      <td>-0.624271</td>\n",
       "      <td>3.600958</td>\n",
       "      <td>1.176489</td>\n",
       "      <td>-0.182776</td>\n",
       "      <td>-0.228391</td>\n",
       "      <td>1.682263</td>\n",
       "      <td>-0.833236</td>\n",
       "      <td>-4.377688</td>\n",
       "      <td>-5.372410</td>\n",
       "      <td>-0.477742</td>\n",
       "      <td>-0.179005</td>\n",
       "      <td>-0.516475</td>\n",
       "      <td>0.127391</td>\n",
       "      <td>-0.857591</td>\n",
       "      <td>-0.461500</td>\n",
       "      <td>2.160303</td>\n",
       "      <td>-2.118371</td>\n",
       "      <td>0.515493</td>\n",
       "      <td>-1.201493</td>\n",
       "      <td>-0.027377</td>\n",
       "      <td>-1.154024</td>\n",
       "      <td>0.753204</td>\n",
       "      <td>-0.179651</td>\n",
       "      <td>-0.807341</td>\n",
       "      <td>-1.663626</td>\n",
       "      <td>0.893806</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.829848</td>\n",
       "      <td>2.347131</td>\n",
       "      <td>0.082462</td>\n",
       "      <td>-1.012654</td>\n",
       "      <td>0.593752</td>\n",
       "      <td>2.904654</td>\n",
       "      <td>-0.428974</td>\n",
       "      <td>-0.919979</td>\n",
       "      <td>2.849575</td>\n",
       "      <td>-0.906744</td>\n",
       "      <td>0.729459</td>\n",
       "      <td>0.386140</td>\n",
       "      <td>0.319814</td>\n",
       "      <td>-0.407682</td>\n",
       "      <td>-0.170667</td>\n",
       "      <td>-1.242919</td>\n",
       "      <td>-1.719046</td>\n",
       "      <td>-0.132395</td>\n",
       "      <td>-0.368991</td>\n",
       "      <td>-5.112553</td>\n",
       "      <td>-2.085988</td>\n",
       "      <td>-0.897257</td>\n",
       "      <td>1.080671</td>\n",
       "      <td>-0.273262</td>\n",
       "      <td>0.342824</td>\n",
       "      <td>0.640177</td>\n",
       "      <td>-0.415298</td>\n",
       "      <td>-0.483126</td>\n",
       "      <td>-0.080799</td>\n",
       "      <td>2.416224</td>\n",
       "      <td>0.351895</td>\n",
       "      <td>0.618824</td>\n",
       "      <td>-1.542423</td>\n",
       "      <td>0.598175</td>\n",
       "      <td>0.611757</td>\n",
       "      <td>0.678772</td>\n",
       "      <td>0.247059</td>\n",
       "      <td>-0.806677</td>\n",
       "      <td>-0.193649</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5880c03c6582a7b42248668e56b4bdec</td>\n",
       "      <td>-0.491702</td>\n",
       "      <td>0.082645</td>\n",
       "      <td>-0.011193</td>\n",
       "      <td>1.071266</td>\n",
       "      <td>-0.346347</td>\n",
       "      <td>-0.082209</td>\n",
       "      <td>0.110579</td>\n",
       "      <td>-0.382374</td>\n",
       "      <td>-0.229620</td>\n",
       "      <td>0.783980</td>\n",
       "      <td>-1.280579</td>\n",
       "      <td>-1.003480</td>\n",
       "      <td>-7.753201</td>\n",
       "      <td>-1.320547</td>\n",
       "      <td>0.919078</td>\n",
       "      <td>-1.036068</td>\n",
       "      <td>0.030213</td>\n",
       "      <td>0.910172</td>\n",
       "      <td>-0.905345</td>\n",
       "      <td>0.646641</td>\n",
       "      <td>-0.465291</td>\n",
       "      <td>-0.531735</td>\n",
       "      <td>-0.756781</td>\n",
       "      <td>0.193724</td>\n",
       "      <td>0.224277</td>\n",
       "      <td>-0.474412</td>\n",
       "      <td>1.885805</td>\n",
       "      <td>0.205439</td>\n",
       "      <td>-6.481422</td>\n",
       "      <td>1.035620</td>\n",
       "      <td>-0.453623</td>\n",
       "      <td>0.375936</td>\n",
       "      <td>-0.320670</td>\n",
       "      <td>-0.144646</td>\n",
       "      <td>-0.220129</td>\n",
       "      <td>0.577826</td>\n",
       "      <td>-0.360512</td>\n",
       "      <td>-0.600107</td>\n",
       "      <td>0.008111</td>\n",
       "      <td>...</td>\n",
       "      <td>0.982205</td>\n",
       "      <td>-1.161978</td>\n",
       "      <td>0.532269</td>\n",
       "      <td>1.133215</td>\n",
       "      <td>0.003503</td>\n",
       "      <td>-1.390962</td>\n",
       "      <td>0.158572</td>\n",
       "      <td>0.143794</td>\n",
       "      <td>-0.317185</td>\n",
       "      <td>1.017192</td>\n",
       "      <td>-0.395342</td>\n",
       "      <td>-0.642357</td>\n",
       "      <td>-0.627209</td>\n",
       "      <td>0.257271</td>\n",
       "      <td>-1.461564</td>\n",
       "      <td>0.325613</td>\n",
       "      <td>1.628369</td>\n",
       "      <td>0.640040</td>\n",
       "      <td>0.750735</td>\n",
       "      <td>1.164573</td>\n",
       "      <td>0.900373</td>\n",
       "      <td>0.063489</td>\n",
       "      <td>0.948158</td>\n",
       "      <td>0.273014</td>\n",
       "      <td>-1.269147</td>\n",
       "      <td>-0.251101</td>\n",
       "      <td>-2.271731</td>\n",
       "      <td>-0.044167</td>\n",
       "      <td>-0.443766</td>\n",
       "      <td>-1.144794</td>\n",
       "      <td>-0.645115</td>\n",
       "      <td>-1.246090</td>\n",
       "      <td>2.613357</td>\n",
       "      <td>-0.479664</td>\n",
       "      <td>1.581289</td>\n",
       "      <td>0.931258</td>\n",
       "      <td>0.151937</td>\n",
       "      <td>-0.766595</td>\n",
       "      <td>0.474351</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4ccbcb3d13e5072ff1d9c61afe2c4f77</td>\n",
       "      <td>-1.680473</td>\n",
       "      <td>0.860529</td>\n",
       "      <td>-1.076195</td>\n",
       "      <td>0.740124</td>\n",
       "      <td>3.678445</td>\n",
       "      <td>0.288558</td>\n",
       "      <td>0.515875</td>\n",
       "      <td>0.920590</td>\n",
       "      <td>-1.223277</td>\n",
       "      <td>-1.029780</td>\n",
       "      <td>-2.203397</td>\n",
       "      <td>-7.088717</td>\n",
       "      <td>0.438218</td>\n",
       "      <td>-0.848173</td>\n",
       "      <td>1.542666</td>\n",
       "      <td>-2.166858</td>\n",
       "      <td>-0.867670</td>\n",
       "      <td>-0.980947</td>\n",
       "      <td>0.567793</td>\n",
       "      <td>1.323430</td>\n",
       "      <td>-2.076700</td>\n",
       "      <td>-0.291598</td>\n",
       "      <td>-1.564816</td>\n",
       "      <td>-8.718695</td>\n",
       "      <td>0.340144</td>\n",
       "      <td>-0.566402</td>\n",
       "      <td>0.844324</td>\n",
       "      <td>0.816421</td>\n",
       "      <td>-1.019114</td>\n",
       "      <td>-0.881431</td>\n",
       "      <td>-2.285710</td>\n",
       "      <td>-0.090958</td>\n",
       "      <td>-0.898440</td>\n",
       "      <td>-0.584417</td>\n",
       "      <td>-0.143660</td>\n",
       "      <td>-0.182084</td>\n",
       "      <td>0.798516</td>\n",
       "      <td>0.010756</td>\n",
       "      <td>-0.347155</td>\n",
       "      <td>...</td>\n",
       "      <td>0.829467</td>\n",
       "      <td>0.588236</td>\n",
       "      <td>0.427946</td>\n",
       "      <td>-0.563037</td>\n",
       "      <td>-0.103990</td>\n",
       "      <td>-0.817698</td>\n",
       "      <td>1.251046</td>\n",
       "      <td>-0.977157</td>\n",
       "      <td>2.732600</td>\n",
       "      <td>1.997984</td>\n",
       "      <td>-0.214285</td>\n",
       "      <td>-0.389428</td>\n",
       "      <td>-1.007633</td>\n",
       "      <td>0.336435</td>\n",
       "      <td>-0.851292</td>\n",
       "      <td>-0.024184</td>\n",
       "      <td>0.455908</td>\n",
       "      <td>0.458753</td>\n",
       "      <td>-0.267230</td>\n",
       "      <td>-2.032402</td>\n",
       "      <td>0.203082</td>\n",
       "      <td>0.654107</td>\n",
       "      <td>-3.512338</td>\n",
       "      <td>-0.840937</td>\n",
       "      <td>0.519407</td>\n",
       "      <td>-0.028053</td>\n",
       "      <td>-1.621083</td>\n",
       "      <td>0.142132</td>\n",
       "      <td>1.514664</td>\n",
       "      <td>0.828815</td>\n",
       "      <td>0.516422</td>\n",
       "      <td>0.130521</td>\n",
       "      <td>-0.459210</td>\n",
       "      <td>2.028205</td>\n",
       "      <td>-0.093968</td>\n",
       "      <td>-0.218274</td>\n",
       "      <td>-0.163136</td>\n",
       "      <td>-0.870289</td>\n",
       "      <td>0.064038</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>e350f17a357f12a1941f0837afb7eb8d</td>\n",
       "      <td>0.183774</td>\n",
       "      <td>0.919134</td>\n",
       "      <td>-0.946958</td>\n",
       "      <td>0.918492</td>\n",
       "      <td>0.862278</td>\n",
       "      <td>1.155287</td>\n",
       "      <td>0.911106</td>\n",
       "      <td>0.562598</td>\n",
       "      <td>-1.349685</td>\n",
       "      <td>-1.182729</td>\n",
       "      <td>0.003159</td>\n",
       "      <td>-0.626847</td>\n",
       "      <td>0.368980</td>\n",
       "      <td>1.560784</td>\n",
       "      <td>0.502851</td>\n",
       "      <td>-0.108050</td>\n",
       "      <td>0.633208</td>\n",
       "      <td>-0.411502</td>\n",
       "      <td>-3.201592</td>\n",
       "      <td>-0.710612</td>\n",
       "      <td>0.786816</td>\n",
       "      <td>0.500979</td>\n",
       "      <td>-1.040048</td>\n",
       "      <td>-1.369170</td>\n",
       "      <td>0.987666</td>\n",
       "      <td>-0.681838</td>\n",
       "      <td>-0.331372</td>\n",
       "      <td>2.254289</td>\n",
       "      <td>-0.009330</td>\n",
       "      <td>2.007067</td>\n",
       "      <td>1.203750</td>\n",
       "      <td>-2.003928</td>\n",
       "      <td>-0.566088</td>\n",
       "      <td>0.223452</td>\n",
       "      <td>0.434202</td>\n",
       "      <td>-1.203766</td>\n",
       "      <td>-0.103490</td>\n",
       "      <td>0.441111</td>\n",
       "      <td>1.818458</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.231836</td>\n",
       "      <td>0.833236</td>\n",
       "      <td>-0.454226</td>\n",
       "      <td>-1.614694</td>\n",
       "      <td>0.159948</td>\n",
       "      <td>-0.150059</td>\n",
       "      <td>-1.570599</td>\n",
       "      <td>0.960839</td>\n",
       "      <td>0.102214</td>\n",
       "      <td>0.077236</td>\n",
       "      <td>0.852834</td>\n",
       "      <td>-1.265608</td>\n",
       "      <td>-3.219190</td>\n",
       "      <td>0.251194</td>\n",
       "      <td>0.215861</td>\n",
       "      <td>-0.009520</td>\n",
       "      <td>1.611203</td>\n",
       "      <td>1.679806</td>\n",
       "      <td>-0.008419</td>\n",
       "      <td>0.658384</td>\n",
       "      <td>-0.132437</td>\n",
       "      <td>-1.466823</td>\n",
       "      <td>-1.577080</td>\n",
       "      <td>-0.800346</td>\n",
       "      <td>1.960795</td>\n",
       "      <td>-4.042900</td>\n",
       "      <td>1.722143</td>\n",
       "      <td>-0.261888</td>\n",
       "      <td>-1.145005</td>\n",
       "      <td>-1.864582</td>\n",
       "      <td>-1.168967</td>\n",
       "      <td>1.385089</td>\n",
       "      <td>-0.353028</td>\n",
       "      <td>3.316150</td>\n",
       "      <td>-0.524087</td>\n",
       "      <td>-0.794327</td>\n",
       "      <td>3.936365</td>\n",
       "      <td>0.682989</td>\n",
       "      <td>-2.521211</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a8f910ea6075b6376af079055965ff68</td>\n",
       "      <td>-0.203933</td>\n",
       "      <td>-0.177252</td>\n",
       "      <td>0.368074</td>\n",
       "      <td>-0.701320</td>\n",
       "      <td>-1.104391</td>\n",
       "      <td>0.735760</td>\n",
       "      <td>0.894273</td>\n",
       "      <td>-1.375826</td>\n",
       "      <td>-5.144946</td>\n",
       "      <td>-2.048711</td>\n",
       "      <td>0.629773</td>\n",
       "      <td>-4.252669</td>\n",
       "      <td>-0.087420</td>\n",
       "      <td>-0.794367</td>\n",
       "      <td>-1.063963</td>\n",
       "      <td>0.115997</td>\n",
       "      <td>0.895180</td>\n",
       "      <td>3.184848</td>\n",
       "      <td>2.057840</td>\n",
       "      <td>-0.950821</td>\n",
       "      <td>0.961059</td>\n",
       "      <td>-1.837828</td>\n",
       "      <td>-0.437156</td>\n",
       "      <td>-0.828433</td>\n",
       "      <td>0.373747</td>\n",
       "      <td>-0.099787</td>\n",
       "      <td>-0.976280</td>\n",
       "      <td>-0.165921</td>\n",
       "      <td>3.297221</td>\n",
       "      <td>3.914132</td>\n",
       "      <td>-4.971376</td>\n",
       "      <td>-0.286520</td>\n",
       "      <td>-0.160133</td>\n",
       "      <td>-3.301453</td>\n",
       "      <td>-1.021032</td>\n",
       "      <td>-0.562744</td>\n",
       "      <td>0.574065</td>\n",
       "      <td>-0.368194</td>\n",
       "      <td>-0.507458</td>\n",
       "      <td>...</td>\n",
       "      <td>0.178099</td>\n",
       "      <td>-0.410396</td>\n",
       "      <td>-1.184236</td>\n",
       "      <td>1.681727</td>\n",
       "      <td>0.589606</td>\n",
       "      <td>0.064222</td>\n",
       "      <td>0.258885</td>\n",
       "      <td>0.560241</td>\n",
       "      <td>-1.545597</td>\n",
       "      <td>0.822283</td>\n",
       "      <td>1.518209</td>\n",
       "      <td>0.460143</td>\n",
       "      <td>0.822488</td>\n",
       "      <td>1.362718</td>\n",
       "      <td>0.218560</td>\n",
       "      <td>-1.038514</td>\n",
       "      <td>1.000763</td>\n",
       "      <td>-0.975878</td>\n",
       "      <td>-0.551268</td>\n",
       "      <td>-0.133044</td>\n",
       "      <td>-0.393092</td>\n",
       "      <td>1.236473</td>\n",
       "      <td>1.657100</td>\n",
       "      <td>0.833020</td>\n",
       "      <td>0.665379</td>\n",
       "      <td>-0.900025</td>\n",
       "      <td>0.291908</td>\n",
       "      <td>0.482727</td>\n",
       "      <td>0.552399</td>\n",
       "      <td>0.970496</td>\n",
       "      <td>-0.279168</td>\n",
       "      <td>1.544356</td>\n",
       "      <td>2.959727</td>\n",
       "      <td>1.641201</td>\n",
       "      <td>-0.130818</td>\n",
       "      <td>-0.264292</td>\n",
       "      <td>-0.748668</td>\n",
       "      <td>0.964218</td>\n",
       "      <td>0.087079</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id   ...    target\n",
       "0  707b395ecdcbb4dc2eabea00e4d1b179   ...         0\n",
       "1  5880c03c6582a7b42248668e56b4bdec   ...         0\n",
       "2  4ccbcb3d13e5072ff1d9c61afe2c4f77   ...         1\n",
       "3  e350f17a357f12a1941f0837afb7eb8d   ...         0\n",
       "4  a8f910ea6075b6376af079055965ff68   ...         0\n",
       "\n",
       "[5 rows x 258 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Train set shape:', train.shape)\n",
    "print('Test set shape:', test.shape)\n",
    "print('Train set overview:')\n",
    "display(train.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On both our train and test sets all columns are numerical, and we also don't have any missing data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# EDA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Target distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, ax = plt.subplots(figsize=(6, 6))\n",
    "ax = sns.countplot(x=\"target\", data=train, label=\"Label count\")\n",
    "sns.despine(bottom=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also we have a balanced target distribution.\n",
    "\n",
    "## Feature distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_distribution():\n",
    "    f, axes = plt.subplots(1, 3, figsize=(20, 8), sharex=True)\n",
    "    for feature in train.columns[1:31]:\n",
    "        sns.distplot(train[feature], ax=axes[0], axlabel='First 30 features').set_title(\"Complete set\")\n",
    "        sns.distplot(train[train['target']==1][feature], ax=axes[1], axlabel='First 30 features').set_title(\"target = 1\")\n",
    "        sns.distplot(train[train['target']==0][feature], ax=axes[2], axlabel='First 30 features').set_title(\"target = 0\")\n",
    "    sns.despine(left=True)\n",
    "    plt.tight_layout()\n",
    "\n",
    "plot_distribution()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The features seems to be normalized.\n",
    "\n",
    "## Process data for model\n",
    "\n",
    "### Turn \"wheezy-copper-turtle-magic\" into a categorical feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = pd.concat([train, pd.get_dummies(train['wheezy-copper-turtle-magic'], prefix='magic', drop_first=True)], axis=1).drop(['wheezy-copper-turtle-magic'], axis=1)\n",
    "test = pd.concat([test, pd.get_dummies(test['wheezy-copper-turtle-magic'], prefix='magic', drop_first=True)], axis=1).drop(['wheezy-copper-turtle-magic'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train/validation random split (80% train / 20% validation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = train['target']\n",
    "train.drop('target', axis=1, inplace=True)\n",
    "train.drop('id', axis=1, inplace=True)\n",
    "test.drop('id', axis=1, inplace=True)\n",
    "\n",
    "X_train, X_val, Y_train, Y_val = train_test_split(train, labels, test_size=0.2, random_state=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Normalize data using MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "non_cat_features = list(train.filter(regex='^(?!magic_)'))\n",
    "scaler = MinMaxScaler()\n",
    "X_train[non_cat_features] = scaler.fit_transform(X_train[non_cat_features])\n",
    "X_val[non_cat_features] = scaler.transform(X_val[non_cat_features])\n",
    "test[non_cat_features] = scaler.transform(test[non_cat_features])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model\n",
    "\n",
    "## Model parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "BATCH_SIZE = 128\n",
    "EPOCHS = 30\n",
    "LEARNING_RATE = 0.01\n",
    "ES_PATIENCE = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_1 (Dense)              (None, 1024)              785408    \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 1024)              4096      \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 1024)              0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 1024)              0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 256)               262400    \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 256)               1024      \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 1)                 257       \n",
      "=================================================================\n",
      "Total params: 1,053,185\n",
      "Trainable params: 1,050,625\n",
      "Non-trainable params: 2,560\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(1024, input_dim=X_train.shape[1]))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(256))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(1, activation=\"sigmoid\"))\n",
    "es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=ES_PATIENCE)\n",
    "callback_list = [es]\n",
    "\n",
    "optimizer = optimizers.Adam(lr=0.0001)\n",
    "model.compile(optimizer=optimizer, loss=\"binary_crossentropy\",  metrics=['binary_accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "_kg_hide-output": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Train on 209715 samples, validate on 52429 samples\n",
      "Epoch 1/30\n",
      " - 21s - loss: 0.7833 - binary_accuracy: 0.5054 - val_loss: 0.6878 - val_binary_accuracy: 0.5402\n",
      "Epoch 2/30\n",
      " - 20s - loss: 0.7194 - binary_accuracy: 0.5202 - val_loss: 0.6826 - val_binary_accuracy: 0.5693\n",
      "Epoch 3/30\n",
      " - 20s - loss: 0.6963 - binary_accuracy: 0.5354 - val_loss: 0.6803 - val_binary_accuracy: 0.5820\n",
      "Epoch 4/30\n",
      " - 20s - loss: 0.6859 - binary_accuracy: 0.5517 - val_loss: 0.6771 - val_binary_accuracy: 0.5865\n",
      "Epoch 5/30\n",
      " - 19s - loss: 0.6780 - binary_accuracy: 0.5705 - val_loss: 0.6705 - val_binary_accuracy: 0.5997\n",
      "Epoch 6/30\n",
      " - 19s - loss: 0.6684 - binary_accuracy: 0.5898 - val_loss: 0.6577 - val_binary_accuracy: 0.6297\n",
      "Epoch 7/30\n",
      " - 19s - loss: 0.6540 - binary_accuracy: 0.6123 - val_loss: 0.6560 - val_binary_accuracy: 0.5923\n",
      "Epoch 8/30\n",
      " - 19s - loss: 0.6358 - binary_accuracy: 0.6368 - val_loss: 0.6509 - val_binary_accuracy: 0.5965\n",
      "Epoch 9/30\n",
      " - 19s - loss: 0.6182 - binary_accuracy: 0.6560 - val_loss: 0.6097 - val_binary_accuracy: 0.6811\n",
      "Epoch 10/30\n",
      " - 20s - loss: 0.6007 - binary_accuracy: 0.6741 - val_loss: 0.5951 - val_binary_accuracy: 0.6976\n",
      "Epoch 11/30\n",
      " - 19s - loss: 0.5859 - binary_accuracy: 0.6876 - val_loss: 0.6986 - val_binary_accuracy: 0.5685\n",
      "Epoch 12/30\n",
      " - 19s - loss: 0.5716 - binary_accuracy: 0.7008 - val_loss: 0.5874 - val_binary_accuracy: 0.6903\n",
      "Epoch 13/30\n",
      " - 19s - loss: 0.5574 - binary_accuracy: 0.7125 - val_loss: 0.5698 - val_binary_accuracy: 0.7109\n",
      "Epoch 14/30\n",
      " - 19s - loss: 0.5448 - binary_accuracy: 0.7228 - val_loss: 0.7322 - val_binary_accuracy: 0.5716\n",
      "Epoch 15/30\n",
      " - 19s - loss: 0.5365 - binary_accuracy: 0.7282 - val_loss: 0.5618 - val_binary_accuracy: 0.7145\n",
      "Epoch 16/30\n",
      " - 19s - loss: 0.5240 - binary_accuracy: 0.7381 - val_loss: 0.6038 - val_binary_accuracy: 0.6682\n",
      "Epoch 17/30\n",
      " - 19s - loss: 0.5171 - binary_accuracy: 0.7447 - val_loss: 0.5601 - val_binary_accuracy: 0.7115\n",
      "Epoch 18/30\n",
      " - 19s - loss: 0.5074 - binary_accuracy: 0.7509 - val_loss: 0.5479 - val_binary_accuracy: 0.7245\n",
      "Epoch 19/30\n",
      " - 19s - loss: 0.4979 - binary_accuracy: 0.7578 - val_loss: 0.5519 - val_binary_accuracy: 0.7207\n",
      "Epoch 20/30\n",
      " - 19s - loss: 0.4886 - binary_accuracy: 0.7639 - val_loss: 0.5555 - val_binary_accuracy: 0.7163\n",
      "Epoch 21/30\n",
      " - 19s - loss: 0.4815 - binary_accuracy: 0.7682 - val_loss: 0.5431 - val_binary_accuracy: 0.7267\n",
      "Epoch 22/30\n",
      " - 19s - loss: 0.4738 - binary_accuracy: 0.7726 - val_loss: 0.5443 - val_binary_accuracy: 0.7253\n",
      "Epoch 23/30\n",
      " - 19s - loss: 0.4663 - binary_accuracy: 0.7774 - val_loss: 0.6160 - val_binary_accuracy: 0.6748\n",
      "Epoch 24/30\n",
      " - 19s - loss: 0.4604 - binary_accuracy: 0.7821 - val_loss: 0.5421 - val_binary_accuracy: 0.7270\n",
      "Epoch 25/30\n",
      " - 19s - loss: 0.4544 - binary_accuracy: 0.7857 - val_loss: 0.5836 - val_binary_accuracy: 0.6998\n",
      "Epoch 26/30\n",
      " - 19s - loss: 0.4471 - binary_accuracy: 0.7902 - val_loss: 0.5548 - val_binary_accuracy: 0.7204\n",
      "Epoch 27/30\n",
      " - 19s - loss: 0.4409 - binary_accuracy: 0.7940 - val_loss: 0.5550 - val_binary_accuracy: 0.7211\n",
      "Epoch 28/30\n",
      " - 19s - loss: 0.4355 - binary_accuracy: 0.7975 - val_loss: 0.5353 - val_binary_accuracy: 0.7335\n",
      "Epoch 29/30\n",
      " - 19s - loss: 0.4304 - binary_accuracy: 0.8006 - val_loss: 0.5379 - val_binary_accuracy: 0.7326\n",
      "Epoch 30/30\n",
      " - 19s - loss: 0.4232 - binary_accuracy: 0.8053 - val_loss: 0.8033 - val_binary_accuracy: 0.6076\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, Y_train, validation_data=(X_val, Y_val), \n",
    "                    callbacks=callback_list, \n",
    "                    epochs=EPOCHS, \n",
    "                    batch_size=BATCH_SIZE, \n",
    "                    verbose=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model graph loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style(\"whitegrid\")\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2, sharex='col', figsize=(20, 7))\n",
    "\n",
    "ax1.plot(history.history['loss'], label='Train loss')\n",
    "ax1.plot(history.history['val_loss'], label='Validation loss')\n",
    "ax1.legend(loc='best')\n",
    "ax1.set_title('Loss')\n",
    "\n",
    "ax2.plot(history.history['binary_accuracy'], label='Train Accuracy')\n",
    "ax2.plot(history.history['val_binary_accuracy'], label='Validation accuracy')\n",
    "ax2.legend(loc='best')\n",
    "ax2.set_title('Accuracy')\n",
    "\n",
    "plt.xlabel('Epochs')\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model evaluation\n",
    "\n",
    "## Confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_pred = model.predict_classes(X_train)\n",
    "val_pred = model.predict_classes(X_val)\n",
    "\n",
    "f, axes = plt.subplots(1, 2, figsize=(16, 5), sharex=True)\n",
    "train_cnf_matrix = confusion_matrix(Y_train, train_pred)\n",
    "val_cnf_matrix = confusion_matrix(Y_val, val_pred)\n",
    "\n",
    "train_cnf_matrix_norm = train_cnf_matrix / train_cnf_matrix.sum(axis=1)[:, np.newaxis]\n",
    "val_cnf_matrix_norm = val_cnf_matrix / val_cnf_matrix.sum(axis=1)[:, np.newaxis]\n",
    "\n",
    "train_df_cm = pd.DataFrame(train_cnf_matrix_norm, index=[0, 1], columns=[0, 1])\n",
    "val_df_cm = pd.DataFrame(val_cnf_matrix_norm, index=[0, 1], columns=[0, 1])\n",
    "\n",
    "sns.heatmap(train_df_cm, annot=True, fmt='.2f', cmap=\"Blues\", ax=axes[0]).set_title(\"Train\")\n",
    "sns.heatmap(val_df_cm, annot=True, fmt='.2f', cmap=\"Blues\", ax=axes[1]).set_title(\"Validation\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Metrics ROC AUC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train AUC 0.69\n",
      "Validation AUC 0.61\n"
     ]
    }
   ],
   "source": [
    "print('Train AUC %.2f' % roc_auc_score(Y_train.values, train_pred))\n",
    "print('Validation AUC %.2f' % roc_auc_score(Y_val.values, val_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "_kg_hide-input": true
   },
   "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>id</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1c13f2701648e0b0d46d8a2a5a131a53</td>\n",
       "      <td>0.406026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ba88c155ba898fc8b5099893036ef205</td>\n",
       "      <td>0.297863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7cbab5cea99169139e7e6d8ff74ebb77</td>\n",
       "      <td>0.299330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ca820ad57809f62eb7b4d13f5d4371a0</td>\n",
       "      <td>0.048772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7baaf361537fbd8a1aaa2c97a6d4ccc7</td>\n",
       "      <td>0.007759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>8b3116e5e3e92e971dac305d1a093bf6</td>\n",
       "      <td>0.225391</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>35cfd7cab9bfa29bc963d1b8c94dd280</td>\n",
       "      <td>0.005694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>83cf532df8ff4642a3e33a70fffdac37</td>\n",
       "      <td>0.199151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2e1350fe94ec9f2220bec5245e5e9265</td>\n",
       "      <td>0.555559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>e62020afa72eb54a15725473e3a8475b</td>\n",
       "      <td>0.044808</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id    target\n",
       "0  1c13f2701648e0b0d46d8a2a5a131a53  0.406026\n",
       "1  ba88c155ba898fc8b5099893036ef205  0.297863\n",
       "2  7cbab5cea99169139e7e6d8ff74ebb77  0.299330\n",
       "3  ca820ad57809f62eb7b4d13f5d4371a0  0.048772\n",
       "4  7baaf361537fbd8a1aaa2c97a6d4ccc7  0.007759\n",
       "5  8b3116e5e3e92e971dac305d1a093bf6  0.225391\n",
       "6  35cfd7cab9bfa29bc963d1b8c94dd280  0.005694\n",
       "7  83cf532df8ff4642a3e33a70fffdac37  0.199151\n",
       "8  2e1350fe94ec9f2220bec5245e5e9265  0.555559\n",
       "9  e62020afa72eb54a15725473e3a8475b  0.044808"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = model.predict(test)\n",
    "df = pd.read_csv('../input/sample_submission.csv')\n",
    "df['target'] = predictions\n",
    "df.to_csv('submission.csv', index=False)\n",
    "df.head(10)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}