SimpleSequentialModel.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from random import randint\n",
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_labels =  []\n",
    "train_samples = []"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Example data: \n",
    "- An experiemental drug was tested on individuals from ages 13 to 100. \n",
    "- The trial had 2100 participants. Half were under 65 years old, half were over 65 years old.\n",
    "- 95% of patientes 65 or older experienced side effects.\n",
    "- 95% of patients under 65 experienced no side effects."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for i in range(50):\n",
    "    # The 5% of younger individuals who did experience side effects\n",
    "    random_younger = randint(13,64)\n",
    "    train_samples.append(random_younger)\n",
    "    train_labels.append(1)\n",
    "    \n",
    "    # The 5% of older individuals who did not experience side effects\n",
    "    random_older = randint(65,100)\n",
    "    train_samples.append(random_older)\n",
    "    train_labels.append(0)\n",
    "\n",
    "for i in range(1000):\n",
    "    # The 95% of younger individuals who did not experience side effects\n",
    "    random_younger = randint(13,64)\n",
    "    train_samples.append(random_younger)\n",
    "    train_labels.append(0)\n",
    "    \n",
    "    # The 95% of older individuals who did experience side effects\n",
    "    random_older = randint(65,100)\n",
    "    train_samples.append(random_older)\n",
    "    train_labels.append(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "49\n",
      "94\n",
      "31\n",
      "83\n",
      "13\n",
      "97\n",
      "23\n",
      "73\n",
      "30\n",
      "72\n",
      "29\n",
      "69\n",
      "54\n",
      "97\n",
      "44\n",
      "100\n",
      "50\n",
      "100\n",
      "48\n",
      "94\n",
      "39\n",
      "90\n",
      "35\n",
      "92\n",
      "35\n",
      "81\n",
      "48\n",
      "69\n",
      "58\n",
      "99\n",
      "47\n",
      "66\n",
      "33\n",
      "98\n",
      "20\n",
      "76\n",
      "14\n",
      "76\n",
      "14\n",
      "99\n",
      "28\n",
      "67\n",
      "54\n",
      "76\n",
      "29\n",
      "77\n",
      "64\n",
      "94\n",
      "58\n",
      "95\n",
      "57\n",
      "78\n",
      "22\n",
      "85\n",
      "47\n",
      "99\n",
      "22\n",
      "86\n",
      "44\n",
      "66\n",
      "58\n",
      "94\n",
      "64\n",
      "88\n",
      "20\n",
      "67\n",
      "58\n",
      "69\n",
      "47\n",
      "82\n",
      "52\n",
      "90\n",
      "50\n",
      "86\n",
      "39\n",
      "83\n",
      "55\n",
      "79\n",
      "63\n",
      "68\n",
      "36\n",
      "99\n",
      "45\n",
      "75\n",
      "41\n",
      "86\n",
      "42\n",
      "81\n",
      "22\n",
      "78\n",
      "29\n",
      "68\n",
      "63\n",
      "66\n",
      "52\n",
      "99\n",
      "64\n",
      "88\n",
      "53\n",
      "72\n",
      "38\n",
      "70\n",
      "49\n",
      "69\n",
      "44\n",
      "75\n",
      "54\n",
      "77\n",
      "42\n",
      "96\n",
      "17\n",
      "80\n",
      "55\n",
      "88\n",
      "38\n",
      "97\n",
      "23\n",
      "92\n",
      "28\n",
      "85\n",
      "60\n",
      "89\n",
      "59\n",
      "69\n",
      "28\n",
      "71\n",
      "33\n",
      "84\n",
      "55\n",
      "70\n",
      "34\n",
      "85\n",
      "37\n",
      "88\n",
      "16\n",
      "66\n",
      "56\n",
      "71\n",
      "44\n",
      "96\n",
      "23\n",
      "82\n",
      "26\n",
      "95\n",
      "49\n",
      "69\n",
      "57\n",
      "84\n",
      "31\n",
      "93\n",
      "35\n",
      "65\n",
      "32\n",
      "80\n",
      "31\n",
      "65\n",
      "54\n",
      "84\n",
      "20\n",
      "65\n",
      "25\n",
      "98\n",
      "34\n",
      "77\n",
      "45\n",
      "98\n",
      "29\n",
      "83\n",
      "46\n",
      "98\n",
      "38\n",
      "78\n",
      "28\n",
      "71\n",
      "48\n",
      "90\n",
      "64\n",
      "84\n",
      "54\n",
      "95\n",
      "31\n",
      "99\n",
      "15\n",
      "83\n",
      "17\n",
      "96\n",
      "59\n",
      "97\n",
      "50\n",
      "78\n",
      "60\n",
      "97\n",
      "38\n",
      "73\n",
      "53\n",
      "94\n",
      "15\n",
      "84\n",
      "60\n",
      "75\n",
      "46\n",
      "94\n",
      "44\n",
      "73\n",
      "16\n",
      "91\n",
      "21\n",
      "95\n",
      "18\n",
      "76\n",
      "17\n",
      "93\n",
      "14\n",
      "76\n",
      "55\n",
      "73\n",
      "29\n",
      "73\n",
      "25\n",
      "85\n",
      "13\n",
      "83\n",
      "37\n",
      "86\n",
      "48\n",
      "72\n",
      "37\n",
      "65\n",
      "46\n",
      "98\n",
      "53\n",
      "83\n",
      "44\n",
      "73\n",
      "21\n",
      "76\n",
      "24\n",
      "71\n",
      "53\n",
      "95\n",
      "45\n",
      "87\n",
      "44\n",
      "85\n",
      "38\n",
      "74\n",
      "47\n",
      "70\n",
      "28\n",
      "65\n",
      "47\n",
      "83\n",
      "42\n",
      "75\n",
      "45\n",
      "86\n",
      "62\n",
      "66\n",
      "62\n",
      "71\n",
      "48\n",
      "66\n",
      "39\n",
      "91\n",
      "41\n",
      "95\n",
      "46\n",
      "85\n",
      "61\n",
      "96\n",
      "30\n",
      "74\n",
      "29\n",
      "70\n",
      "44\n",
      "71\n",
      "36\n",
      "83\n",
      "47\n",
      "94\n",
      "37\n",
      "77\n",
      "17\n",
      "98\n",
      "53\n",
      "73\n",
      "51\n",
      "69\n",
      "47\n",
      "87\n",
      "18\n",
      "79\n",
      "62\n",
      "70\n",
      "15\n",
      "96\n",
      "16\n",
      "76\n",
      "57\n",
      "100\n",
      "37\n",
      "82\n",
      "29\n",
      "81\n",
      "51\n",
      "65\n",
      "17\n",
      "100\n",
      "43\n",
      "88\n",
      "56\n",
      "76\n",
      "26\n",
      "71\n",
      "28\n",
      "78\n",
      "64\n",
      "99\n",
      "46\n",
      "98\n",
      "31\n",
      "67\n",
      "24\n",
      "68\n",
      "52\n",
      "90\n",
      "20\n",
      "74\n",
      "16\n",
      "78\n",
      "58\n",
      "100\n",
      "30\n",
      "83\n",
      "56\n",
      "76\n",
      "56\n",
      "84\n",
      "44\n",
      "73\n",
      "31\n",
      "75\n",
      "13\n",
      "87\n",
      "26\n",
      "81\n",
      "52\n",
      "81\n",
      "41\n",
      "87\n",
      "32\n",
      "91\n",
      "30\n",
      "69\n",
      "31\n",
      "92\n",
      "16\n",
      "92\n",
      "63\n",
      "95\n",
      "24\n",
      "98\n",
      "13\n",
      "76\n",
      "22\n",
      "81\n",
      "20\n",
      "89\n",
      "46\n",
      "67\n",
      "45\n",
      "85\n",
      "28\n",
      "72\n",
      "60\n",
      "85\n",
      "27\n",
      "76\n",
      "20\n",
      "100\n",
      "23\n",
      "69\n",
      "14\n",
      "81\n",
      "18\n",
      "74\n",
      "46\n",
      "75\n",
      "49\n",
      "85\n",
      "64\n",
      "73\n",
      "56\n",
      "76\n",
      "52\n",
      "69\n",
      "18\n",
      "84\n",
      "27\n",
      "91\n",
      "64\n",
      "75\n",
      "58\n",
      "90\n",
      "33\n",
      "96\n",
      "18\n",
      "69\n",
      "39\n",
      "71\n",
      "34\n",
      "91\n",
      "19\n",
      "90\n",
      "38\n",
      "70\n",
      "40\n",
      "92\n",
      "18\n",
      "89\n",
      "43\n",
      "100\n",
      "31\n",
      "68\n",
      "52\n",
      "97\n",
      "46\n",
      "100\n",
      "40\n",
      "71\n",
      "26\n",
      "75\n",
      "52\n",
      "79\n",
      "52\n",
      "94\n",
      "13\n",
      "87\n",
      "56\n",
      "93\n",
      "59\n",
      "92\n",
      "17\n",
      "84\n",
      "19\n",
      "80\n",
      "22\n",
      "84\n",
      "39\n",
      "73\n",
      "24\n",
      "66\n",
      "21\n",
      "70\n",
      "14\n",
      "81\n",
      "56\n",
      "74\n",
      "50\n",
      "98\n",
      "27\n",
      "69\n",
      "41\n",
      "98\n",
      "21\n",
      "71\n",
      "37\n",
      "80\n",
      "27\n",
      "83\n",
      "39\n",
      "86\n",
      "46\n",
      "76\n",
      "55\n",
      "77\n",
      "54\n",
      "76\n",
      "39\n",
      "78\n",
      "64\n",
      "74\n",
      "25\n",
      "84\n",
      "56\n",
      "74\n",
      "27\n",
      "90\n",
      "61\n",
      "78\n",
      "30\n",
      "70\n",
      "29\n",
      "67\n",
      "61\n",
      "77\n",
      "37\n",
      "69\n",
      "63\n",
      "83\n",
      "26\n",
      "88\n",
      "55\n",
      "100\n",
      "29\n",
      "77\n",
      "35\n",
      "82\n",
      "14\n",
      "74\n",
      "29\n",
      "90\n",
      "63\n",
      "80\n",
      "55\n",
      "67\n",
      "14\n",
      "71\n",
      "59\n",
      "97\n",
      "19\n",
      "71\n",
      "18\n",
      "81\n",
      "22\n",
      "70\n",
      "19\n",
      "91\n",
      "37\n",
      "85\n",
      "60\n",
      "70\n",
      "20\n",
      "97\n",
      "29\n",
      "78\n",
      "41\n",
      "94\n",
      "52\n",
      "68\n",
      "61\n",
      "90\n",
      "53\n",
      "81\n",
      "59\n",
      "72\n",
      "40\n",
      "99\n",
      "36\n",
      "81\n",
      "26\n",
      "74\n",
      "16\n",
      "95\n",
      "52\n",
      "67\n",
      "39\n",
      "95\n",
      "34\n",
      "65\n",
      "55\n",
      "86\n",
      "14\n",
      "75\n",
      "27\n",
      "76\n",
      "52\n",
      "84\n",
      "24\n",
      "99\n",
      "53\n",
      "78\n",
      "48\n",
      "73\n",
      "38\n",
      "76\n",
      "45\n",
      "86\n",
      "45\n",
      "69\n",
      "19\n",
      "92\n",
      "55\n",
      "89\n",
      "23\n",
      "95\n",
      "49\n",
      "70\n",
      "62\n",
      "75\n",
      "31\n",
      "99\n",
      "46\n",
      "98\n",
      "26\n",
      "85\n",
      "33\n",
      "94\n",
      "58\n",
      "68\n",
      "58\n",
      "78\n",
      "39\n",
      "71\n",
      "24\n",
      "77\n",
      "61\n",
      "94\n",
      "18\n",
      "67\n",
      "14\n",
      "70\n",
      "62\n",
      "76\n",
      "18\n",
      "80\n",
      "24\n",
      "85\n",
      "53\n",
      "72\n",
      "22\n",
      "73\n",
      "51\n",
      "75\n",
      "51\n",
      "65\n",
      "61\n",
      "69\n",
      "16\n",
      "98\n",
      "47\n",
      "98\n",
      "17\n",
      "79\n",
      "21\n",
      "88\n",
      "58\n",
      "80\n",
      "22\n",
      "96\n",
      "56\n",
      "73\n",
      "46\n",
      "85\n",
      "36\n",
      "67\n",
      "24\n",
      "95\n",
      "53\n",
      "79\n",
      "61\n",
      "68\n",
      "61\n",
      "70\n",
      "24\n",
      "72\n",
      "13\n",
      "80\n",
      "28\n",
      "68\n",
      "46\n",
      "87\n",
      "57\n",
      "79\n",
      "43\n",
      "96\n",
      "49\n",
      "91\n",
      "25\n",
      "82\n",
      "59\n",
      "87\n",
      "38\n",
      "66\n",
      "14\n",
      "67\n",
      "22\n",
      "92\n",
      "37\n",
      "85\n",
      "26\n",
      "76\n",
      "50\n",
      "90\n",
      "18\n",
      "97\n",
      "40\n",
      "79\n",
      "36\n",
      "87\n",
      "56\n",
      "87\n",
      "33\n",
      "84\n",
      "62\n",
      "100\n",
      "51\n",
      "77\n",
      "34\n",
      "97\n",
      "53\n",
      "94\n",
      "25\n",
      "93\n",
      "52\n",
      "91\n",
      "13\n",
      "80\n",
      "17\n",
      "70\n",
      "15\n",
      "84\n",
      "44\n",
      "96\n",
      "61\n",
      "86\n",
      "32\n",
      "92\n",
      "43\n",
      "65\n",
      "17\n",
      "97\n",
      "39\n",
      "100\n",
      "19\n",
      "78\n",
      "39\n",
      "72\n",
      "37\n",
      "68\n",
      "56\n",
      "87\n",
      "15\n",
      "91\n",
      "58\n",
      "95\n",
      "40\n",
      "100\n",
      "43\n",
      "72\n",
      "17\n",
      "79\n",
      "42\n",
      "65\n",
      "53\n",
      "96\n",
      "27\n",
      "83\n",
      "48\n",
      "88\n",
      "22\n",
      "70\n",
      "15\n",
      "80\n",
      "26\n",
      "83\n",
      "48\n",
      "93\n",
      "52\n",
      "83\n",
      "58\n",
      "84\n",
      "17\n",
      "88\n",
      "23\n",
      "99\n",
      "46\n",
      "95\n",
      "60\n",
      "78\n",
      "48\n",
      "81\n",
      "58\n",
      "87\n",
      "25\n",
      "83\n",
      "27\n",
      "92\n",
      "37\n",
      "76\n",
      "47\n",
      "71\n",
      "62\n",
      "84\n",
      "14\n",
      "69\n",
      "47\n",
      "83\n",
      "54\n",
      "73\n",
      "53\n",
      "98\n",
      "14\n",
      "73\n",
      "55\n",
      "72\n",
      "54\n",
      "67\n",
      "20\n",
      "82\n",
      "37\n",
      "70\n",
      "38\n",
      "100\n",
      "44\n",
      "74\n",
      "24\n",
      "88\n",
      "16\n",
      "91\n",
      "40\n",
      "74\n",
      "49\n",
      "90\n",
      "57\n",
      "69\n",
      "23\n",
      "80\n",
      "18\n",
      "93\n",
      "34\n",
      "69\n",
      "22\n",
      "82\n",
      "25\n",
      "78\n",
      "25\n",
      "89\n",
      "40\n",
      "67\n",
      "37\n",
      "72\n",
      "31\n",
      "89\n",
      "50\n",
      "73\n",
      "47\n",
      "84\n",
      "38\n",
      "80\n",
      "26\n",
      "80\n",
      "17\n",
      "74\n",
      "24\n",
      "98\n",
      "61\n",
      "80\n",
      "55\n",
      "100\n",
      "33\n",
      "84\n",
      "45\n",
      "66\n",
      "57\n",
      "86\n",
      "37\n",
      "99\n",
      "36\n",
      "70\n",
      "57\n",
      "68\n",
      "64\n",
      "65\n",
      "22\n",
      "72\n",
      "32\n",
      "94\n",
      "30\n",
      "82\n",
      "45\n",
      "97\n",
      "43\n",
      "90\n",
      "30\n",
      "70\n",
      "64\n",
      "100\n",
      "41\n",
      "99\n",
      "28\n",
      "84\n",
      "20\n",
      "70\n",
      "33\n",
      "80\n",
      "27\n",
      "91\n",
      "13\n",
      "72\n",
      "16\n",
      "68\n",
      "59\n",
      "89\n",
      "32\n",
      "99\n",
      "46\n",
      "91\n",
      "16\n",
      "78\n",
      "31\n",
      "80\n",
      "36\n",
      "90\n",
      "54\n",
      "99\n",
      "22\n",
      "86\n",
      "31\n",
      "72\n",
      "31\n",
      "65\n",
      "27\n",
      "66\n",
      "48\n",
      "78\n",
      "34\n",
      "100\n",
      "45\n",
      "92\n",
      "41\n",
      "81\n",
      "17\n",
      "94\n",
      "51\n",
      "79\n",
      "21\n",
      "91\n",
      "27\n",
      "70\n",
      "43\n",
      "80\n",
      "60\n",
      "83\n",
      "62\n",
      "90\n",
      "24\n",
      "94\n",
      "50\n",
      "98\n",
      "19\n",
      "67\n",
      "31\n",
      "90\n",
      "14\n",
      "83\n",
      "32\n",
      "81\n",
      "60\n",
      "90\n",
      "18\n",
      "72\n",
      "56\n",
      "92\n",
      "24\n",
      "79\n",
      "38\n",
      "65\n",
      "43\n",
      "88\n",
      "28\n",
      "86\n",
      "46\n",
      "66\n",
      "14\n",
      "90\n",
      "49\n",
      "79\n",
      "43\n",
      "89\n",
      "39\n",
      "97\n",
      "47\n",
      "66\n",
      "50\n",
      "89\n",
      "63\n",
      "75\n",
      "13\n",
      "78\n",
      "27\n",
      "88\n",
      "64\n",
      "91\n",
      "30\n",
      "73\n",
      "36\n",
      "71\n",
      "33\n",
      "75\n",
      "26\n",
      "69\n",
      "48\n",
      "67\n",
      "22\n",
      "73\n",
      "23\n",
      "93\n",
      "14\n",
      "93\n",
      "32\n",
      "68\n",
      "34\n",
      "74\n",
      "52\n",
      "72\n",
      "60\n",
      "97\n",
      "26\n",
      "95\n",
      "51\n",
      "84\n",
      "31\n",
      "92\n",
      "45\n",
      "74\n",
      "23\n",
      "86\n",
      "23\n",
      "84\n",
      "51\n",
      "80\n",
      "49\n",
      "69\n",
      "15\n",
      "66\n",
      "63\n",
      "83\n",
      "39\n",
      "80\n",
      "24\n",
      "98\n",
      "15\n",
      "75\n",
      "63\n",
      "98\n",
      "15\n",
      "79\n",
      "54\n",
      "97\n",
      "27\n",
      "68\n",
      "20\n",
      "97\n",
      "42\n",
      "71\n",
      "29\n",
      "70\n",
      "40\n",
      "99\n",
      "50\n",
      "66\n",
      "31\n",
      "80\n",
      "53\n",
      "65\n",
      "16\n",
      "98\n",
      "34\n",
      "66\n",
      "36\n",
      "85\n",
      "33\n",
      "75\n",
      "32\n",
      "74\n",
      "27\n",
      "91\n",
      "19\n",
      "100\n",
      "39\n",
      "70\n",
      "31\n",
      "73\n",
      "32\n",
      "82\n",
      "41\n",
      "75\n",
      "45\n",
      "82\n",
      "45\n",
      "74\n",
      "19\n",
      "87\n",
      "49\n",
      "93\n",
      "17\n",
      "87\n",
      "40\n",
      "96\n",
      "21\n",
      "70\n",
      "59\n",
      "93\n",
      "17\n",
      "83\n",
      "44\n",
      "68\n",
      "49\n",
      "88\n",
      "58\n",
      "87\n",
      "56\n",
      "78\n",
      "35\n",
      "81\n",
      "13\n",
      "84\n",
      "22\n",
      "69\n",
      "24\n",
      "71\n",
      "32\n",
      "65\n",
      "23\n",
      "100\n",
      "41\n",
      "70\n",
      "34\n",
      "76\n",
      "63\n",
      "98\n",
      "16\n",
      "79\n",
      "22\n",
      "66\n",
      "61\n",
      "81\n",
      "23\n",
      "71\n",
      "64\n",
      "66\n",
      "50\n",
      "91\n",
      "30\n",
      "81\n",
      "19\n",
      "74\n",
      "53\n",
      "70\n",
      "58\n",
      "67\n",
      "25\n",
      "98\n",
      "16\n",
      "85\n",
      "50\n",
      "73\n",
      "29\n",
      "80\n",
      "54\n",
      "93\n",
      "30\n",
      "76\n",
      "60\n",
      "94\n",
      "16\n",
      "95\n",
      "19\n",
      "91\n",
      "19\n",
      "94\n",
      "42\n",
      "68\n",
      "25\n",
      "93\n",
      "32\n",
      "87\n",
      "57\n",
      "67\n",
      "23\n",
      "76\n",
      "32\n",
      "68\n",
      "59\n",
      "88\n",
      "18\n",
      "84\n",
      "35\n",
      "93\n",
      "38\n",
      "72\n",
      "61\n",
      "74\n",
      "43\n",
      "88\n",
      "48\n",
      "73\n",
      "22\n",
      "66\n",
      "29\n",
      "72\n",
      "40\n",
      "92\n",
      "47\n",
      "88\n",
      "15\n",
      "65\n",
      "15\n",
      "80\n",
      "18\n",
      "99\n",
      "20\n",
      "66\n",
      "41\n",
      "87\n",
      "45\n",
      "79\n",
      "60\n",
      "72\n",
      "56\n",
      "97\n",
      "23\n",
      "92\n",
      "34\n",
      "68\n",
      "63\n",
      "87\n",
      "29\n",
      "97\n",
      "48\n",
      "76\n",
      "13\n",
      "67\n",
      "60\n",
      "81\n",
      "45\n",
      "84\n",
      "41\n",
      "68\n",
      "16\n",
      "83\n",
      "24\n",
      "70\n",
      "16\n",
      "85\n",
      "20\n",
      "83\n",
      "62\n",
      "96\n",
      "18\n",
      "69\n",
      "54\n",
      "97\n",
      "41\n",
      "78\n",
      "31\n",
      "98\n",
      "45\n",
      "68\n",
      "28\n",
      "100\n",
      "13\n",
      "81\n",
      "15\n",
      "70\n",
      "16\n",
      "87\n",
      "13\n",
      "65\n",
      "32\n",
      "80\n",
      "28\n",
      "66\n",
      "49\n",
      "98\n",
      "29\n",
      "74\n",
      "42\n",
      "94\n",
      "45\n",
      "86\n",
      "56\n",
      "91\n",
      "64\n",
      "96\n",
      "27\n",
      "82\n",
      "59\n",
      "100\n",
      "62\n",
      "75\n",
      "45\n",
      "65\n",
      "13\n",
      "100\n",
      "29\n",
      "90\n",
      "54\n",
      "70\n",
      "15\n",
      "71\n",
      "43\n",
      "92\n",
      "14\n",
      "72\n",
      "57\n",
      "69\n",
      "50\n",
      "78\n",
      "54\n",
      "87\n",
      "22\n",
      "93\n",
      "46\n",
      "100\n",
      "37\n",
      "88\n",
      "31\n",
      "95\n",
      "37\n",
      "80\n",
      "34\n",
      "93\n",
      "24\n",
      "87\n",
      "16\n",
      "75\n",
      "13\n",
      "71\n",
      "36\n",
      "87\n",
      "20\n",
      "72\n",
      "28\n",
      "90\n",
      "25\n",
      "80\n",
      "36\n",
      "94\n",
      "31\n",
      "97\n",
      "23\n",
      "74\n",
      "58\n",
      "71\n",
      "59\n",
      "87\n",
      "49\n",
      "100\n",
      "29\n",
      "93\n",
      "20\n",
      "99\n",
      "59\n",
      "65\n",
      "56\n",
      "84\n",
      "23\n",
      "96\n",
      "33\n",
      "84\n",
      "37\n",
      "93\n",
      "42\n",
      "80\n",
      "22\n",
      "74\n",
      "58\n",
      "80\n",
      "38\n",
      "79\n",
      "43\n",
      "65\n",
      "24\n",
      "68\n",
      "21\n",
      "79\n",
      "53\n",
      "79\n",
      "17\n",
      "88\n",
      "23\n",
      "89\n",
      "53\n",
      "86\n",
      "51\n",
      "75\n",
      "17\n",
      "72\n",
      "45\n",
      "75\n",
      "59\n",
      "72\n",
      "23\n",
      "76\n",
      "18\n",
      "94\n",
      "20\n",
      "79\n",
      "53\n",
      "72\n",
      "36\n",
      "96\n",
      "63\n",
      "80\n",
      "32\n",
      "84\n",
      "43\n",
      "70\n",
      "13\n",
      "99\n",
      "49\n",
      "66\n",
      "47\n",
      "92\n",
      "23\n",
      "90\n",
      "37\n",
      "68\n",
      "17\n",
      "78\n",
      "59\n",
      "69\n",
      "46\n",
      "93\n",
      "13\n",
      "84\n",
      "24\n",
      "65\n",
      "27\n",
      "80\n",
      "21\n",
      "96\n",
      "52\n",
      "69\n",
      "42\n",
      "66\n",
      "45\n",
      "65\n",
      "49\n",
      "68\n",
      "14\n",
      "83\n",
      "45\n",
      "66\n",
      "33\n",
      "72\n",
      "58\n",
      "73\n",
      "15\n",
      "69\n",
      "60\n",
      "91\n",
      "49\n",
      "95\n",
      "16\n",
      "81\n",
      "39\n",
      "72\n",
      "23\n",
      "74\n",
      "18\n",
      "71\n",
      "49\n",
      "92\n",
      "53\n",
      "100\n",
      "44\n",
      "76\n",
      "49\n",
      "87\n",
      "27\n",
      "75\n",
      "31\n",
      "96\n",
      "33\n",
      "86\n",
      "46\n",
      "93\n",
      "27\n",
      "87\n",
      "55\n",
      "79\n",
      "26\n",
      "66\n",
      "57\n",
      "84\n",
      "50\n",
      "79\n",
      "16\n",
      "66\n",
      "25\n",
      "83\n",
      "31\n",
      "91\n",
      "64\n",
      "83\n",
      "61\n",
      "70\n",
      "20\n",
      "87\n",
      "52\n",
      "68\n",
      "22\n",
      "65\n",
      "38\n",
      "70\n",
      "48\n",
      "80\n",
      "20\n",
      "72\n",
      "28\n",
      "71\n",
      "42\n",
      "95\n",
      "36\n",
      "95\n",
      "17\n",
      "81\n",
      "57\n",
      "66\n",
      "58\n",
      "79\n",
      "27\n",
      "89\n",
      "54\n",
      "91\n",
      "36\n",
      "76\n",
      "56\n",
      "66\n",
      "48\n",
      "90\n",
      "53\n",
      "78\n",
      "25\n",
      "97\n",
      "18\n",
      "91\n",
      "39\n",
      "70\n",
      "23\n",
      "77\n",
      "33\n",
      "92\n",
      "37\n",
      "74\n",
      "45\n",
      "87\n",
      "36\n",
      "70\n",
      "35\n",
      "85\n",
      "60\n",
      "78\n",
      "49\n",
      "77\n",
      "45\n",
      "98\n",
      "44\n",
      "67\n",
      "54\n",
      "92\n",
      "32\n",
      "78\n",
      "54\n",
      "86\n",
      "24\n",
      "73\n",
      "59\n",
      "99\n",
      "38\n",
      "76\n",
      "43\n",
      "78\n",
      "52\n",
      "80\n",
      "43\n",
      "92\n",
      "36\n",
      "73\n",
      "49\n",
      "77\n",
      "39\n",
      "76\n",
      "27\n",
      "72\n",
      "55\n",
      "90\n",
      "33\n",
      "98\n",
      "46\n",
      "89\n",
      "17\n",
      "100\n",
      "47\n",
      "83\n",
      "24\n",
      "95\n",
      "39\n",
      "89\n",
      "20\n",
      "86\n",
      "34\n",
      "86\n",
      "15\n",
      "66\n",
      "50\n",
      "91\n",
      "29\n",
      "93\n",
      "15\n",
      "86\n",
      "41\n",
      "86\n",
      "14\n",
      "83\n",
      "20\n",
      "76\n",
      "60\n",
      "77\n",
      "63\n",
      "99\n",
      "43\n",
      "100\n",
      "13\n",
      "70\n",
      "36\n",
      "90\n",
      "25\n",
      "100\n",
      "60\n",
      "80\n",
      "41\n",
      "79\n",
      "32\n",
      "71\n",
      "64\n",
      "65\n",
      "62\n",
      "70\n",
      "17\n",
      "65\n",
      "50\n",
      "86\n",
      "56\n",
      "95\n",
      "14\n",
      "85\n",
      "40\n",
      "89\n",
      "49\n",
      "87\n",
      "63\n",
      "89\n",
      "55\n",
      "91\n",
      "61\n",
      "79\n",
      "56\n",
      "72\n",
      "17\n",
      "99\n",
      "29\n",
      "79\n",
      "38\n",
      "73\n",
      "29\n",
      "78\n",
      "62\n",
      "94\n",
      "25\n",
      "100\n",
      "18\n",
      "66\n",
      "64\n",
      "95\n",
      "49\n",
      "76\n",
      "18\n",
      "69\n",
      "28\n",
      "76\n",
      "25\n",
      "70\n",
      "19\n",
      "76\n",
      "26\n",
      "76\n",
      "23\n",
      "70\n",
      "29\n",
      "76\n",
      "32\n",
      "91\n",
      "57\n",
      "68\n",
      "58\n",
      "99\n",
      "31\n",
      "82\n",
      "20\n",
      "85\n",
      "32\n",
      "67\n",
      "28\n",
      "78\n",
      "47\n",
      "96\n",
      "24\n",
      "87\n",
      "41\n",
      "72\n",
      "60\n",
      "67\n",
      "54\n",
      "72\n",
      "32\n",
      "94\n",
      "44\n",
      "81\n",
      "61\n",
      "77\n",
      "48\n",
      "95\n",
      "21\n",
      "93\n",
      "54\n",
      "78\n",
      "53\n",
      "70\n",
      "23\n",
      "84\n",
      "26\n",
      "87\n",
      "62\n",
      "78\n",
      "55\n",
      "99\n",
      "42\n",
      "91\n",
      "40\n",
      "78\n",
      "52\n",
      "91\n",
      "16\n",
      "74\n",
      "49\n",
      "91\n",
      "35\n",
      "68\n",
      "21\n",
      "94\n",
      "26\n",
      "86\n",
      "39\n",
      "99\n",
      "13\n",
      "67\n",
      "23\n",
      "88\n",
      "49\n",
      "98\n",
      "22\n",
      "67\n",
      "24\n",
      "65\n",
      "19\n",
      "97\n",
      "58\n",
      "92\n",
      "57\n",
      "89\n",
      "43\n",
      "78\n",
      "50\n",
      "87\n",
      "40\n",
      "89\n",
      "14\n",
      "75\n",
      "63\n",
      "97\n",
      "29\n",
      "81\n",
      "17\n",
      "85\n",
      "49\n",
      "85\n",
      "62\n",
      "95\n",
      "17\n",
      "65\n",
      "33\n",
      "83\n",
      "37\n",
      "73\n",
      "45\n",
      "88\n",
      "17\n",
      "96\n",
      "43\n",
      "99\n",
      "47\n",
      "97\n",
      "50\n",
      "91\n",
      "36\n",
      "100\n",
      "50\n",
      "65\n",
      "29\n",
      "93\n",
      "22\n",
      "81\n",
      "47\n",
      "66\n",
      "55\n",
      "81\n",
      "56\n",
      "72\n",
      "26\n",
      "74\n",
      "48\n",
      "78\n",
      "30\n",
      "84\n",
      "40\n",
      "93\n",
      "45\n",
      "84\n",
      "50\n",
      "69\n",
      "23\n",
      "66\n",
      "14\n",
      "94\n",
      "26\n",
      "90\n",
      "42\n",
      "65\n",
      "51\n",
      "83\n",
      "52\n",
      "100\n",
      "62\n",
      "83\n",
      "22\n",
      "75\n",
      "16\n",
      "94\n",
      "40\n",
      "95\n",
      "26\n",
      "94\n",
      "63\n",
      "91\n",
      "62\n",
      "87\n",
      "43\n",
      "95\n",
      "48\n",
      "76\n",
      "55\n",
      "77\n",
      "31\n",
      "69\n",
      "47\n",
      "96\n",
      "29\n",
      "89\n",
      "37\n",
      "83\n",
      "27\n",
      "98\n",
      "61\n",
      "66\n",
      "40\n",
      "75\n",
      "57\n",
      "68\n",
      "26\n",
      "95\n",
      "50\n",
      "89\n",
      "44\n",
      "91\n",
      "53\n",
      "97\n",
      "19\n",
      "68\n",
      "58\n",
      "86\n",
      "23\n",
      "66\n",
      "42\n",
      "88\n",
      "44\n",
      "100\n",
      "43\n",
      "99\n",
      "19\n",
      "74\n",
      "41\n",
      "86\n",
      "29\n",
      "91\n",
      "58\n",
      "83\n",
      "47\n",
      "91\n",
      "58\n",
      "99\n",
      "25\n",
      "65\n",
      "51\n",
      "82\n",
      "36\n",
      "87\n",
      "16\n",
      "73\n",
      "64\n",
      "85\n",
      "56\n",
      "78\n",
      "19\n",
      "99\n",
      "35\n",
      "82\n",
      "15\n",
      "81\n",
      "63\n",
      "81\n",
      "61\n",
      "87\n",
      "21\n",
      "79\n",
      "22\n",
      "71\n",
      "59\n",
      "71\n",
      "19\n",
      "73\n",
      "13\n",
      "76\n",
      "29\n",
      "77\n",
      "42\n",
      "74\n",
      "27\n",
      "91\n",
      "58\n",
      "97\n",
      "29\n",
      "75\n",
      "57\n",
      "66\n",
      "38\n",
      "71\n",
      "45\n",
      "90\n",
      "20\n",
      "93\n",
      "25\n",
      "94\n",
      "35\n",
      "85\n",
      "24\n",
      "80\n",
      "49\n",
      "100\n",
      "32\n",
      "94\n",
      "30\n",
      "74\n",
      "26\n",
      "69\n",
      "32\n",
      "74\n",
      "50\n",
      "68\n",
      "23\n",
      "84\n",
      "16\n",
      "97\n",
      "13\n",
      "70\n",
      "34\n",
      "87\n",
      "43\n",
      "71\n",
      "18\n",
      "96\n",
      "16\n",
      "91\n",
      "50\n",
      "79\n",
      "40\n",
      "67\n",
      "33\n",
      "98\n",
      "57\n",
      "97\n",
      "58\n",
      "99\n",
      "53\n",
      "98\n",
      "32\n",
      "74\n",
      "53\n",
      "73\n",
      "25\n",
      "98\n",
      "17\n",
      "94\n",
      "39\n",
      "80\n",
      "31\n",
      "97\n",
      "27\n",
      "73\n",
      "26\n",
      "88\n",
      "15\n",
      "93\n",
      "36\n",
      "76\n",
      "34\n",
      "68\n",
      "24\n",
      "82\n",
      "16\n",
      "77\n",
      "25\n",
      "81\n",
      "51\n",
      "71\n",
      "58\n",
      "72\n",
      "29\n",
      "88\n",
      "50\n",
      "71\n",
      "41\n",
      "71\n",
      "49\n",
      "86\n",
      "62\n",
      "81\n",
      "27\n",
      "85\n",
      "64\n",
      "97\n",
      "43\n",
      "97\n",
      "50\n",
      "69\n",
      "62\n",
      "98\n",
      "17\n",
      "87\n",
      "52\n",
      "88\n",
      "26\n",
      "81\n",
      "45\n",
      "94\n",
      "44\n",
      "91\n",
      "48\n",
      "84\n",
      "59\n",
      "74\n",
      "37\n",
      "74\n",
      "33\n",
      "66\n",
      "41\n",
      "78\n",
      "38\n",
      "98\n",
      "44\n",
      "97\n",
      "29\n",
      "89\n",
      "34\n",
      "96\n",
      "34\n",
      "95\n",
      "44\n",
      "77\n",
      "31\n",
      "89\n",
      "32\n",
      "96\n",
      "43\n",
      "79\n",
      "21\n",
      "98\n",
      "47\n",
      "100\n",
      "60\n",
      "98\n",
      "36\n",
      "99\n",
      "38\n",
      "98\n",
      "22\n",
      "88\n"
     ]
    }
   ],
   "source": [
    "#print raw data\n",
    "for i in train_samples:\n",
    "    print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "for i in train_labels:\n",
    "    print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_labels = np.array(train_labels)\n",
    "train_samples = np.array(train_samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/epoch/anaconda2/envs/fastai34/lib/python3.4/site-packages/sklearn/utils/validation.py:429: DataConversionWarning: Data with input dtype int64 was converted to float64 by MinMaxScaler.\n",
      "  warnings.warn(msg, _DataConversionWarning)\n"
     ]
    }
   ],
   "source": [
    "scaler = MinMaxScaler(feature_range=(0,1))\n",
    "scaled_train_samples = scaler.fit_transform((train_samples).reshape(-1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.47126437]\n",
      "[ 0.6091954]\n",
      "[ 0.06896552]\n",
      "[ 0.90804598]\n",
      "[ 0.35632184]\n",
      "[ 0.65517241]\n",
      "[ 0.18390805]\n",
      "[ 0.96551724]\n",
      "[ 0.05747126]\n",
      "[ 0.66666667]\n",
      "[ 0.54022989]\n",
      "[ 0.73563218]\n",
      "[ 0.14942529]\n",
      "[ 0.6091954]\n",
      "[ 0.49425287]\n",
      "[ 0.73563218]\n",
      "[ 0.08045977]\n",
      "[ 0.98850575]\n",
      "[ 0.42528736]\n",
      "[ 0.83908046]\n",
      "[ 0.18390805]\n",
      "[ 0.88505747]\n",
      "[ 0.49425287]\n",
      "[ 0.70114943]\n",
      "[ 0.12643678]\n",
      "[ 0.81609195]\n",
      "[ 0.17241379]\n",
      "[ 1.]\n",
      "[ 0.13793103]\n",
      "[ 0.6091954]\n",
      "[ 0.11494253]\n",
      "[ 0.93103448]\n",
      "[ 0.52873563]\n",
      "[ 0.81609195]\n",
      "[ 0.25287356]\n",
      "[ 0.91954023]\n",
      "[ 0.5862069]\n",
      "[ 0.89655172]\n",
      "[ 0.29885057]\n",
      "[ 0.63218391]\n",
      "[ 0.57471264]\n",
      "[ 0.96551724]\n",
      "[ 0.14942529]\n",
      "[ 0.82758621]\n",
      "[ 0.40229885]\n",
      "[ 0.93103448]\n",
      "[ 0.18390805]\n",
      "[ 0.81609195]\n",
      "[ 0.08045977]\n",
      "[ 0.97701149]\n",
      "[ 0.48275862]\n",
      "[ 0.66666667]\n",
      "[ 0.13793103]\n",
      "[ 0.89655172]\n",
      "[ 0.2183908]\n",
      "[ 0.82758621]\n",
      "[ 0.36781609]\n",
      "[ 0.7816092]\n",
      "[ 0.57471264]\n",
      "[ 0.59770115]\n",
      "[ 0.31034483]\n",
      "[ 0.7816092]\n",
      "[ 0.57471264]\n",
      "[ 0.94252874]\n",
      "[ 0.11494253]\n",
      "[ 0.68965517]\n",
      "[ 0.20689655]\n",
      "[ 0.67816092]\n",
      "[ 0.44827586]\n",
      "[ 0.93103448]\n",
      "[ 0.28735632]\n",
      "[ 0.67816092]\n",
      "[ 0.1954023]\n",
      "[ 0.89655172]\n",
      "[ 0.35632184]\n",
      "[ 0.83908046]\n",
      "[ 0.45977011]\n",
      "[ 1.]\n",
      "[ 0.24137931]\n",
      "[ 0.74712644]\n",
      "[ 0.31034483]\n",
      "[ 0.83908046]\n",
      "[ 0.05747126]\n",
      "[ 0.82758621]\n",
      "[ 0.51724138]\n",
      "[ 0.87356322]\n",
      "[ 0.4137931]\n",
      "[ 0.68965517]\n",
      "[ 0.52873563]\n",
      "[ 0.82758621]\n",
      "[ 0.57471264]\n",
      "[ 0.71264368]\n",
      "[ 0.08045977]\n",
      "[ 0.79310345]\n",
      "[ 0.06896552]\n",
      "[ 0.83908046]\n",
      "[ 0.06896552]\n",
      "[ 0.95402299]\n",
      "[ 0.28735632]\n",
      "[ 0.7816092]\n",
      "[ 0.40229885]\n",
      "[ 0.88505747]\n",
      "[ 0.47126437]\n",
      "[ 0.75862069]\n",
      "[ 0.47126437]\n",
      "[ 0.62068966]\n",
      "[ 0.24137931]\n",
      "[ 0.86206897]\n",
      "[ 0.10344828]\n",
      "[ 0.75862069]\n",
      "[ 0.05747126]\n",
      "[ 0.66666667]\n",
      "[ 0.20689655]\n",
      "[ 0.89655172]\n",
      "[ 0.56321839]\n",
      "[ 0.7816092]\n",
      "[ 0.27586207]\n",
      "[ 0.82758621]\n",
      "[ 0.32183908]\n",
      "[ 0.71264368]\n",
      "[ 0.18390805]\n",
      "[ 1.]\n",
      "[ 0.57471264]\n",
      "[ 0.72413793]\n",
      "[ 0.3908046]\n",
      "[ 1.]\n",
      "[ 0.12643678]\n",
      "[ 0.8045977]\n",
      "[ 0.10344828]\n",
      "[ 0.73563218]\n",
      "[ 0.52873563]\n",
      "[ 0.72413793]\n",
      "[ 0.1954023]\n",
      "[ 0.79310345]\n",
      "[ 0.04597701]\n",
      "[ 0.74712644]\n",
      "[ 0.02298851]\n",
      "[ 0.88505747]\n",
      "[ 0.14942529]\n",
      "[ 0.97701149]\n",
      "[ 0.35632184]\n",
      "[ 0.74712644]\n",
      "[ 0.13793103]\n",
      "[ 0.90804598]\n",
      "[ 0.17241379]\n",
      "[ 0.64367816]\n",
      "[ 0.01149425]\n",
      "[ 0.59770115]\n",
      "[ 0.57471264]\n",
      "[ 0.77011494]\n",
      "[ 0.]\n",
      "[ 0.87356322]\n",
      "[ 0.43678161]\n",
      "[ 0.59770115]\n",
      "[ 0.05747126]\n",
      "[ 0.64367816]\n",
      "[ 0.50574713]\n",
      "[ 0.68965517]\n",
      "[ 0.52873563]\n",
      "[ 0.95402299]\n",
      "[ 0.04597701]\n",
      "[ 0.79310345]\n",
      "[ 0.50574713]\n",
      "[ 0.70114943]\n",
      "[ 0.05747126]\n",
      "[ 0.91954023]\n",
      "[ 0.20689655]\n",
      "[ 0.97701149]\n",
      "[ 0.54022989]\n",
      "[ 0.97701149]\n",
      "[ 0.49425287]\n",
      "[ 0.59770115]\n",
      "[ 0.42528736]\n",
      "[ 0.67816092]\n",
      "[ 0.34482759]\n",
      "[ 0.72413793]\n",
      "[ 0.09195402]\n",
      "[ 0.68965517]\n",
      "[ 0.1954023]\n",
      "[ 0.67816092]\n",
      "[ 0.47126437]\n",
      "[ 0.67816092]\n",
      "[ 0.57471264]\n",
      "[ 0.74712644]\n",
      "[ 0.56321839]\n",
      "[ 0.93103448]\n",
      "[ 0.25287356]\n",
      "[ 0.97701149]\n",
      "[ 0.48275862]\n",
      "[ 0.77011494]\n",
      "[ 0.14942529]\n",
      "[ 0.82758621]\n",
      "[ 0.50574713]\n",
      "[ 0.65517241]\n",
      "[ 0.18390805]\n",
      "[ 0.62068966]\n",
      "[ 0.08045977]\n",
      "[ 0.77011494]\n",
      "[ 0.32183908]\n",
      "[ 0.90804598]\n",
      "[ 0.44827586]\n",
      "[ 0.77011494]\n",
      "[ 0.36781609]\n",
      "[ 0.79310345]\n",
      "[ 0.55172414]\n",
      "[ 0.70114943]\n",
      "[ 0.28735632]\n",
      "[ 0.89655172]\n",
      "[ 0.52873563]\n",
      "[ 1.]\n",
      "[ 0.52873563]\n",
      "[ 0.68965517]\n",
      "[ 0.16091954]\n",
      "[ 0.65517241]\n",
      "[ 0.32183908]\n",
      "[ 0.90804598]\n",
      "[ 0.33333333]\n",
      "[ 0.83908046]\n",
      "[ 0.12643678]\n",
      "[ 0.93103448]\n",
      "[ 0.31034483]\n",
      "[ 0.93103448]\n",
      "[ 0.4137931]\n",
      "[ 0.83908046]\n",
      "[ 0.18390805]\n",
      "[ 0.66666667]\n",
      "[ 0.36781609]\n",
      "[ 0.97701149]\n",
      "[ 0.48275862]\n",
      "[ 0.88505747]\n",
      "[ 0.20689655]\n",
      "[ 0.83908046]\n",
      "[ 0.57471264]\n",
      "[ 0.96551724]\n",
      "[ 0.06896552]\n",
      "[ 0.97701149]\n",
      "[ 0.49425287]\n",
      "[ 0.66666667]\n",
      "[ 0.01149425]\n",
      "[ 0.63218391]\n",
      "[ 0.17241379]\n",
      "[ 0.90804598]\n",
      "[ 0.56321839]\n",
      "[ 0.73563218]\n",
      "[ 0.12643678]\n",
      "[ 0.94252874]\n",
      "[ 0.33333333]\n",
      "[ 0.62068966]\n",
      "[ 0.40229885]\n",
      "[ 0.85057471]\n",
      "[ 0.18390805]\n",
      "[ 0.81609195]\n",
      "[ 0.5862069]\n",
      "[ 0.71264368]\n",
      "[ 0.36781609]\n",
      "[ 0.74712644]\n",
      "[ 0.1954023]\n",
      "[ 0.87356322]\n",
      "[ 0.2183908]\n",
      "[ 0.79310345]\n",
      "[ 0.33333333]\n",
      "[ 0.64367816]\n",
      "[ 0.17241379]\n",
      "[ 0.73563218]\n",
      "[ 0.43678161]\n",
      "[ 0.70114943]\n",
      "[ 0.20689655]\n",
      "[ 0.63218391]\n",
      "[ 0.3908046]\n",
      "[ 0.81609195]\n",
      "[ 0.49425287]\n",
      "[ 0.95402299]\n",
      "[ 0.26436782]\n",
      "[ 0.75862069]\n",
      "[ 0.17241379]\n",
      "[ 0.6091954]\n",
      "[ 0.17241379]\n",
      "[ 0.62068966]\n",
      "[ 0.54022989]\n",
      "[ 0.75862069]\n",
      "[ 0.2183908]\n",
      "[ 0.62068966]\n",
      "[ 0.20689655]\n",
      "[ 0.97701149]\n",
      "[ 0.34482759]\n",
      "[ 0.74712644]\n",
      "[ 0.27586207]\n",
      "[ 0.8045977]\n",
      "[ 0.37931034]\n",
      "[ 0.96551724]\n",
      "[ 0.42528736]\n",
      "[ 0.66666667]\n",
      "[ 0.47126437]\n",
      "[ 0.87356322]\n",
      "[ 0.24137931]\n",
      "[ 0.72413793]\n",
      "[ 0.34482759]\n",
      "[ 0.93103448]\n",
      "[ 0.11494253]\n",
      "[ 0.95402299]\n",
      "[ 0.29885057]\n",
      "[ 0.73563218]\n",
      "[ 0.54022989]\n",
      "[ 0.74712644]\n",
      "[ 0.57471264]\n",
      "[ 0.6091954]\n",
      "[ 0.42528736]\n",
      "[ 0.65517241]\n",
      "[ 0.49425287]\n",
      "[ 0.59770115]\n",
      "[ 0.25287356]\n",
      "[ 0.77011494]\n",
      "[ 0.5862069]\n",
      "[ 1.]\n",
      "[ 0.40229885]\n",
      "[ 0.66666667]\n",
      "[ 0.04597701]\n",
      "[ 0.71264368]\n",
      "[ 0.43678161]\n",
      "[ 0.79310345]\n",
      "[ 0.55172414]\n",
      "[ 0.71264368]\n",
      "[ 0.49425287]\n",
      "[ 0.70114943]\n",
      "[ 0.16091954]\n",
      "[ 0.91954023]\n",
      "[ 0.50574713]\n",
      "[ 0.65517241]\n",
      "[ 0.1954023]\n",
      "[ 0.7816092]\n",
      "[ 0.06896552]\n",
      "[ 1.]\n",
      "[ 0.16091954]\n",
      "[ 0.6091954]\n",
      "[ 0.48275862]\n",
      "[ 0.63218391]\n",
      "[ 0.57471264]\n",
      "[ 0.87356322]\n",
      "[ 0.5862069]\n",
      "[ 0.94252874]\n",
      "[ 0.1954023]\n",
      "[ 0.94252874]\n",
      "[ 0.47126437]\n",
      "[ 0.8045977]\n",
      "[ 0.32183908]\n",
      "[ 0.70114943]\n",
      "[ 0.55172414]\n",
      "[ 0.74712644]\n",
      "[ 0.32183908]\n",
      "[ 0.87356322]\n",
      "[ 0.22988506]\n",
      "[ 1.]\n",
      "[ 0.01149425]\n",
      "[ 0.59770115]\n",
      "[ 0.11494253]\n",
      "[ 0.83908046]\n",
      "[ 0.3908046]\n",
      "[ 0.95402299]\n",
      "[ 0.33333333]\n",
      "[ 0.93103448]\n",
      "[ 0.52873563]\n",
      "[ 0.97701149]\n",
      "[ 0.36781609]\n",
      "[ 0.96551724]\n",
      "[ 0.50574713]\n",
      "[ 0.6091954]\n",
      "[ 0.28735632]\n",
      "[ 0.82758621]\n",
      "[ 0.12643678]\n",
      "[ 0.83908046]\n",
      "[ 0.56321839]\n",
      "[ 0.6091954]\n",
      "[ 0.08045977]\n",
      "[ 0.70114943]\n",
      "[ 0.14942529]\n",
      "[ 0.65517241]\n",
      "[ 0.]\n",
      "[ 0.7816092]\n",
      "[ 0.29885057]\n",
      "[ 0.62068966]\n",
      "[ 0.08045977]\n",
      "[ 0.90804598]\n",
      "[ 0.18390805]\n",
      "[ 0.90804598]\n",
      "[ 0.26436782]\n",
      "[ 0.90804598]\n",
      "[ 0.20689655]\n",
      "[ 0.73563218]\n",
      "[ 0.11494253]\n",
      "[ 0.88505747]\n",
      "[ 0.06896552]\n",
      "[ 0.85057471]\n",
      "[ 0.10344828]\n",
      "[ 0.65517241]\n",
      "[ 0.57471264]\n",
      "[ 0.98850575]\n",
      "[ 0.4137931]\n",
      "[ 0.98850575]\n",
      "[ 0.04597701]\n",
      "[ 0.70114943]\n",
      "[ 0.31034483]\n",
      "[ 0.93103448]\n",
      "[ 0.31034483]\n",
      "[ 0.6091954]\n",
      "[ 0.52873563]\n",
      "[ 0.94252874]\n",
      "[ 0.35632184]\n",
      "[ 0.91954023]\n",
      "[ 0.13793103]\n",
      "[ 0.64367816]\n",
      "[ 0.03448276]\n",
      "[ 0.59770115]\n",
      "[ 0.]\n",
      "[ 0.85057471]\n",
      "[ 0.43678161]\n",
      "[ 0.62068966]\n",
      "[ 0.40229885]\n",
      "[ 0.96551724]\n",
      "[ 0.43678161]\n",
      "[ 1.]\n",
      "[ 0.52873563]\n",
      "[ 0.96551724]\n",
      "[ 0.42528736]\n",
      "[ 0.96551724]\n",
      "[ 0.]\n",
      "[ 1.]\n",
      "[ 0.09195402]\n",
      "[ 0.89655172]\n",
      "[ 0.16091954]\n",
      "[ 0.94252874]\n",
      "[ 0.3908046]\n",
      "[ 0.75862069]\n",
      "[ 0.5862069]\n",
      "[ 0.73563218]\n",
      "[ 0.25287356]\n",
      "[ 0.70114943]\n",
      "[ 0.43678161]\n",
      "[ 0.81609195]\n",
      "[ 0.24137931]\n",
      "[ 0.70114943]\n",
      "[ 0.3908046]\n",
      "[ 0.68965517]\n",
      "[ 0.20689655]\n",
      "[ 0.98850575]\n",
      "[ 0.06896552]\n",
      "[ 0.93103448]\n",
      "[ 0.05747126]\n",
      "[ 0.83908046]\n",
      "[ 0.02298851]\n",
      "[ 0.90804598]\n",
      "[ 0.42528736]\n",
      "[ 0.93103448]\n",
      "[ 0.42528736]\n",
      "[ 0.91954023]\n",
      "[ 0.28735632]\n",
      "[ 0.73563218]\n",
      "[ 0.05747126]\n",
      "[ 0.67816092]\n",
      "[ 0.37931034]\n",
      "[ 0.71264368]\n",
      "[ 0.25287356]\n",
      "[ 0.79310345]\n",
      "[ 0.25287356]\n",
      "[ 0.90804598]\n",
      "[ 0.32183908]\n",
      "[ 0.73563218]\n",
      "[ 0.50574713]\n",
      "[ 0.96551724]\n",
      "[ 0.56321839]\n",
      "[ 0.68965517]\n",
      "[ 0.14942529]\n",
      "[ 0.96551724]\n",
      "[ 0.09195402]\n",
      "[ 0.95402299]\n",
      "[ 0.3908046]\n",
      "[ 0.62068966]\n",
      "[ 0.47126437]\n",
      "[ 1.]\n",
      "[ 0.44827586]\n",
      "[ 0.62068966]\n",
      "[ 0.08045977]\n",
      "[ 0.63218391]\n",
      "[ 0.]\n",
      "[ 0.93103448]\n",
      "[ 0.57471264]\n",
      "[ 0.96551724]\n",
      "[ 0.49425287]\n",
      "[ 0.94252874]\n",
      "[ 0.44827586]\n",
      "[ 0.66666667]\n",
      "[ 0.29885057]\n",
      "[ 0.70114943]\n",
      "[ 0.27586207]\n",
      "[ 0.67816092]\n",
      "[ 0.02298851]\n",
      "[ 0.79310345]\n",
      "[ 0.2183908]\n",
      "[ 0.71264368]\n",
      "[ 0.10344828]\n",
      "[ 0.77011494]\n",
      "[ 0.55172414]\n",
      "[ 0.87356322]\n",
      "[ 0.02298851]\n",
      "[ 0.81609195]\n",
      "[ 0.25287356]\n",
      "[ 0.81609195]\n",
      "[ 0.03448276]\n",
      "[ 0.59770115]\n",
      "[ 0.1954023]\n",
      "[ 0.82758621]\n",
      "[ 0.25287356]\n",
      "[ 0.82758621]\n",
      "[ 0.50574713]\n",
      "[ 0.83908046]\n",
      "[ 0.18390805]\n",
      "[ 0.91954023]\n",
      "[ 0.57471264]\n",
      "[ 0.85057471]\n",
      "[ 0.4137931]\n",
      "[ 0.71264368]\n",
      "[ 0.09195402]\n",
      "[ 0.97701149]\n",
      "[ 0.2183908]\n",
      "[ 0.75862069]\n",
      "[ 0.22988506]\n",
      "[ 0.95402299]\n",
      "[ 0.47126437]\n",
      "[ 0.90804598]\n",
      "[ 0.05747126]\n",
      "[ 0.68965517]\n",
      "[ 0.37931034]\n",
      "[ 0.74712644]\n",
      "[ 0.20689655]\n",
      "[ 0.74712644]\n",
      "[ 0.49425287]\n",
      "[ 0.87356322]\n",
      "[ 0.56321839]\n",
      "[ 0.89655172]\n",
      "[ 0.12643678]\n",
      "[ 0.6091954]\n",
      "[ 0.49425287]\n",
      "[ 0.74712644]\n",
      "[ 0.05747126]\n",
      "[ 0.83908046]\n",
      "[ 0.20689655]\n",
      "[ 0.82758621]\n",
      "[ 0.13793103]\n",
      "[ 0.85057471]\n",
      "[ 0.31034483]\n",
      "[ 0.89655172]\n",
      "[ 0.37931034]\n",
      "[ 0.72413793]\n",
      "[ 0.04597701]\n",
      "[ 0.6091954]\n",
      "[ 0.20689655]\n",
      "[ 0.8045977]\n",
      "[ 0.10344828]\n",
      "[ 0.77011494]\n",
      "[ 0.37931034]\n",
      "[ 0.72413793]\n",
      "[ 0.03448276]\n",
      "[ 0.90804598]\n",
      "[ 0.13793103]\n",
      "[ 0.68965517]\n",
      "[ 0.1954023]\n",
      "[ 0.68965517]\n",
      "[ 0.37931034]\n",
      "[ 0.66666667]\n",
      "[ 0.3908046]\n",
      "[ 0.75862069]\n",
      "[ 0.50574713]\n",
      "[ 0.96551724]\n",
      "[ 0.25287356]\n",
      "[ 0.65517241]\n",
      "[ 0.27586207]\n",
      "[ 0.98850575]\n",
      "[ 0.57471264]\n",
      "[ 0.65517241]\n",
      "[ 0.55172414]\n",
      "[ 0.93103448]\n",
      "[ 0.55172414]\n",
      "[ 0.91954023]\n",
      "[ 0.2183908]\n",
      "[ 0.65517241]\n",
      "[ 0.28735632]\n",
      "[ 1.]\n",
      "[ 0.14942529]\n",
      "[ 0.94252874]\n",
      "[ 0.10344828]\n",
      "[ 0.98850575]\n",
      "[ 0.11494253]\n",
      "[ 0.72413793]\n",
      "[ 0.52873563]\n",
      "[ 0.98850575]\n",
      "[ 0.2183908]\n",
      "[ 0.7816092]\n",
      "[ 0.26436782]\n",
      "[ 0.83908046]\n",
      "[ 0.03448276]\n",
      "[ 0.6091954]\n",
      "[ 0.54022989]\n",
      "[ 0.94252874]\n",
      "[ 0.24137931]\n",
      "[ 0.6091954]\n",
      "[ 0.45977011]\n",
      "[ 0.82758621]\n",
      "[ 0.04597701]\n",
      "[ 0.88505747]\n",
      "[ 0.54022989]\n",
      "[ 1.]\n",
      "[ 0.28735632]\n",
      "[ 0.95402299]\n",
      "[ 0.32183908]\n",
      "[ 0.94252874]\n",
      "[ 0.50574713]\n",
      "[ 0.89655172]\n",
      "[ 0.48275862]\n",
      "[ 0.88505747]\n",
      "[ 0.]\n",
      "[ 0.81609195]\n",
      "[ 0.48275862]\n",
      "[ 0.72413793]\n",
      "[ 0.2183908]\n",
      "[ 0.6091954]\n",
      "[ 0.14942529]\n",
      "[ 0.90804598]\n",
      "[ 0.09195402]\n",
      "[ 0.74712644]\n",
      "[ 0.20689655]\n",
      "[ 0.89655172]\n",
      "[ 0.51724138]\n",
      "[ 0.70114943]\n",
      "[ 0.56321839]\n",
      "[ 0.7816092]\n",
      "[ 0.34482759]\n",
      "[ 0.89655172]\n",
      "[ 0.20689655]\n",
      "[ 0.77011494]\n",
      "[ 0.40229885]\n",
      "[ 0.70114943]\n",
      "[ 0.28735632]\n",
      "[ 0.88505747]\n",
      "[ 0.09195402]\n",
      "[ 0.62068966]\n",
      "[ 0.18390805]\n",
      "[ 0.65517241]\n",
      "[ 0.52873563]\n",
      "[ 0.93103448]\n",
      "[ 0.06896552]\n",
      "[ 1.]\n",
      "[ 0.47126437]\n",
      "[ 0.66666667]\n",
      "[ 0.14942529]\n",
      "[ 0.68965517]\n",
      "[ 0.5862069]\n",
      "[ 0.77011494]\n",
      "[ 0.10344828]\n",
      "[ 0.97701149]\n",
      "[ 0.52873563]\n",
      "[ 0.97701149]\n",
      "[ 0.04597701]\n",
      "[ 0.98850575]\n",
      "[ 0.31034483]\n",
      "[ 0.6091954]\n",
      "[ 0.1954023]\n",
      "[ 0.93103448]\n",
      "[ 0.20689655]\n",
      "[ 0.59770115]\n",
      "[ 0.43678161]\n",
      "[ 0.98850575]\n",
      "[ 0.33333333]\n",
      "[ 0.87356322]\n",
      "[ 0.35632184]\n",
      "[ 0.72413793]\n",
      "[ 0.56321839]\n",
      "[ 0.87356322]\n",
      "[ 0.09195402]\n",
      "[ 0.75862069]\n",
      "[ 0.51724138]\n",
      "[ 0.83908046]\n",
      "[ 0.32183908]\n",
      "[ 0.81609195]\n",
      "[ 0.47126437]\n",
      "[ 0.65517241]\n",
      "[ 0.01149425]\n",
      "[ 1.]\n",
      "[ 0.47126437]\n",
      "[ 0.83908046]\n",
      "[ 0.06896552]\n",
      "[ 1.]\n",
      "[ 0.45977011]\n",
      "[ 0.91954023]\n",
      "[ 0.]\n",
      "[ 0.82758621]\n",
      "[ 0.25287356]\n",
      "[ 1.]\n",
      "[ 0.01149425]\n",
      "[ 0.74712644]\n",
      "[ 0.57471264]\n",
      "[ 0.94252874]\n",
      "[ 0.18390805]\n",
      "[ 0.66666667]\n",
      "[ 0.28735632]\n",
      "[ 0.98850575]\n",
      "[ 0.3908046]\n",
      "[ 0.6091954]\n",
      "[ 0.37931034]\n",
      "[ 0.65517241]\n",
      "[ 0.]\n",
      "[ 0.82758621]\n",
      "[ 0.55172414]\n",
      "[ 0.59770115]\n",
      "[ 0.3908046]\n",
      "[ 0.82758621]\n",
      "[ 0.33333333]\n",
      "[ 0.6091954]\n",
      "[ 0.27586207]\n",
      "[ 0.97701149]\n",
      "[ 0.05747126]\n",
      "[ 0.74712644]\n",
      "[ 0.25287356]\n",
      "[ 0.67816092]\n",
      "[ 0.5862069]\n",
      "[ 0.89655172]\n",
      "[ 0.47126437]\n",
      "[ 0.91954023]\n",
      "[ 0.3908046]\n",
      "[ 0.90804598]\n",
      "[ 0.57471264]\n",
      "[ 0.82758621]\n",
      "[ 0.45977011]\n",
      "[ 0.7816092]\n",
      "[ 0.29885057]\n",
      "[ 0.89655172]\n",
      "[ 0.47126437]\n",
      "[ 0.93103448]\n",
      "[ 0.01149425]\n",
      "[ 0.72413793]\n",
      "[ 0.09195402]\n",
      "[ 0.98850575]\n",
      "[ 0.18390805]\n",
      "[ 0.90804598]\n",
      "[ 0.37931034]\n",
      "[ 0.94252874]\n",
      "[ 0.14942529]\n",
      "[ 0.94252874]\n",
      "[ 0.45977011]\n",
      "[ 0.7816092]\n",
      "[ 0.33333333]\n",
      "[ 0.77011494]\n",
      "[ 0.24137931]\n",
      "[ 0.89655172]\n",
      "[ 0.4137931]\n",
      "[ 0.91954023]\n",
      "[ 0.08045977]\n",
      "[ 0.74712644]\n",
      "[ 0.28735632]\n",
      "[ 0.98850575]\n",
      "[ 0.28735632]\n",
      "[ 0.90804598]\n",
      "[ 0.06896552]\n",
      "[ 0.70114943]\n",
      "[ 0.47126437]\n",
      "[ 0.75862069]\n",
      "[ 0.57471264]\n",
      "[ 0.66666667]\n",
      "[ 0.34482759]\n",
      "[ 0.64367816]\n",
      "[ 0.33333333]\n",
      "[ 0.98850575]\n",
      "[ 0.04597701]\n",
      "[ 0.8045977]\n",
      "[ 0.27586207]\n",
      "[ 0.65517241]\n",
      "[ 0.49425287]\n",
      "[ 0.98850575]\n",
      "[ 0.5862069]\n",
      "[ 0.81609195]\n",
      "[ 0.57471264]\n",
      "[ 0.91954023]\n",
      "[ 0.03448276]\n",
      "[ 0.87356322]\n",
      "[ 0.54022989]\n",
      "[ 0.66666667]\n",
      "[ 0.16091954]\n",
      "[ 0.83908046]\n",
      "[ 0.17241379]\n",
      "[ 0.96551724]\n",
      "[ 0.09195402]\n",
      "[ 0.68965517]\n",
      "[ 0.06896552]\n",
      "[ 0.91954023]\n",
      "[ 0.13793103]\n",
      "[ 0.74712644]\n",
      "[ 0.18390805]\n",
      "[ 0.77011494]\n",
      "[ 0.10344828]\n",
      "[ 0.73563218]\n",
      "[ 0.34482759]\n",
      "[ 0.83908046]\n",
      "[ 0.44827586]\n",
      "[ 0.89655172]\n",
      "[ 0.36781609]\n",
      "[ 0.72413793]\n",
      "[ 0.20689655]\n",
      "[ 0.94252874]\n",
      "[ 0.16091954]\n",
      "[ 0.88505747]\n",
      "[ 0.2183908]\n",
      "[ 1.]\n",
      "[ 0.20689655]\n",
      "[ 0.71264368]\n",
      "[ 0.34482759]\n",
      "[ 0.83908046]\n",
      "[ 0.13793103]\n",
      "[ 0.81609195]\n",
      "[ 0.05747126]\n",
      "[ 0.65517241]\n",
      "[ 0.29885057]\n",
      "[ 0.94252874]\n",
      "[ 0.03448276]\n",
      "[ 0.79310345]\n",
      "[ 0.11494253]\n",
      "[ 1.]\n",
      "[ 0.5862069]\n",
      "[ 0.63218391]\n",
      "[ 0.56321839]\n",
      "[ 0.87356322]\n",
      "[ 0.03448276]\n",
      "[ 0.6091954]\n",
      "[ 0.35632184]\n",
      "[ 0.7816092]\n",
      "[ 0.01149425]\n",
      "[ 0.88505747]\n",
      "[ 0.14942529]\n",
      "[ 0.94252874]\n",
      "[ 0.45977011]\n",
      "[ 0.85057471]\n",
      "[ 0.37931034]\n",
      "[ 0.73563218]\n",
      "[ 0.36781609]\n",
      "[ 0.87356322]\n",
      "[ 0.05747126]\n",
      "[ 0.77011494]\n",
      "[ 0.24137931]\n",
      "[ 0.96551724]\n",
      "[ 0.48275862]\n",
      "[ 0.75862069]\n",
      "[ 0.4137931]\n",
      "[ 0.7816092]\n",
      "[ 0.52873563]\n",
      "[ 0.93103448]\n",
      "[ 0.24137931]\n",
      "[ 0.93103448]\n",
      "[ 0.14942529]\n",
      "[ 0.72413793]\n",
      "[ 0.25287356]\n",
      "[ 0.77011494]\n",
      "[ 0.49425287]\n",
      "[ 0.6091954]\n",
      "[ 0.32183908]\n",
      "[ 0.7816092]\n",
      "[ 0.20689655]\n",
      "[ 0.7816092]\n",
      "[ 0.40229885]\n",
      "[ 0.68965517]\n",
      "[ 0.48275862]\n",
      "[ 0.89655172]\n",
      "[ 0.27586207]\n",
      "[ 0.91954023]\n",
      "[ 0.51724138]\n",
      "[ 0.72413793]\n",
      "[ 0.09195402]\n",
      "[ 0.96551724]\n",
      "[ 0.31034483]\n",
      "[ 0.94252874]\n",
      "[ 0.36781609]\n",
      "[ 0.66666667]\n",
      "[ 0.40229885]\n",
      "[ 0.65517241]\n",
      "[ 0.5862069]\n",
      "[ 0.86206897]\n",
      "[ 0.32183908]\n",
      "[ 0.63218391]\n",
      "[ 0.35632184]\n",
      "[ 0.91954023]\n",
      "[ 0.05747126]\n",
      "[ 0.81609195]\n",
      "[ 0.22988506]\n",
      "[ 0.75862069]\n",
      "[ 0.11494253]\n",
      "[ 0.96551724]\n",
      "[ 0.11494253]\n",
      "[ 0.98850575]\n",
      "[ 0.57471264]\n",
      "[ 0.95402299]\n",
      "[ 0.31034483]\n",
      "[ 0.6091954]\n",
      "[ 0.56321839]\n",
      "[ 0.6091954]\n",
      "[ 0.4137931]\n",
      "[ 0.96551724]\n",
      "[ 0.05747126]\n",
      "[ 0.8045977]\n",
      "[ 0.4137931]\n",
      "[ 0.7816092]\n",
      "[ 0.06896552]\n",
      "[ 0.88505747]\n",
      "[ 0.54022989]\n",
      "[ 0.91954023]\n",
      "[ 0.]\n",
      "[ 0.8045977]\n",
      "[ 0.45977011]\n",
      "[ 0.95402299]\n",
      "[ 0.06896552]\n",
      "[ 0.96551724]\n",
      "[ 0.36781609]\n",
      "[ 0.85057471]\n",
      "[ 0.50574713]\n",
      "[ 0.65517241]\n",
      "[ 0.01149425]\n",
      "[ 0.94252874]\n",
      "[ 0.54022989]\n",
      "[ 0.86206897]\n",
      "[ 0.4137931]\n",
      "[ 0.79310345]\n",
      "[ 0.09195402]\n",
      "[ 0.91954023]\n",
      "[ 0.09195402]\n",
      "[ 1.]\n",
      "[ 0.51724138]\n",
      "[ 0.82758621]\n",
      "[ 0.52873563]\n",
      "[ 0.74712644]\n",
      "[ 0.22988506]\n",
      "[ 0.73563218]\n",
      "[ 0.09195402]\n",
      "[ 0.73563218]\n",
      "[ 0.26436782]\n",
      "[ 0.89655172]\n",
      "[ 0.05747126]\n",
      "[ 0.95402299]\n",
      "[ 0.09195402]\n",
      "[ 0.7816092]\n",
      "[ 0.01149425]\n",
      "[ 0.85057471]\n",
      "[ 0.]\n",
      "[ 0.94252874]\n",
      "[ 0.34482759]\n",
      "[ 0.6091954]\n",
      "[ 0.44827586]\n",
      "[ 0.81609195]\n",
      "[ 0.33333333]\n",
      "[ 0.89655172]\n",
      "[ 0.36781609]\n",
      "[ 0.63218391]\n",
      "[ 0.33333333]\n",
      "[ 0.93103448]\n",
      "[ 0.45977011]\n",
      "[ 0.95402299]\n",
      "[ 0.26436782]\n",
      "[ 0.72413793]\n",
      "[ 0.22988506]\n",
      "[ 0.94252874]\n",
      "[ 0.47126437]\n",
      "[ 0.88505747]\n",
      "[ 0.34482759]\n",
      "[ 0.74712644]\n",
      "[ 0.43678161]\n",
      "[ 0.89655172]\n",
      "[ 0.48275862]\n",
      "[ 0.97701149]\n",
      "[ 0.3908046]\n",
      "[ 0.8045977]\n",
      "[ 0.52873563]\n",
      "[ 0.59770115]\n",
      "[ 0.26436782]\n",
      "[ 0.97701149]\n",
      "[ 0.51724138]\n",
      "[ 1.]\n",
      "[ 0.43678161]\n",
      "[ 0.67816092]\n",
      "[ 0.48275862]\n",
      "[ 0.65517241]\n",
      "[ 0.29885057]\n",
      "[ 0.77011494]\n",
      "[ 0.5862069]\n",
      "[ 0.94252874]\n",
      "[ 0.5862069]\n",
      "[ 0.73563218]\n",
      "[ 0.26436782]\n",
      "[ 0.59770115]\n",
      "[ 0.13793103]\n",
      "[ 0.81609195]\n",
      "[ 0.05747126]\n",
      "[ 0.95402299]\n",
      "[ 0.2183908]\n",
      "[ 0.8045977]\n",
      "[ 0.29885057]\n",
      "[ 0.88505747]\n",
      "[ 0.03448276]\n",
      "[ 0.97701149]\n",
      "[ 0.10344828]\n",
      "[ 0.93103448]\n",
      "[ 0.03448276]\n",
      "[ 0.85057471]\n",
      "[ 0.08045977]\n",
      "[ 0.98850575]\n",
      "[ 0.5862069]\n",
      "[ 0.88505747]\n",
      "[ 0.09195402]\n",
      "[ 0.88505747]\n",
      "[ 0.14942529]\n",
      "[ 0.90804598]\n",
      "[ 0.08045977]\n",
      "[ 0.75862069]\n",
      "[ 0.06896552]\n",
      "[ 0.90804598]\n",
      "[ 0.25287356]\n",
      "[ 0.98850575]\n",
      "[ 0.20689655]\n",
      "[ 0.86206897]\n",
      "[ 0.44827586]\n",
      "[ 0.83908046]\n",
      "[ 0.14942529]\n",
      "[ 0.98850575]\n",
      "[ 0.29885057]\n",
      "[ 0.87356322]\n",
      "[ 0.16091954]\n",
      "[ 0.75862069]\n",
      "[ 0.17241379]\n",
      "[ 0.71264368]\n",
      "[ 0.49425287]\n",
      "[ 0.75862069]\n",
      "[ 0.12643678]\n",
      "[ 0.6091954]\n",
      "[ 0.11494253]\n",
      "[ 0.87356322]\n",
      "[ 0.54022989]\n",
      "[ 0.62068966]\n",
      "[ 0.48275862]\n",
      "[ 0.88505747]\n",
      "[ 0.18390805]\n",
      "[ 0.8045977]\n",
      "[ 0.35632184]\n",
      "[ 0.65517241]\n",
      "[ 0.01149425]\n",
      "[ 0.67816092]\n",
      "[ 0.20689655]\n",
      "[ 0.74712644]\n",
      "[ 0.04597701]\n",
      "[ 0.87356322]\n",
      "[ 0.17241379]\n",
      "[ 0.95402299]\n",
      "[ 0.09195402]\n",
      "[ 0.82758621]\n",
      "[ 0.44827586]\n",
      "[ 0.68965517]\n",
      "[ 0.16091954]\n",
      "[ 0.81609195]\n",
      "[ 0.52873563]\n",
      "[ 0.71264368]\n",
      "[ 0.49425287]\n",
      "[ 0.65517241]\n",
      "[ 0.05747126]\n",
      "[ 0.59770115]\n",
      "[ 0.17241379]\n",
      "[ 0.87356322]\n",
      "[ 0.5862069]\n",
      "[ 0.70114943]\n",
      "[ 0.45977011]\n",
      "[ 0.77011494]\n",
      "[ 0.24137931]\n",
      "[ 0.97701149]\n",
      "[ 0.50574713]\n",
      "[ 0.94252874]\n",
      "[ 0.27586207]\n",
      "[ 0.83908046]\n",
      "[ 0.49425287]\n",
      "[ 0.6091954]\n",
      "[ 0.01149425]\n",
      "[ 0.6091954]\n",
      "[ 0.12643678]\n",
      "[ 0.71264368]\n",
      "[ 0.17241379]\n",
      "[ 0.66666667]\n",
      "[ 0.40229885]\n",
      "[ 0.67816092]\n",
      "[ 0.45977011]\n",
      "[ 0.71264368]\n",
      "[ 0.29885057]\n",
      "[ 0.79310345]\n",
      "[ 0.09195402]\n",
      "[ 0.93103448]\n",
      "[ 0.02298851]\n",
      "[ 0.88505747]\n",
      "[ 0.10344828]\n",
      "[ 0.98850575]\n",
      "[ 0.50574713]\n",
      "[ 0.89655172]\n",
      "[ 0.10344828]\n",
      "[ 0.62068966]\n",
      "[ 0.5862069]\n",
      "[ 0.64367816]\n",
      "[ 0.05747126]\n",
      "[ 0.98850575]\n",
      "[ 0.16091954]\n",
      "[ 0.7816092]\n",
      "[ 0.37931034]\n",
      "[ 0.7816092]\n",
      "[ 0.2183908]\n",
      "[ 0.88505747]\n",
      "[ 0.09195402]\n",
      "[ 0.90804598]\n",
      "[ 0.24137931]\n",
      "[ 0.97701149]\n",
      "[ 0.36781609]\n",
      "[ 0.98850575]\n",
      "[ 0.55172414]\n",
      "[ 0.71264368]\n",
      "[ 0.48275862]\n",
      "[ 0.59770115]\n",
      "[ 0.03448276]\n",
      "[ 0.59770115]\n",
      "[ 0.34482759]\n",
      "[ 0.93103448]\n",
      "[ 0.55172414]\n",
      "[ 0.82758621]\n",
      "[ 0.18390805]\n",
      "[ 0.68965517]\n",
      "[ 0.17241379]\n",
      "[ 0.73563218]\n",
      "[ 0.56321839]\n",
      "[ 0.79310345]\n",
      "[ 0.52873563]\n",
      "[ 0.81609195]\n",
      "[ 0.52873563]\n",
      "[ 0.97701149]\n",
      "[ 0.47126437]\n",
      "[ 0.73563218]\n",
      "[ 0.5862069]\n",
      "[ 0.87356322]\n",
      "[ 0.1954023]\n",
      "[ 0.94252874]\n",
      "[ 0.49425287]\n",
      "[ 0.90804598]\n",
      "[ 0.47126437]\n",
      "[ 0.90804598]\n",
      "[ 0.25287356]\n",
      "[ 0.94252874]\n",
      "[ 0.48275862]\n",
      "[ 0.73563218]\n",
      "[ 0.1954023]\n",
      "[ 0.82758621]\n",
      "[ 0.29885057]\n",
      "[ 0.77011494]\n",
      "[ 0.28735632]\n",
      "[ 0.8045977]\n",
      "[ 0.35632184]\n",
      "[ 0.79310345]\n",
      "[ 0.45977011]\n",
      "[ 0.66666667]\n",
      "[ 0.27586207]\n",
      "[ 0.59770115]\n",
      "[ 0.47126437]\n",
      "[ 0.91954023]\n",
      "[ 0.29885057]\n",
      "[ 0.95402299]\n",
      "[ 0.49425287]\n",
      "[ 0.75862069]\n",
      "[ 0.03448276]\n",
      "[ 0.96551724]\n",
      "[ 0.35632184]\n",
      "[ 0.72413793]\n",
      "[ 0.03448276]\n",
      "[ 0.62068966]\n",
      "[ 0.51724138]\n",
      "[ 0.87356322]\n",
      "[ 0.56321839]\n",
      "[ 0.90804598]\n",
      "[ 0.28735632]\n",
      "[ 0.83908046]\n",
      "[ 0.2183908]\n",
      "[ 0.89655172]\n",
      "[ 0.13793103]\n",
      "[ 0.94252874]\n",
      "[ 0.40229885]\n",
      "[ 0.63218391]\n",
      "[ 0.25287356]\n",
      "[ 0.82758621]\n",
      "[ 0.26436782]\n",
      "[ 0.66666667]\n",
      "[ 0.51724138]\n",
      "[ 0.97701149]\n",
      "[ 0.44827586]\n",
      "[ 0.72413793]\n",
      "[ 0.02298851]\n",
      "[ 0.81609195]\n",
      "[ 0.02298851]\n",
      "[ 0.96551724]\n",
      "[ 0.12643678]\n",
      "[ 0.83908046]\n",
      "[ 0.37931034]\n",
      "[ 0.7816092]\n",
      "[ 0.33333333]\n",
      "[ 0.75862069]\n",
      "[ 0.27586207]\n",
      "[ 0.64367816]\n",
      "[ 0.01149425]\n",
      "[ 0.73563218]\n",
      "[ 0.17241379]\n",
      "[ 0.91954023]\n",
      "[ 0.36781609]\n",
      "[ 0.88505747]\n",
      "[ 0.05747126]\n",
      "[ 0.74712644]\n",
      "[ 0.16091954]\n",
      "[ 0.82758621]\n",
      "[ 0.10344828]\n",
      "[ 0.94252874]\n",
      "[ 0.43678161]\n",
      "[ 0.65517241]\n",
      "[ 0.26436782]\n",
      "[ 0.75862069]\n",
      "[ 0.29885057]\n",
      "[ 1.]\n",
      "[ 0.26436782]\n",
      "[ 0.83908046]\n",
      "[ 0.4137931]\n",
      "[ 0.82758621]\n",
      "[ 0.09195402]\n",
      "[ 0.74712644]\n",
      "[ 0.14942529]\n",
      "[ 0.71264368]\n",
      "[ 0.08045977]\n",
      "[ 0.63218391]\n",
      "[ 0.36781609]\n",
      "[ 0.72413793]\n",
      "[ 0.28735632]\n",
      "[ 0.88505747]\n",
      "[ 0.13793103]\n",
      "[ 1.]\n",
      "[ 0.05747126]\n",
      "[ 1.]\n",
      "[ 0.45977011]\n",
      "[ 0.71264368]\n",
      "[ 0.43678161]\n",
      "[ 0.77011494]\n",
      "[ 0.03448276]\n",
      "[ 0.75862069]\n",
      "[ 0.3908046]\n",
      "[ 0.81609195]\n",
      "[ 0.57471264]\n",
      "[ 0.6091954]\n",
      "[ 0.10344828]\n",
      "[ 0.63218391]\n",
      "[ 0.11494253]\n",
      "[ 0.83908046]\n",
      "[ 0.06896552]\n",
      "[ 0.91954023]\n",
      "[ 0.50574713]\n",
      "[ 0.64367816]\n",
      "[ 0.36781609]\n",
      "[ 0.97701149]\n",
      "[ 0.12643678]\n",
      "[ 0.94252874]\n",
      "[ 0.51724138]\n",
      "[ 0.70114943]\n",
      "[ 0.02298851]\n",
      "[ 0.73563218]\n",
      "[ 0.03448276]\n",
      "[ 0.95402299]\n",
      "[ 0.26436782]\n",
      "[ 0.82758621]\n",
      "[ 0.36781609]\n",
      "[ 0.8045977]\n",
      "[ 0.27586207]\n",
      "[ 0.68965517]\n",
      "[ 0.57471264]\n",
      "[ 0.87356322]\n",
      "[ 0.05747126]\n",
      "[ 0.74712644]\n",
      "[ 0.02298851]\n",
      "[ 0.81609195]\n",
      "[ 0.01149425]\n",
      "[ 0.95402299]\n",
      "[ 0.57471264]\n",
      "[ 0.67816092]\n",
      "[ 0.05747126]\n",
      "[ 0.65517241]\n",
      "[ 0.18390805]\n",
      "[ 0.94252874]\n",
      "[ 0.]\n",
      "[ 0.88505747]\n",
      "[ 0.17241379]\n",
      "[ 0.97701149]\n",
      "[ 0.18390805]\n",
      "[ 0.62068966]\n",
      "[ 0.09195402]\n",
      "[ 0.74712644]\n",
      "[ 0.25287356]\n",
      "[ 0.86206897]\n",
      "[ 0.55172414]\n",
      "[ 0.73563218]\n",
      "[ 0.05747126]\n",
      "[ 0.77011494]\n",
      "[ 0.13793103]\n",
      "[ 0.73563218]\n",
      "[ 0.48275862]\n",
      "[ 0.6091954]\n",
      "[ 0.17241379]\n",
      "[ 0.67816092]\n",
      "[ 0.45977011]\n",
      "[ 0.67816092]\n",
      "[ 0.43678161]\n",
      "[ 0.89655172]\n",
      "[ 0.26436782]\n",
      "[ 0.6091954]\n",
      "[ 0.14942529]\n",
      "[ 0.97701149]\n",
      "[ 0.29885057]\n",
      "[ 0.79310345]\n",
      "[ 0.02298851]\n",
      "[ 0.96551724]\n",
      "[ 0.1954023]\n",
      "[ 0.74712644]\n",
      "[ 0.25287356]\n",
      "[ 0.94252874]\n",
      "[ 0.12643678]\n",
      "[ 0.81609195]\n",
      "[ 0.04597701]\n",
      "[ 0.63218391]\n",
      "[ 0.45977011]\n",
      "[ 0.90804598]\n",
      "[ 0.20689655]\n",
      "[ 0.8045977]\n",
      "[ 0.]\n",
      "[ 0.8045977]\n",
      "[ 0.05747126]\n",
      "[ 0.70114943]\n",
      "[ 0.1954023]\n",
      "[ 0.82758621]\n",
      "[ 0.10344828]\n",
      "[ 0.81609195]\n",
      "[ 0.06896552]\n",
      "[ 0.67816092]\n",
      "[ 0.4137931]\n",
      "[ 0.86206897]\n",
      "[ 0.22988506]\n",
      "[ 0.97701149]\n",
      "[ 0.48275862]\n",
      "[ 0.72413793]\n",
      "[ 0.50574713]\n",
      "[ 0.75862069]\n",
      "[ 0.01149425]\n",
      "[ 0.63218391]\n",
      "[ 0.01149425]\n",
      "[ 0.77011494]\n",
      "[ 0.36781609]\n",
      "[ 0.89655172]\n",
      "[ 0.20689655]\n",
      "[ 0.6091954]\n",
      "[ 0.17241379]\n",
      "[ 0.66666667]\n",
      "[ 0.28735632]\n",
      "[ 0.6091954]\n",
      "[ 0.56321839]\n",
      "[ 0.97701149]\n",
      "[ 0.25287356]\n",
      "[ 0.85057471]\n",
      "[ 0.10344828]\n",
      "[ 0.89655172]\n",
      "[ 0.03448276]\n",
      "[ 0.89655172]\n",
      "[ 0.37931034]\n",
      "[ 0.89655172]\n",
      "[ 0.14942529]\n",
      "[ 0.94252874]\n",
      "[ 0.17241379]\n",
      "[ 0.85057471]\n",
      "[ 0.12643678]\n",
      "[ 0.82758621]\n",
      "[ 0.2183908]\n",
      "[ 0.7816092]\n",
      "[ 0.4137931]\n",
      "[ 0.94252874]\n",
      "[ 0.22988506]\n",
      "[ 0.93103448]\n",
      "[ 0.14942529]\n",
      "[ 0.66666667]\n",
      "[ 0.51724138]\n",
      "[ 0.6091954]\n",
      "[ 0.33333333]\n",
      "[ 1.]\n",
      "[ 0.10344828]\n",
      "[ 0.73563218]\n",
      "[ 0.]\n",
      "[ 0.96551724]\n",
      "[ 0.09195402]\n",
      "[ 0.73563218]\n",
      "[ 0.48275862]\n",
      "[ 0.98850575]\n",
      "[ 0.5862069]\n",
      "[ 0.93103448]\n",
      "[ 0.34482759]\n",
      "[ 0.94252874]\n",
      "[ 0.20689655]\n",
      "[ 0.88505747]\n",
      "[ 0.24137931]\n",
      "[ 0.66666667]\n",
      "[ 0.14942529]\n",
      "[ 0.77011494]\n",
      "[ 0.29885057]\n",
      "[ 0.94252874]\n",
      "[ 0.06896552]\n",
      "[ 0.67816092]\n",
      "[ 0.3908046]\n",
      "[ 0.8045977]\n",
      "[ 0.06896552]\n",
      "[ 0.65517241]\n",
      "[ 0.]\n",
      "[ 0.71264368]\n",
      "[ 0.49425287]\n",
      "[ 0.77011494]\n",
      "[ 0.35632184]\n",
      "[ 0.77011494]\n",
      "[ 0.3908046]\n",
      "[ 0.64367816]\n",
      "[ 0.24137931]\n",
      "[ 0.64367816]\n",
      "[ 0.37931034]\n",
      "[ 0.64367816]\n",
      "[ 0.18390805]\n",
      "[ 0.88505747]\n",
      "[ 0.50574713]\n",
      "[ 0.90804598]\n",
      "[ 0.27586207]\n",
      "[ 0.91954023]\n",
      "[ 0.27586207]\n",
      "[ 0.81609195]\n",
      "[ 0.25287356]\n",
      "[ 0.83908046]\n",
      "[ 0.26436782]\n",
      "[ 0.86206897]\n",
      "[ 0.45977011]\n",
      "[ 0.6091954]\n",
      "[ 0.51724138]\n",
      "[ 0.68965517]\n",
      "[ 0.44827586]\n",
      "[ 0.95402299]\n",
      "[ 0.13793103]\n",
      "[ 0.64367816]\n",
      "[ 0.56321839]\n",
      "[ 0.70114943]\n",
      "[ 0.56321839]\n",
      "[ 0.6091954]\n",
      "[ 0.29885057]\n",
      "[ 0.97701149]\n",
      "[ 0.12643678]\n",
      "[ 0.8045977]\n",
      "[ 0.04597701]\n",
      "[ 0.95402299]\n",
      "[ 0.06896552]\n",
      "[ 0.62068966]\n",
      "[ 0.44827586]\n",
      "[ 0.97701149]\n",
      "[ 0.57471264]\n",
      "[ 0.65517241]\n",
      "[ 0.5862069]\n",
      "[ 0.70114943]\n",
      "[ 0.51724138]\n",
      "[ 0.93103448]\n",
      "[ 0.11494253]\n",
      "[ 0.97701149]\n",
      "[ 0.1954023]\n",
      "[ 0.93103448]\n",
      "[ 0.4137931]\n",
      "[ 0.93103448]\n",
      "[ 0.57471264]\n",
      "[ 0.59770115]\n",
      "[ 0.52873563]\n",
      "[ 0.85057471]\n",
      "[ 0.]\n",
      "[ 0.71264368]\n",
      "[ 0.27586207]\n",
      "[ 0.79310345]\n",
      "[ 0.16091954]\n",
      "[ 0.59770115]\n",
      "[ 0.27586207]\n",
      "[ 0.87356322]\n",
      "[ 0.45977011]\n",
      "[ 0.95402299]\n",
      "[ 0.40229885]\n",
      "[ 0.68965517]\n",
      "[ 0.05747126]\n",
      "[ 0.64367816]\n",
      "[ 0.44827586]\n",
      "[ 0.65517241]\n",
      "[ 0.37931034]\n",
      "[ 0.95402299]\n",
      "[ 0.56321839]\n",
      "[ 0.63218391]\n",
      "[ 0.01149425]\n",
      "[ 0.75862069]\n",
      "[ 0.57471264]\n",
      "[ 0.88505747]\n",
      "[ 0.20689655]\n",
      "[ 0.62068966]\n",
      "[ 0.37931034]\n",
      "[ 0.90804598]\n",
      "[ 0.37931034]\n",
      "[ 0.96551724]\n",
      "[ 0.35632184]\n",
      "[ 0.65517241]\n",
      "[ 0.26436782]\n",
      "[ 1.]\n",
      "[ 0.13793103]\n",
      "[ 0.70114943]\n",
      "[ 0.09195402]\n",
      "[ 0.87356322]\n",
      "[ 0.35632184]\n",
      "[ 0.82758621]\n",
      "[ 0.]\n",
      "[ 0.81609195]\n",
      "[ 0.34482759]\n",
      "[ 0.79310345]\n",
      "[ 0.16091954]\n",
      "[ 0.63218391]\n",
      "[ 0.25287356]\n",
      "[ 0.90804598]\n",
      "[ 0.47126437]\n",
      "[ 0.93103448]\n",
      "[ 0.22988506]\n",
      "[ 0.59770115]\n",
      "[ 0.04597701]\n",
      "[ 0.82758621]\n",
      "[ 0.28735632]\n",
      "[ 0.7816092]\n",
      "[ 0.57471264]\n",
      "[ 0.70114943]\n",
      "[ 0.32183908]\n",
      "[ 0.62068966]\n",
      "[ 0.33333333]\n",
      "[ 0.68965517]\n",
      "[ 0.32183908]\n",
      "[ 0.71264368]\n",
      "[ 0.17241379]\n",
      "[ 0.71264368]\n",
      "[ 0.56321839]\n",
      "[ 0.63218391]\n",
      "[ 0.52873563]\n",
      "[ 0.86206897]\n",
      "[ 0.3908046]\n",
      "[ 0.87356322]\n",
      "[ 0.55172414]\n",
      "[ 0.72413793]\n",
      "[ 0.20689655]\n",
      "[ 0.77011494]\n",
      "[ 0.14942529]\n",
      "[ 0.87356322]\n",
      "[ 0.40229885]\n",
      "[ 0.98850575]\n",
      "[ 0.57471264]\n",
      "[ 0.95402299]\n",
      "[ 0.20689655]\n",
      "[ 0.72413793]\n",
      "[ 0.44827586]\n",
      "[ 0.71264368]\n",
      "[ 0.01149425]\n",
      "[ 0.72413793]\n",
      "[ 0.22988506]\n",
      "[ 0.75862069]\n",
      "[ 0.17241379]\n",
      "[ 0.67816092]\n",
      "[ 0.34482759]\n",
      "[ 0.72413793]\n",
      "[ 0.50574713]\n",
      "[ 0.86206897]\n",
      "[ 0.08045977]\n",
      "[ 0.74712644]\n",
      "[ 0.48275862]\n",
      "[ 0.94252874]\n",
      "[ 0.12643678]\n",
      "[ 0.89655172]\n",
      "[ 0.10344828]\n",
      "[ 1.]\n",
      "[ 0.13793103]\n",
      "[ 0.68965517]\n",
      "[ 0.31034483]\n",
      "[ 0.79310345]\n",
      "[ 0.2183908]\n",
      "[ 0.90804598]\n",
      "[ 0.5862069]\n",
      "[ 0.71264368]\n",
      "[ 0.18390805]\n",
      "[ 0.73563218]\n",
      "[ 0.02298851]\n",
      "[ 0.6091954]\n",
      "[ 0.06896552]\n",
      "[ 0.98850575]\n",
      "[ 0.52873563]\n",
      "[ 0.81609195]\n",
      "[ 0.24137931]\n",
      "[ 0.94252874]\n",
      "[ 0.01149425]\n",
      "[ 0.93103448]\n",
      "[ 0.49425287]\n",
      "[ 0.67816092]\n",
      "[ 0.5862069]\n",
      "[ 0.90804598]\n",
      "[ 0.06896552]\n",
      "[ 0.67816092]\n",
      "[ 0.56321839]\n",
      "[ 0.96551724]\n",
      "[ 0.50574713]\n",
      "[ 0.77011494]\n",
      "[ 0.28735632]\n",
      "[ 0.86206897]\n",
      "[ 0.03448276]\n",
      "[ 0.63218391]\n",
      "[ 0.06896552]\n",
      "[ 0.70114943]\n",
      "[ 0.29885057]\n",
      "[ 0.83908046]\n",
      "[ 0.18390805]\n",
      "[ 0.74712644]\n",
      "[ 0.09195402]\n",
      "[ 0.62068966]\n",
      "[ 0.35632184]\n",
      "[ 0.8045977]\n",
      "[ 0.54022989]\n",
      "[ 0.81609195]\n",
      "[ 0.03448276]\n",
      "[ 0.81609195]\n",
      "[ 0.1954023]\n",
      "[ 0.98850575]\n",
      "[ 0.25287356]\n",
      "[ 0.59770115]\n",
      "[ 0.04597701]\n",
      "[ 0.98850575]\n",
      "[ 0.17241379]\n",
      "[ 0.81609195]\n",
      "[ 0.52873563]\n",
      "[ 0.65517241]\n",
      "[ 0.56321839]\n",
      "[ 0.85057471]\n",
      "[ 0.03448276]\n",
      "[ 0.59770115]\n",
      "[ 0.43678161]\n",
      "[ 0.88505747]\n",
      "[ 0.16091954]\n",
      "[ 0.90804598]\n",
      "[ 0.]\n",
      "[ 0.87356322]\n",
      "[ 0.20689655]\n",
      "[ 0.70114943]\n",
      "[ 0.5862069]\n",
      "[ 0.85057471]\n",
      "[ 0.16091954]\n",
      "[ 0.70114943]\n",
      "[ 0.54022989]\n",
      "[ 0.8045977]\n",
      "[ 0.49425287]\n",
      "[ 0.98850575]\n",
      "[ 0.4137931]\n",
      "[ 0.90804598]\n",
      "[ 0.24137931]\n",
      "[ 0.71264368]\n",
      "[ 0.32183908]\n",
      "[ 0.6091954]\n",
      "[ 0.27586207]\n",
      "[ 0.72413793]\n",
      "[ 0.02298851]\n",
      "[ 0.72413793]\n",
      "[ 0.49425287]\n",
      "[ 1.]\n",
      "[ 0.45977011]\n",
      "[ 0.71264368]\n",
      "[ 0.2183908]\n",
      "[ 0.79310345]\n",
      "[ 0.47126437]\n",
      "[ 0.68965517]\n",
      "[ 0.22988506]\n",
      "[ 0.8045977]\n",
      "[ 0.01149425]\n",
      "[ 0.94252874]\n",
      "[ 0.26436782]\n",
      "[ 0.59770115]\n",
      "[ 0.37931034]\n",
      "[ 0.70114943]\n",
      "[ 0.]\n",
      "[ 0.82758621]\n",
      "[ 0.13793103]\n",
      "[ 0.7816092]\n",
      "[ 0.18390805]\n",
      "[ 0.70114943]\n",
      "[ 0.45977011]\n",
      "[ 0.63218391]\n",
      "[ 0.49425287]\n",
      "[ 0.89655172]\n",
      "[ 0.3908046]\n",
      "[ 0.68965517]\n",
      "[ 0.11494253]\n",
      "[ 0.91954023]\n",
      "[ 0.52873563]\n",
      "[ 0.85057471]\n",
      "[ 0.47126437]\n",
      "[ 0.88505747]\n",
      "[ 0.33333333]\n",
      "[ 0.7816092]\n",
      "[ 0.09195402]\n",
      "[ 0.90804598]\n",
      "[ 0.26436782]\n",
      "[ 0.8045977]\n",
      "[ 0.14942529]\n",
      "[ 0.86206897]\n",
      "[ 0.56321839]\n",
      "[ 0.89655172]\n",
      "[ 0.51724138]\n",
      "[ 0.70114943]\n",
      "[ 0.54022989]\n",
      "[ 0.98850575]\n",
      "[ 0.24137931]\n",
      "[ 0.97701149]\n",
      "[ 0.32183908]\n",
      "[ 0.71264368]\n",
      "[ 0.34482759]\n",
      "[ 0.74712644]\n",
      "[ 0.13793103]\n",
      "[ 0.67816092]\n",
      "[ 0.29885057]\n",
      "[ 0.86206897]\n",
      "[ 0.04597701]\n",
      "[ 0.91954023]\n",
      "[ 0.02298851]\n",
      "[ 0.74712644]\n",
      "[ 0.33333333]\n",
      "[ 0.7816092]\n",
      "[ 0.18390805]\n",
      "[ 0.7816092]\n",
      "[ 0.48275862]\n",
      "[ 0.87356322]\n",
      "[ 0.26436782]\n",
      "[ 0.59770115]\n",
      "[ 0.3908046]\n",
      "[ 0.70114943]\n",
      "[ 0.25287356]\n",
      "[ 0.96551724]\n",
      "[ 0.05747126]\n",
      "[ 0.71264368]\n",
      "[ 0.47126437]\n",
      "[ 0.85057471]\n",
      "[ 0.32183908]\n",
      "[ 0.66666667]\n",
      "[ 0.52873563]\n",
      "[ 0.62068966]\n",
      "[ 0.1954023]\n",
      "[ 0.77011494]\n",
      "[ 0.40229885]\n",
      "[ 0.62068966]\n",
      "[ 0.36781609]\n",
      "[ 0.82758621]\n",
      "[ 0.5862069]\n",
      "[ 0.74712644]\n",
      "[ 0.02298851]\n",
      "[ 0.88505747]\n",
      "[ 0.36781609]\n",
      "[ 0.94252874]\n",
      "[ 0.57471264]\n",
      "[ 0.93103448]\n",
      "[ 0.34482759]\n",
      "[ 0.68965517]\n",
      "[ 0.29885057]\n",
      "[ 0.89655172]\n",
      "[ 0.33333333]\n",
      "[ 0.68965517]\n",
      "[ 0.]\n",
      "[ 0.70114943]\n",
      "[ 0.3908046]\n",
      "[ 0.96551724]\n",
      "[ 0.01149425]\n",
      "[ 0.71264368]\n",
      "[ 0.10344828]\n",
      "[ 0.59770115]\n",
      "[ 0.10344828]\n",
      "[ 0.90804598]\n",
      "[ 0.]\n",
      "[ 0.94252874]\n",
      "[ 0.11494253]\n",
      "[ 0.88505747]\n",
      "[ 0.31034483]\n",
      "[ 0.85057471]\n",
      "[ 0.09195402]\n",
      "[ 0.88505747]\n",
      "[ 0.52873563]\n",
      "[ 0.71264368]\n",
      "[ 0.55172414]\n",
      "[ 0.71264368]\n",
      "[ 0.2183908]\n",
      "[ 0.91954023]\n",
      "[ 0.20689655]\n",
      "[ 0.94252874]\n",
      "[ 0.50574713]\n",
      "[ 0.81609195]\n",
      "[ 0.29885057]\n",
      "[ 0.6091954]\n",
      "[ 0.13793103]\n",
      "[ 0.88505747]\n",
      "[ 0.2183908]\n",
      "[ 0.98850575]\n",
      "[ 0.42528736]\n",
      "[ 0.68965517]\n",
      "[ 0.50574713]\n",
      "[ 0.70114943]\n",
      "[ 0.14942529]\n",
      "[ 0.82758621]\n",
      "[ 0.12643678]\n",
      "[ 0.79310345]\n",
      "[ 0.43678161]\n",
      "[ 0.87356322]\n",
      "[ 0.48275862]\n",
      "[ 0.6091954]\n",
      "[ 0.13793103]\n",
      "[ 0.91954023]\n",
      "[ 0.31034483]\n",
      "[ 0.8045977]\n",
      "[ 0.49425287]\n",
      "[ 0.75862069]\n",
      "[ 0.22988506]\n",
      "[ 0.62068966]\n",
      "[ 0.50574713]\n",
      "[ 0.81609195]\n",
      "[ 0.50574713]\n",
      "[ 0.81609195]\n",
      "[ 0.57471264]\n",
      "[ 0.67816092]\n",
      "[ 0.16091954]\n",
      "[ 0.79310345]\n",
      "[ 0.24137931]\n",
      "[ 0.72413793]\n",
      "[ 0.]\n",
      "[ 0.67816092]\n",
      "[ 0.43678161]\n",
      "[ 0.96551724]\n",
      "[ 0.32183908]\n",
      "[ 0.94252874]\n",
      "[ 0.26436782]\n",
      "[ 0.95402299]\n",
      "[ 0.43678161]\n",
      "[ 0.66666667]\n",
      "[ 0.4137931]\n",
      "[ 0.65517241]\n",
      "[ 0.14942529]\n",
      "[ 0.88505747]\n",
      "[ 0.09195402]\n",
      "[ 0.83908046]\n",
      "[ 0.44827586]\n",
      "[ 0.65517241]\n",
      "[ 0.44827586]\n",
      "[ 0.71264368]\n",
      "[ 0.20689655]\n",
      "[ 0.71264368]\n",
      "[ 0.1954023]\n",
      "[ 0.82758621]\n",
      "[ 0.55172414]\n",
      "[ 0.97701149]\n",
      "[ 0.24137931]\n",
      "[ 0.67816092]\n",
      "[ 0.10344828]\n",
      "[ 0.8045977]\n",
      "[ 0.06896552]\n",
      "[ 0.90804598]\n",
      "[ 0.14942529]\n",
      "[ 0.68965517]\n",
      "[ 0.50574713]\n",
      "[ 0.68965517]\n",
      "[ 0.50574713]\n",
      "[ 0.81609195]\n",
      "[ 0.04597701]\n",
      "[ 0.72413793]\n",
      "[ 0.55172414]\n",
      "[ 0.81609195]\n",
      "[ 0.13793103]\n",
      "[ 0.95402299]\n",
      "[ 0.20689655]\n",
      "[ 0.79310345]\n",
      "[ 0.40229885]\n",
      "[ 0.62068966]\n",
      "[ 0.35632184]\n",
      "[ 0.66666667]\n",
      "[ 0.10344828]\n",
      "[ 0.73563218]\n",
      "[ 0.4137931]\n",
      "[ 0.59770115]\n",
      "[ 0.03448276]\n",
      "[ 0.74712644]\n",
      "[ 0.22988506]\n",
      "[ 0.81609195]\n",
      "[ 0.32183908]\n",
      "[ 0.6091954]\n",
      "[ 0.37931034]\n",
      "[ 0.6091954]\n",
      "[ 0.45977011]\n",
      "[ 0.71264368]\n",
      "[ 0.42528736]\n",
      "[ 0.64367816]\n",
      "[ 0.52873563]\n",
      "[ 0.89655172]\n",
      "[ 0.14942529]\n",
      "[ 0.59770115]\n",
      "[ 0.25287356]\n",
      "[ 0.97701149]\n",
      "[ 0.51724138]\n",
      "[ 0.85057471]\n",
      "[ 0.01149425]\n",
      "[ 0.67816092]\n",
      "[ 0.55172414]\n",
      "[ 0.88505747]\n",
      "[ 0.14942529]\n",
      "[ 0.97701149]\n",
      "[ 0.48275862]\n",
      "[ 0.66666667]\n",
      "[ 0.1954023]\n",
      "[ 0.70114943]\n",
      "[ 0.17241379]\n",
      "[ 0.79310345]\n",
      "[ 0.45977011]\n",
      "[ 0.93103448]\n",
      "[ 0.2183908]\n",
      "[ 0.94252874]\n",
      "[ 0.11494253]\n",
      "[ 1.]\n",
      "[ 0.10344828]\n",
      "[ 0.81609195]\n",
      "[ 0.17241379]\n",
      "[ 0.77011494]\n",
      "[ 0.32183908]\n",
      "[ 0.68965517]\n",
      "[ 0.08045977]\n",
      "[ 0.6091954]\n",
      "[ 0.56321839]\n",
      "[ 0.67816092]\n",
      "[ 0.50574713]\n",
      "[ 0.70114943]\n",
      "[ 0.31034483]\n",
      "[ 0.81609195]\n",
      "[ 0.4137931]\n",
      "[ 0.73563218]\n",
      "[ 0.5862069]\n",
      "[ 0.67816092]\n",
      "[ 0.35632184]\n",
      "[ 0.91954023]\n",
      "[ 0.]\n",
      "[ 0.74712644]\n",
      "[ 0.11494253]\n",
      "[ 0.59770115]\n",
      "[ 0.04597701]\n",
      "[ 0.74712644]\n",
      "[ 0.42528736]\n",
      "[ 0.77011494]\n",
      "[ 0.51724138]\n",
      "[ 0.66666667]\n",
      "[ 0.14942529]\n",
      "[ 0.79310345]\n",
      "[ 0.28735632]\n",
      "[ 0.82758621]\n",
      "[ 0.20689655]\n",
      "[ 0.79310345]\n",
      "[ 0.44827586]\n",
      "[ 0.64367816]\n",
      "[ 0.09195402]\n",
      "[ 0.62068966]\n",
      "[ 0.47126437]\n",
      "[ 0.75862069]\n",
      "[ 0.40229885]\n",
      "[ 0.6091954]\n",
      "[ 0.52873563]\n",
      "[ 0.85057471]\n",
      "[ 0.51724138]\n",
      "[ 0.79310345]\n",
      "[ 0.24137931]\n",
      "[ 0.73563218]\n",
      "[ 0.55172414]\n",
      "[ 0.66666667]\n",
      "[ 0.22988506]\n",
      "[ 0.72413793]\n",
      "[ 0.2183908]\n",
      "[ 0.83908046]\n",
      "[ 0.33333333]\n",
      "[ 0.77011494]\n",
      "[ 0.25287356]\n",
      "[ 0.67816092]\n",
      "[ 0.35632184]\n",
      "[ 0.70114943]\n",
      "[ 0.16091954]\n",
      "[ 0.62068966]\n",
      "[ 0.50574713]\n",
      "[ 0.73563218]\n",
      "[ 0.20689655]\n",
      "[ 0.88505747]\n",
      "[ 0.51724138]\n",
      "[ 0.81609195]\n",
      "[ 0.40229885]\n",
      "[ 0.82758621]\n",
      "[ 0.48275862]\n",
      "[ 0.65517241]\n",
      "[ 0.4137931]\n",
      "[ 0.65517241]\n",
      "[ 0.04597701]\n",
      "[ 0.59770115]\n",
      "[ 0.3908046]\n",
      "[ 0.79310345]\n",
      "[ 0.11494253]\n",
      "[ 0.65517241]\n",
      "[ 0.49425287]\n",
      "[ 0.75862069]\n",
      "[ 0.42528736]\n",
      "[ 0.72413793]\n",
      "[ 0.05747126]\n",
      "[ 0.63218391]\n",
      "[ 0.17241379]\n",
      "[ 0.79310345]\n",
      "[ 0.29885057]\n",
      "[ 0.96551724]\n",
      "[ 0.05747126]\n",
      "[ 0.65517241]\n",
      "[ 0.25287356]\n",
      "[ 0.91954023]\n",
      "[ 0.27586207]\n",
      "[ 0.91954023]\n",
      "[ 0.02298851]\n",
      "[ 0.87356322]\n",
      "[ 0.55172414]\n",
      "[ 1.]\n",
      "[ 0.11494253]\n",
      "[ 0.7816092]\n",
      "[ 0.09195402]\n",
      "[ 0.77011494]\n",
      "[ 0.47126437]\n",
      "[ 0.71264368]\n",
      "[ 0.4137931]\n",
      "[ 0.83908046]\n",
      "[ 0.57471264]\n",
      "[ 0.7816092]\n",
      "[ 0.44827586]\n",
      "[ 0.64367816]\n",
      "[ 0.12643678]\n",
      "[ 0.87356322]\n",
      "[ 0.42528736]\n",
      "[ 0.85057471]\n",
      "[ 0.16091954]\n",
      "[ 0.88505747]\n",
      "[ 0.04597701]\n",
      "[ 0.95402299]\n",
      "[ 0.5862069]\n",
      "[ 0.81609195]\n",
      "[ 0.5862069]\n",
      "[ 0.59770115]\n",
      "[ 0.31034483]\n",
      "[ 0.75862069]\n",
      "[ 0.17241379]\n",
      "[ 0.86206897]\n",
      "[ 0.18390805]\n",
      "[ 0.62068966]\n",
      "[ 0.01149425]\n",
      "[ 0.75862069]\n",
      "[ 0.05747126]\n",
      "[ 0.63218391]\n",
      "[ 0.34482759]\n",
      "[ 0.91954023]\n",
      "[ 0.10344828]\n",
      "[ 0.7816092]\n",
      "[ 0.24137931]\n",
      "[ 0.77011494]\n",
      "[ 0.08045977]\n",
      "[ 0.8045977]\n",
      "[ 0.17241379]\n",
      "[ 0.93103448]\n",
      "[ 0.02298851]\n",
      "[ 0.81609195]\n",
      "[ 0.14942529]\n",
      "[ 0.93103448]\n",
      "[ 0.48275862]\n",
      "[ 0.89655172]\n",
      "[ 0.5862069]\n",
      "[ 0.68965517]\n",
      "[ 0.48275862]\n",
      "[ 0.6091954]\n",
      "[ 0.26436782]\n",
      "[ 0.62068966]\n",
      "[ 0.2183908]\n",
      "[ 0.73563218]\n",
      "[ 0.16091954]\n",
      "[ 0.85057471]\n",
      "[ 0.29885057]\n",
      "[ 0.64367816]\n",
      "[ 0.06896552]\n",
      "[ 0.8045977]\n",
      "[ 0.42528736]\n",
      "[ 0.86206897]\n"
     ]
    }
   ],
   "source": [
    "#print scaled data\n",
    "for i in scaled_train_samples:\n",
    "    print(i)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Simple Sequential Model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using Theano backend.\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "from keras import backend as K\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Activation\n",
    "from keras.layers.core import Dense\n",
    "from keras.optimizers import Adam\n",
    "from keras.metrics import categorical_crossentropy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model = Sequential([\n",
    "    Dense(16, input_shape=(1,), activation='relu'),\n",
    "    Dense(32, activation='relu'),\n",
    "    Dense(2, activation='softmax')\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_10 (Dense)             (None, 16)                32        \n",
      "_________________________________________________________________\n",
      "dense_11 (Dense)             (None, 32)                544       \n",
      "_________________________________________________________________\n",
      "dense_12 (Dense)             (None, 2)                 66        \n",
      "=================================================================\n",
      "Total params: 642\n",
      "Trainable params: 642\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model.compile(Adam(lr=.0001), loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 1890 samples, validate on 210 samples\n",
      "Epoch 1/20\n",
      "0s - loss: 0.6405 - acc: 0.5444 - val_loss: 0.6190 - val_acc: 0.6286\n",
      "Epoch 2/20\n",
      "0s - loss: 0.6153 - acc: 0.6561 - val_loss: 0.5697 - val_acc: 0.7667\n",
      "Epoch 3/20\n",
      "0s - loss: 0.5619 - acc: 0.7746 - val_loss: 0.5011 - val_acc: 0.8810\n",
      "Epoch 4/20\n",
      "0s - loss: 0.5040 - acc: 0.8434 - val_loss: 0.4403 - val_acc: 0.9333\n",
      "Epoch 5/20\n",
      "0s - loss: 0.4597 - acc: 0.8704 - val_loss: 0.3905 - val_acc: 0.9476\n",
      "Epoch 6/20\n",
      "0s - loss: 0.4224 - acc: 0.8868 - val_loss: 0.3467 - val_acc: 0.9476\n",
      "Epoch 7/20\n",
      "0s - loss: 0.3908 - acc: 0.8825 - val_loss: 0.3068 - val_acc: 0.9571\n",
      "Epoch 8/20\n",
      "0s - loss: 0.3609 - acc: 0.8963 - val_loss: 0.2637 - val_acc: 0.9762\n",
      "Epoch 9/20\n",
      "0s - loss: 0.3343 - acc: 0.9085 - val_loss: 0.2383 - val_acc: 0.9952\n",
      "Epoch 10/20\n",
      "0s - loss: 0.3190 - acc: 0.9180 - val_loss: 0.2153 - val_acc: 0.9952\n",
      "Epoch 11/20\n",
      "0s - loss: 0.3067 - acc: 0.9169 - val_loss: 0.1974 - val_acc: 0.9952\n",
      "Epoch 12/20\n",
      "0s - loss: 0.2984 - acc: 0.9249 - val_loss: 0.1853 - val_acc: 0.9952\n",
      "Epoch 13/20\n",
      "0s - loss: 0.2932 - acc: 0.9190 - val_loss: 0.1764 - val_acc: 0.9952\n",
      "Epoch 14/20\n",
      "0s - loss: 0.2891 - acc: 0.9265 - val_loss: 0.1689 - val_acc: 1.0000\n",
      "Epoch 15/20\n",
      "0s - loss: 0.2857 - acc: 0.9228 - val_loss: 0.1617 - val_acc: 0.9952\n",
      "Epoch 16/20\n",
      "0s - loss: 0.2830 - acc: 0.9249 - val_loss: 0.1560 - val_acc: 0.9952\n",
      "Epoch 17/20\n",
      "0s - loss: 0.2808 - acc: 0.9196 - val_loss: 0.1527 - val_acc: 1.0000\n",
      "Epoch 18/20\n",
      "0s - loss: 0.2790 - acc: 0.9286 - val_loss: 0.1476 - val_acc: 1.0000\n",
      "Epoch 19/20\n",
      "0s - loss: 0.2773 - acc: 0.9265 - val_loss: 0.1442 - val_acc: 1.0000\n",
      "Epoch 20/20\n",
      "0s - loss: 0.2760 - acc: 0.9286 - val_loss: 0.1412 - val_acc: 1.0000\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7fdf3881cf98>"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(scaled_train_samples, train_labels, validation_split=0.1, batch_size=10, epochs=20, shuffle=True, verbose=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Save and Load Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. model.save()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.save('medical_trial_model.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This save functions saves:\n",
    "- The architecture of the model, allowing to re-create the model.\n",
    "- The weights of the model.\n",
    "- The training configuration (loss, optimizer).\n",
    "- The state of the optimizer, allowing to resume training exactly where you left off.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.models import load_model\n",
    "new_model = load_model('medical_trial_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_7 (Dense)              (None, 16)                32        \n",
      "_________________________________________________________________\n",
      "dense_8 (Dense)              (None, 32)                544       \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 2)                 66        \n",
      "=================================================================\n",
      "Total params: 642\n",
      "Trainable params: 642\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "new_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([[-0.55824059, -0.23213108, -0.23506787, -0.18476903, -0.02869798,\n",
       "          0.22048838, -0.24348828, -0.20775913, -0.04001847,  0.52669042,\n",
       "         -0.57981801,  0.35746267, -0.17660122, -0.22007732,  0.69854277,\n",
       "          0.52346867]], dtype=float32),\n",
       " array([ -5.05928067e-04,   1.90466806e-01,  -5.11594524e-04,\n",
       "         -5.11574442e-04,   1.53560624e-01,   2.13311628e-01,\n",
       "          1.95102409e-01,  -1.62209501e-04,  -5.11592429e-04,\n",
       "         -6.52560219e-02,  -5.11568855e-04,  -2.15423163e-02,\n",
       "          1.51852086e-01,  -5.11587015e-04,  -8.08358714e-02,\n",
       "          1.78475305e-01], dtype=float32),\n",
       " array([[-0.00450879, -0.08100165,  0.11644009,  0.13092437, -0.28577736,\n",
       "          0.08238146, -0.21489237,  0.19505553, -0.10658678, -0.27693301,\n",
       "         -0.24013698, -0.30782962,  0.26164085, -0.32277834,  0.32263884,\n",
       "          0.06146641, -0.06125453, -0.32100996,  0.09845912,  0.08729291,\n",
       "         -0.34265009, -0.18574393, -0.26289314, -0.08126694, -0.11828163,\n",
       "         -0.10521819,  0.15823284, -0.15937917,  0.20078856, -0.21436444,\n",
       "          0.18525884, -0.00997262],\n",
       "        [ 0.17016929, -0.33530334,  0.5720734 , -0.11036614,  0.26005536,\n",
       "          0.63021338, -0.21984364,  0.15586203,  0.24155881,  0.45153794,\n",
       "          0.66650838,  0.16898219, -0.19299753, -0.28423494,  0.06565986,\n",
       "         -0.17698295,  0.22551182,  0.4560805 , -0.38985988,  0.05320776,\n",
       "         -0.03031705,  0.03303786, -0.10267868,  0.09674571,  0.20863932,\n",
       "          0.40641576, -0.19950488, -0.12717701, -0.45470032,  0.5555678 ,\n",
       "          0.43399239, -0.28200743],\n",
       "        [-0.31182763, -0.14840369, -0.32003853, -0.1944935 , -0.30310264,\n",
       "          0.00804722, -0.131934  , -0.16671041, -0.30733803,  0.29831067,\n",
       "          0.21912673, -0.32854858,  0.2708981 ,  0.08126371, -0.11880381,\n",
       "         -0.09630032, -0.11044759, -0.2071186 ,  0.14548047,  0.3437784 ,\n",
       "          0.11400195,  0.08041922, -0.08599925, -0.32868168,  0.04396215,\n",
       "          0.27437201,  0.27707389,  0.25461951, -0.02398746, -0.07973924,\n",
       "          0.10379016, -0.1016148 ],\n",
       "        [ 0.06183511,  0.13754135, -0.14945416, -0.32213345, -0.16256084,\n",
       "          0.08241963, -0.19906548,  0.05490912, -0.2213712 , -0.0496162 ,\n",
       "         -0.13229065, -0.29842815, -0.300704  , -0.27467173,  0.2398017 ,\n",
       "          0.35232079, -0.08191369,  0.19916396,  0.24313134,  0.02212684,\n",
       "         -0.24171938,  0.0121428 ,  0.30573049,  0.04220042,  0.25314733,\n",
       "         -0.34841883, -0.0346002 ,  0.16545378,  0.01286912,  0.34340069,\n",
       "         -0.00689168, -0.08564182],\n",
       "        [-0.20235111,  0.0225154 ,  0.34031099, -0.05019691,  0.09740504,\n",
       "          0.26879579, -0.23417027, -0.24146383,  0.18816547,  0.09979913,\n",
       "          0.35134175, -0.17725734, -0.01535286,  0.27368823, -0.27873737,\n",
       "          0.09955252, -0.07511313,  0.50303334,  0.19528468,  0.13374521,\n",
       "          0.25183877, -0.10118014, -0.28176242,  0.33339113, -0.28240824,\n",
       "          0.29094845,  0.21905972,  0.05767601, -0.13956711,  0.40815154,\n",
       "          0.3121247 , -0.05011176],\n",
       "        [-0.14970809,  0.24495275,  0.1047416 , -0.09729525,  0.14872137,\n",
       "         -0.26453623, -0.23264726, -0.25778905, -0.03002889, -0.18532023,\n",
       "          0.23062307, -0.35832223,  0.01431278, -0.15617844, -0.13508339,\n",
       "          0.05614368,  0.24606517,  0.17838511,  0.29530466, -0.00831848,\n",
       "          0.08646956,  0.18879111, -0.14868978,  0.1006019 , -0.08833759,\n",
       "          0.37370208, -0.34140989,  0.14407314,  0.39126319, -0.03222291,\n",
       "          0.0908955 , -0.29631507],\n",
       "        [ 0.01909383,  0.04873857,  0.33499172, -0.07754555,  0.5759207 ,\n",
       "          0.37300068,  0.33012795,  0.2734147 ,  0.58095533,  0.68693936,\n",
       "          0.58154178,  0.14178161,  0.32710826,  0.16182107, -0.14604762,\n",
       "         -0.06305101,  0.55986536,  0.20719233, -0.64900148,  0.25428051,\n",
       "         -0.10862546, -0.26342231,  0.16468994,  0.33343408, -0.09155385,\n",
       "          0.43022513,  0.10432567,  0.11050339, -0.63819271,  0.49636003,\n",
       "          0.38338685,  0.17858429],\n",
       "        [-0.02729988,  0.34721547, -0.21079868, -0.32899478,  0.23637503,\n",
       "          0.18560682, -0.18393975, -0.18574759, -0.32932061,  0.09924154,\n",
       "         -0.07376287, -0.05487867,  0.30748746,  0.14141093, -0.32977217,\n",
       "         -0.31329203, -0.12802921, -0.30797395, -0.16787358, -0.1242087 ,\n",
       "          0.04563713, -0.28599206,  0.15355796,  0.04518862,  0.23328239,\n",
       "         -0.13923004, -0.17843686, -0.31587076,  0.00355497,  0.01324073,\n",
       "         -0.07826556, -0.09502369],\n",
       "        [ 0.22787595,  0.29349273, -0.22358777,  0.14266489, -0.31032324,\n",
       "         -0.09297553, -0.028287  , -0.26012093,  0.14716752,  0.25762901,\n",
       "         -0.15498133,  0.00159998,  0.18281031, -0.3297452 , -0.22841193,\n",
       "          0.31760746,  0.03636422, -0.24072249,  0.08338118,  0.05926563,\n",
       "         -0.1122623 , -0.32774928, -0.16838694, -0.18784595, -0.09553957,\n",
       "         -0.16488467, -0.29022321,  0.16776364,  0.04764381, -0.02386055,\n",
       "         -0.24039805,  0.26965225],\n",
       "        [-0.05421547, -0.06224913, -0.31464729,  0.20289262, -0.26375791,\n",
       "         -0.37074456,  0.08380745,  0.16367063, -0.34921181, -0.20677088,\n",
       "         -0.0143941 , -0.14955333,  0.18997188,  0.17953594,  0.30116633,\n",
       "         -0.21793333, -0.49435574, -0.0351067 ,  0.33743536, -0.30101001,\n",
       "          0.25456139,  0.24242593, -0.29679638, -0.32800341,  0.29541034,\n",
       "         -0.29486042,  0.06201031, -0.18276514,  0.3591865 ,  0.16050486,\n",
       "         -0.37324753, -0.10707548],\n",
       "        [-0.25576997,  0.34670419,  0.29506204,  0.00259772,  0.34097403,\n",
       "         -0.17666328,  0.09437881,  0.06288685,  0.18411164, -0.02436141,\n",
       "          0.04082643, -0.16819176,  0.27843684,  0.09865823, -0.04884493,\n",
       "          0.05584542, -0.06600945,  0.16963634,  0.28526145,  0.24605912,\n",
       "          0.08081295,  0.05591917,  0.20442094, -0.2335003 , -0.26189905,\n",
       "          0.33412698, -0.30014336,  0.16511357, -0.07780965, -0.14503644,\n",
       "          0.18019269, -0.07795009],\n",
       "        [ 0.09180616,  0.15584737, -0.15688387, -0.2633518 ,  0.15649781,\n",
       "          0.11987493, -0.28943092, -0.08874085, -0.28777379, -0.14263831,\n",
       "         -0.37739497,  0.22047633,  0.22450058, -0.10919056, -0.27471218,\n",
       "         -0.12555815,  0.15841003,  0.09850772,  0.61546797, -0.10306536,\n",
       "         -0.0136737 ,  0.12682648, -0.15664555, -0.35586554,  0.31336641,\n",
       "          0.29903847,  0.03693908, -0.32724696,  0.28902718, -0.18594679,\n",
       "         -0.39918312, -0.34144619],\n",
       "        [-0.35355014,  0.26790366,  0.64297116, -0.01155495,  0.4952946 ,\n",
       "          0.56661755, -0.14902067, -0.25349453,  0.42268938,  0.18132725,\n",
       "          0.46608654,  0.17817868, -0.08893404,  0.1036597 , -0.35016462,\n",
       "          0.00542375,  0.00249961,  0.4134478 , -0.09287677, -0.23513836,\n",
       "         -0.38861093, -0.23494716, -0.10182571,  0.65702468,  0.03479677,\n",
       "         -0.00379919, -0.07917044, -0.20299289, -0.43907312,  0.62640876,\n",
       "          0.65353787,  0.15301101],\n",
       "        [-0.1326174 , -0.14158516, -0.05788901, -0.16881312, -0.01385649,\n",
       "          0.03723606,  0.09809472,  0.19997895, -0.20865795,  0.2998676 ,\n",
       "         -0.18332051, -0.03264695, -0.17541637, -0.29058254,  0.24476171,\n",
       "         -0.04817909,  0.11160628,  0.27544469,  0.27050838,  0.31809393,\n",
       "          0.24111128, -0.25677165, -0.26000267, -0.25111404, -0.02032884,\n",
       "          0.17515239,  0.16100585,  0.3434909 , -0.07243025,  0.28695205,\n",
       "         -0.25773871, -0.03327911],\n",
       "        [-0.20399271,  0.16224648, -0.20741509, -0.10722972, -0.35120893,\n",
       "         -0.40945521, -0.015903  ,  0.20118865, -0.01396851, -0.16471729,\n",
       "         -0.38221011, -0.16657056, -0.05939792, -0.16332485,  0.28526324,\n",
       "         -0.26584581, -0.38108489, -0.35284299,  0.44374681,  0.17264162,\n",
       "          0.19153227, -0.2176013 , -0.05073311, -0.22471267, -0.33219528,\n",
       "         -0.1644147 ,  0.25175121, -0.20133238,  0.31328976, -0.06911985,\n",
       "         -0.22218835, -0.0961314 ],\n",
       "        [-0.23766688, -0.28162447, -0.0418895 ,  0.15820318, -0.04171183,\n",
       "          0.32493916, -0.16480827,  0.07182733,  0.07949116,  0.27177209,\n",
       "          0.08987289, -0.19803207, -0.20117016,  0.04912793, -0.2984277 ,\n",
       "         -0.07806439,  0.24703656, -0.29776284,  0.25377566, -0.35613036,\n",
       "          0.54648626, -0.31083006, -0.10712207,  0.194795  ,  0.07477053,\n",
       "         -0.26500338, -0.26904857, -0.09847876,  0.2137872 , -0.32672256,\n",
       "          0.20285016, -0.27029315]], dtype=float32),\n",
       " array([ 0.        , -0.0353566 ,  0.15845792, -0.02182444,  0.15326779,\n",
       "         0.13648668,  0.        , -0.09910487,  0.12002168,  0.06610067,\n",
       "         0.0821787 , -0.01089142, -0.00353345, -0.01393286, -0.02681207,\n",
       "        -0.0005116 ,  0.08223103,  0.16984959,  0.13732612, -0.00499463,\n",
       "         0.10528738,  0.        ,  0.        ,  0.10061042, -0.01535971,\n",
       "         0.06933066, -0.02761294, -0.0005116 ,  0.16029432,  0.13847825,\n",
       "         0.12299038,  0.        ], dtype=float32),\n",
       " array([[-0.16389063,  0.38667735],\n",
       "        [-0.00852075, -0.38989428],\n",
       "        [ 0.52292585, -0.61731607],\n",
       "        [ 0.30819824, -0.11042283],\n",
       "        [ 0.08631014, -0.64174998],\n",
       "        [ 0.67267013, -0.67943281],\n",
       "        [ 0.31509909,  0.17057505],\n",
       "        [ 0.07194477, -0.13460469],\n",
       "        [ 0.20532571, -0.74169219],\n",
       "        [ 0.49102944, -0.37470841],\n",
       "        [ 0.70963132, -0.44804388],\n",
       "        [ 0.08322205,  0.24952687],\n",
       "        [-0.05576908, -0.2816318 ],\n",
       "        [-0.12291811,  0.18271691],\n",
       "        [ 0.36846623, -0.30589214],\n",
       "        [-0.10245264,  0.4019056 ],\n",
       "        [ 0.58062744, -0.47862363],\n",
       "        [ 0.25699303, -0.46828336],\n",
       "        [-0.04142442,  0.5261302 ],\n",
       "        [ 0.32674465,  0.34512082],\n",
       "        [-0.67921579,  0.17416315],\n",
       "        [ 0.06968593,  0.06994438],\n",
       "        [-0.04084613, -0.38771692],\n",
       "        [ 0.48274121, -0.41425401],\n",
       "        [ 0.23731484, -0.29605931],\n",
       "        [ 0.3326996 , -0.16685621],\n",
       "        [ 0.10306972, -0.18045028],\n",
       "        [-0.25480118,  0.10807615],\n",
       "        [-0.27306184,  0.56821913],\n",
       "        [ 0.26719874, -0.64521414],\n",
       "        [ 0.40959126, -0.33349776],\n",
       "        [-0.25054461,  0.07801859]], dtype=float32),\n",
       " array([-0.08852487,  0.08852487], dtype=float32)]"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_model.get_weights()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<keras.optimizers.Adam at 0x7fdf399795c0>"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_model.optimizer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. model.to_json()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you only need to save the architecture of a model, and not its weights or its training configuration, you can use the following function to save the architecture only."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# save as JSON\n",
    "json_string = model.to_json()\n",
    "\n",
    "# save as YAML\n",
    "# yaml_string = model.to_yaml()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'{\"backend\": \"theano\", \"keras_version\": \"2.0.6\", \"config\": [{\"config\": {\"use_bias\": true, \"dtype\": \"float32\", \"kernel_regularizer\": null, \"bias_regularizer\": null, \"activation\": \"relu\", \"batch_input_shape\": [null, 1], \"kernel_initializer\": {\"config\": {\"mode\": \"fan_avg\", \"seed\": null, \"scale\": 1.0, \"distribution\": \"uniform\"}, \"class_name\": \"VarianceScaling\"}, \"trainable\": true, \"bias_constraint\": null, \"bias_initializer\": {\"config\": {}, \"class_name\": \"Zeros\"}, \"activity_regularizer\": null, \"kernel_constraint\": null, \"units\": 16, \"name\": \"dense_7\"}, \"class_name\": \"Dense\"}, {\"config\": {\"use_bias\": true, \"kernel_regularizer\": null, \"bias_regularizer\": null, \"activation\": \"relu\", \"kernel_initializer\": {\"config\": {\"mode\": \"fan_avg\", \"seed\": null, \"scale\": 1.0, \"distribution\": \"uniform\"}, \"class_name\": \"VarianceScaling\"}, \"trainable\": true, \"bias_constraint\": null, \"bias_initializer\": {\"config\": {}, \"class_name\": \"Zeros\"}, \"activity_regularizer\": null, \"name\": \"dense_8\", \"units\": 32, \"kernel_constraint\": null}, \"class_name\": \"Dense\"}, {\"config\": {\"use_bias\": true, \"kernel_regularizer\": null, \"bias_regularizer\": null, \"activation\": \"softmax\", \"kernel_initializer\": {\"config\": {\"mode\": \"fan_avg\", \"seed\": null, \"scale\": 1.0, \"distribution\": \"uniform\"}, \"class_name\": \"VarianceScaling\"}, \"trainable\": true, \"bias_constraint\": null, \"bias_initializer\": {\"config\": {}, \"class_name\": \"Zeros\"}, \"activity_regularizer\": null, \"name\": \"dense_9\", \"units\": 2, \"kernel_constraint\": null}, \"class_name\": \"Dense\"}], \"class_name\": \"Sequential\"}'"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "json_string"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# model reconstruction from JSON:\n",
    "from keras.models import model_from_json\n",
    "model_architecture = model_from_json(json_string)\n",
    "\n",
    "# model reconstruction from YAML\n",
    "# from keras.models import model_from_yaml\n",
    "# model = model_from_yaml(yaml_string)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_7 (Dense)              (None, 16)                32        \n",
      "_________________________________________________________________\n",
      "dense_8 (Dense)              (None, 32)                544       \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 2)                 66        \n",
      "=================================================================\n",
      "Total params: 642\n",
      "Trainable params: 642\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model_architecture.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. model.save_weights()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you only need to save the weights of a model, you can use the following function save the weights only."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.save_weights('my_model_weights.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model2 = Sequential([\n",
    "    Dense(16, input_shape=(1,), activation='relu'),\n",
    "    Dense(32, activation='relu'),\n",
    "    Dense(2, activation='softmax')\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model2.load_weights('my_model_weights.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Preprocess Test Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "test_labels =  []\n",
    "test_samples = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for i in range(10):\n",
    "    # The 5% of younger individuals who did experience side effects\n",
    "    random_younger = randint(13,64)\n",
    "    test_samples.append(random_younger)\n",
    "    test_labels.append(1)\n",
    "    \n",
    "    # The 5% of older individuals who did not experience side effects\n",
    "    random_older = randint(65,100)\n",
    "    test_samples.append(random_older)\n",
    "    test_labels.append(0)\n",
    "\n",
    "for i in range(200):\n",
    "    # The 95% of younger individuals who did not experience side effects\n",
    "    random_younger = randint(13,64)\n",
    "    test_samples.append(random_younger)\n",
    "    test_labels.append(0)\n",
    "    \n",
    "    # The 95% of older individuals who did experience side effects\n",
    "    random_older = randint(65,100)\n",
    "    test_samples.append(random_older)\n",
    "    test_labels.append(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "test_labels = np.array(test_labels)\n",
    "test_samples = np.array(test_samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/epoch/anaconda2/envs/fastai34/lib/python3.4/site-packages/sklearn/utils/validation.py:429: DataConversionWarning: Data with input dtype int64 was converted to float64 by MinMaxScaler.\n",
      "  warnings.warn(msg, _DataConversionWarning)\n"
     ]
    }
   ],
   "source": [
    "scaler = MinMaxScaler(feature_range=(0,1))\n",
    "scaled_test_samples = scaler.fit_transform((test_samples).reshape(-1,1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "predictions = model.predict(scaled_test_samples, batch_size=10, verbose=0) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.7410683  0.2589317]\n",
      "[ 0.14958295  0.85041702]\n",
      "[ 0.96918124  0.03081879]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.88596725  0.11403273]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.97202832  0.02797166]\n",
      "[ 0.13397652  0.86602348]\n",
      "[ 0.96605438  0.03394562]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.97461933  0.02538065]\n",
      "[ 0.11597294  0.88402706]\n",
      "[ 0.53810829  0.46189171]\n",
      "[ 0.11046991  0.88953006]\n",
      "[ 0.92053664  0.07946334]\n",
      "[ 0.41420627  0.58579373]\n",
      "[ 0.53810829  0.46189171]\n",
      "[ 0.09690166  0.90309834]\n",
      "[ 0.94527358  0.05472645]\n",
      "[ 0.11227837  0.88772166]\n",
      "[ 0.97461933  0.02538065]\n",
      "[ 0.41420627  0.58579373]\n",
      "[ 0.98343712  0.01656286]\n",
      "[ 0.22349688  0.77650309]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.11227837  0.88772166]\n",
      "[ 0.98343712  0.01656286]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.95885855  0.04114145]\n",
      "[ 0.12674139  0.87325859]\n",
      "[ 0.51321256  0.48678744]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.96605438  0.03394562]\n",
      "[ 0.30028591  0.69971412]\n",
      "[ 0.98595065  0.01404935]\n",
      "[ 0.10868704  0.89131296]\n",
      "[ 0.48825112  0.51174885]\n",
      "[ 0.22349688  0.77650309]\n",
      "[ 0.89567453  0.10432549]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.89567453  0.10432549]\n",
      "[ 0.13397652  0.86602348]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.36671475  0.63328528]\n",
      "[ 0.96262252  0.03737745]\n",
      "[ 0.12674139  0.87325859]\n",
      "[ 0.96605438  0.03394562]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.63464487  0.36535516]\n",
      "[ 0.34384555  0.65615445]\n",
      "[ 0.65747505  0.34252495]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.70094359  0.29905638]\n",
      "[ 0.36671475  0.63328528]\n",
      "[ 0.96918124  0.03081879]\n",
      "[ 0.11798808  0.88201195]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.95885855  0.04114145]\n",
      "[ 0.11798808  0.88201195]\n",
      "[ 0.95021594  0.04978407]\n",
      "[ 0.17580917  0.82419086]\n",
      "[ 0.91291791  0.08708212]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.95021594  0.04978407]\n",
      "[ 0.19075122  0.80924881]\n",
      "[ 0.96918124  0.03081879]\n",
      "[ 0.12006459  0.87993544]\n",
      "[ 0.89567453  0.10432549]\n",
      "[ 0.20664477  0.79335523]\n",
      "[ 0.97697604  0.02302396]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.87548238  0.12451764]\n",
      "[ 0.11227837  0.88772166]\n",
      "[ 0.95473331  0.04526669]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.96605438  0.03394562]\n",
      "[ 0.30028591  0.69971412]\n",
      "[ 0.83898944  0.16101058]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.67959881  0.32040116]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.7410683  0.2589317]\n",
      "[ 0.11227837  0.88772166]\n",
      "[ 0.88596725  0.11403273]\n",
      "[ 0.36671475  0.63328528]\n",
      "[ 0.98414063  0.01585935]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.46334821  0.53665179]\n",
      "[ 0.14028686  0.85971314]\n",
      "[ 0.83898944  0.16101058]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.72144932  0.27855071]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.53810829  0.46189171]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.11046991  0.88953006]\n",
      "[ 0.96262252  0.03737745]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.95885855  0.04114145]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.34384555  0.65615445]\n",
      "[ 0.96262252  0.03737745]\n",
      "[ 0.0985126   0.90148741]\n",
      "[ 0.83898944  0.16101058]\n",
      "[ 0.09531429  0.90468574]\n",
      "[ 0.85202777  0.14797221]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.11597294  0.88402706]\n",
      "[ 0.97697604  0.02302396]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.43862674  0.56137329]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.43862674  0.56137329]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.79431409  0.20568593]\n",
      "[ 0.11046991  0.88953006]\n",
      "[ 0.63464487  0.36535516]\n",
      "[ 0.3216781  0.6783219]\n",
      "[ 0.92754179  0.07245822]\n",
      "[ 0.41420627  0.58579373]\n",
      "[ 0.72144932  0.27855071]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.93397367  0.06602635]\n",
      "[ 0.0985126   0.90148741]\n",
      "[ 0.98540163  0.01459837]\n",
      "[ 0.19075122  0.80924881]\n",
      "[ 0.91291791  0.08708212]\n",
      "[ 0.22349688  0.77650309]\n",
      "[ 0.98540163  0.01459837]\n",
      "[ 0.30028591  0.69971412]\n",
      "[ 0.56281531  0.43718472]\n",
      "[ 0.11798808  0.88201195]\n",
      "[ 0.98540163  0.01459837]\n",
      "[ 0.3216781  0.6783219]\n",
      "[ 0.88596725  0.11403273]\n",
      "[ 0.09690166  0.90309834]\n",
      "[ 0.82503814  0.17496185]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.98481476  0.01518524]\n",
      "[ 0.41420627  0.58579373]\n",
      "[ 0.67959881  0.32040116]\n",
      "[ 0.10868704  0.89131296]\n",
      "[ 0.95021594  0.04978407]\n",
      "[ 0.39020029  0.60979974]\n",
      "[ 0.86418122  0.13581876]\n",
      "[ 0.3216781  0.6783219]\n",
      "[ 0.86418122  0.13581876]\n",
      "[ 0.17580917  0.82419086]\n",
      "[ 0.70094359  0.29905638]\n",
      "[ 0.09531429  0.90468574]\n",
      "[ 0.58721459  0.41278541]\n",
      "[ 0.20664477  0.79335523]\n",
      "[ 0.98343712  0.01656286]\n",
      "[ 0.30028591  0.69971412]\n",
      "[ 0.92754179  0.07245822]\n",
      "[ 0.12006459  0.87993544]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.09531429  0.90468574]\n",
      "[ 0.58721459  0.41278541]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.97697604  0.02302396]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.13397652  0.86602348]\n",
      "[ 0.98595065  0.01404935]\n",
      "[ 0.36671475  0.63328528]\n",
      "[ 0.98077482  0.01922515]\n",
      "[ 0.19075122  0.80924881]\n",
      "[ 0.90464431  0.0953557 ]\n",
      "[ 0.11597294  0.88402706]\n",
      "[ 0.82503814  0.17496185]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.77751815  0.22248183]\n",
      "[ 0.09531429  0.90468574]\n",
      "[ 0.91291791  0.08708212]\n",
      "[ 0.26005745  0.73994255]\n",
      "[ 0.65747505  0.34252495]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.10868704  0.89131296]\n",
      "[ 0.63464487  0.36535516]\n",
      "[ 0.39020029  0.60979974]\n",
      "[ 0.53810829  0.46189171]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.43862674  0.56137329]\n",
      "[ 0.26005745  0.73994255]\n",
      "[ 0.98343712  0.01656286]\n",
      "[ 0.10348926  0.89651072]\n",
      "[ 0.96262252  0.03737745]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.98595065  0.01404935]\n",
      "[ 0.12438267  0.87561733]\n",
      "[ 0.95885855  0.04114145]\n",
      "[ 0.3216781  0.6783219]\n",
      "[ 0.98540163  0.01459837]\n",
      "[ 0.30028591  0.69971412]\n",
      "[ 0.92754179  0.07245822]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.98540163  0.01459837]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.65747505  0.34252495]\n",
      "[ 0.14028686  0.85971314]\n",
      "[ 0.97202832  0.02797166]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.98647934  0.01352068]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.83898944  0.16101058]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.92754179  0.07245822]\n",
      "[ 0.12674139  0.87325859]\n",
      "[ 0.67959881  0.32040116]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.85202777  0.14797221]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.20664477  0.79335523]\n",
      "[ 0.77751815  0.22248183]\n",
      "[ 0.13397652  0.86602348]\n",
      "[ 0.91291791  0.08708212]\n",
      "[ 0.16190827  0.83809173]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.97697604  0.02302396]\n",
      "[ 0.14958295  0.85041702]\n",
      "[ 0.85202777  0.14797221]\n",
      "[ 0.10014733  0.89985269]\n",
      "[ 0.98261213  0.01738786]\n",
      "[ 0.17580917  0.82419086]\n",
      "[ 0.56281531  0.43718472]\n",
      "[ 0.13397652  0.86602348]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.98540163  0.01459837]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.87548238  0.12451764]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.9797861  0.0202139]\n",
      "[ 0.27972952  0.72027051]\n",
      "[ 0.98343712  0.01656286]\n",
      "[ 0.11597294  0.88402706]\n",
      "[ 0.72144932  0.27855071]\n",
      "[ 0.41420627  0.58579373]\n",
      "[ 0.98343712  0.01656286]\n",
      "[ 0.10348926  0.89651072]\n",
      "[ 0.77751815  0.22248183]\n",
      "[ 0.41420627  0.58579373]\n",
      "[ 0.96605438  0.03394562]\n",
      "[ 0.26005745  0.73994255]\n",
      "[ 0.75976562  0.24023436]\n",
      "[ 0.11597294  0.88402706]\n",
      "[ 0.97461933  0.02538065]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.98540163  0.01459837]\n",
      "[ 0.09531429  0.90468574]\n",
      "[ 0.98171616  0.01828387]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.95473331  0.04526669]\n",
      "[ 0.20664477  0.79335523]\n",
      "[ 0.95885855  0.04114145]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.79431409  0.20568593]\n",
      "[ 0.10348926  0.89651072]\n",
      "[ 0.63464487  0.36535516]\n",
      "[ 0.09690166  0.90309834]\n",
      "[ 0.86418122  0.13581876]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.11227837  0.88772166]\n",
      "[ 0.98077482  0.01922515]\n",
      "[ 0.16190827  0.83809173]\n",
      "[ 0.98647934  0.01352068]\n",
      "[ 0.12438267  0.87561733]\n",
      "[ 0.87548238  0.12451764]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.16190827  0.83809173]\n",
      "[ 0.96262252  0.03737745]\n",
      "[ 0.12006459  0.87993544]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.98481476  0.01518524]\n",
      "[ 0.12438267  0.87561733]\n",
      "[ 0.96605438  0.03394562]\n",
      "[ 0.10348926  0.89651072]\n",
      "[ 0.98595065  0.01404935]\n",
      "[ 0.3216781  0.6783219]\n",
      "[ 0.85202777  0.14797221]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.94527358  0.05472645]\n",
      "[ 0.10868704  0.89131296]\n",
      "[ 0.77751815  0.22248183]\n",
      "[ 0.16190827  0.83809173]\n",
      "[ 0.58721459  0.41278541]\n",
      "[ 0.19075122  0.80924881]\n",
      "[ 0.79431409  0.20568593]\n",
      "[ 0.10348926  0.89651072]\n",
      "[ 0.98414063  0.01585935]\n",
      "[ 0.19075122  0.80924881]\n",
      "[ 0.98077482  0.01922515]\n",
      "[ 0.34384555  0.65615445]\n",
      "[ 0.70094359  0.29905638]\n",
      "[ 0.39020029  0.60979974]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.46334821  0.53665179]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.9797861  0.0202139]\n",
      "[ 0.10014733  0.89985269]\n",
      "[ 0.86418122  0.13581876]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.95473331  0.04526669]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.53810829  0.46189171]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.86418122  0.13581876]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.70094359  0.29905638]\n",
      "[ 0.17580917  0.82419086]\n",
      "[ 0.70094359  0.29905638]\n",
      "[ 0.36671475  0.63328528]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.20664477  0.79335523]\n",
      "[ 0.93397367  0.06602635]\n",
      "[ 0.13397652  0.86602348]\n",
      "[ 0.58721459  0.41278541]\n",
      "[ 0.36671475  0.63328528]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.11798808  0.88201195]\n",
      "[ 0.98343712  0.01656286]\n",
      "[ 0.12006459  0.87993544]\n",
      "[ 0.81015152  0.18984847]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.96262252  0.03737745]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.63464487  0.36535516]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.98595065  0.01404935]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.98540163  0.01459837]\n",
      "[ 0.19075122  0.80924881]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.34384555  0.65615445]\n",
      "[ 0.7410683  0.2589317]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.98171616  0.01828387]\n",
      "[ 0.14958295  0.85041702]\n",
      "[ 0.89567453  0.10432549]\n",
      "[ 0.24130528  0.75869471]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.14028686  0.85971314]\n",
      "[ 0.46334821  0.53665179]\n",
      "[ 0.16190827  0.83809173]\n",
      "[ 0.98261213  0.01738786]\n",
      "[ 0.12006459  0.87993544]\n",
      "[ 0.93397367  0.06602635]\n",
      "[ 0.12985019  0.87014979]\n",
      "[ 0.94527358  0.05472645]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.83898944  0.16101058]\n",
      "[ 0.20664477  0.79335523]\n",
      "[ 0.61119276  0.38880724]\n",
      "[ 0.10519696  0.89480305]\n",
      "[ 0.98481476  0.01518524]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.96605438  0.03394562]\n",
      "[ 0.34384555  0.65615445]\n",
      "[ 0.95473331  0.04526669]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.56281531  0.43718472]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.98595065  0.01404935]\n",
      "[ 0.09690166  0.90309834]\n",
      "[ 0.93397367  0.06602635]\n",
      "[ 0.11597294  0.88402706]\n",
      "[ 0.46334821  0.53665179]\n",
      "[ 0.34384555  0.65615445]\n",
      "[ 0.56281531  0.43718472]\n",
      "[ 0.39020029  0.60979974]\n",
      "[ 0.89567453  0.10432549]\n",
      "[ 0.12006459  0.87993544]\n",
      "[ 0.9797861  0.0202139]\n",
      "[ 0.12006459  0.87993544]\n",
      "[ 0.43862674  0.56137329]\n",
      "[ 0.11046991  0.88953006]\n",
      "[ 0.95885855  0.04114145]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.72144932  0.27855071]\n",
      "[ 0.41420627  0.58579373]\n",
      "[ 0.97874761  0.02125239]\n",
      "[ 0.13397652  0.86602348]\n",
      "[ 0.97697604  0.02302396]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.70094359  0.29905638]\n",
      "[ 0.12217253  0.87782747]\n",
      "[ 0.79431409  0.20568593]\n",
      "[ 0.0985126   0.90148741]\n",
      "[ 0.98171616  0.01828387]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.53810829  0.46189171]\n",
      "[ 0.30028591  0.69971412]\n",
      "[ 0.98481476  0.01518524]\n",
      "[ 0.11411262  0.88588738]\n",
      "[ 0.93397367  0.06602635]\n",
      "[ 0.10692947  0.89307052]\n",
      "[ 0.93987161  0.0601284 ]\n",
      "[ 0.10180613  0.8981939 ]\n",
      "[ 0.81015152  0.18984847]\n",
      "[ 0.12217253  0.87782747]\n"
     ]
    }
   ],
   "source": [
    "for i in predictions:\n",
    "    print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "rounded_predictions = model.predict_classes(scaled_test_samples, batch_size=10, verbose=0) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n",
      "0\n",
      "1\n"
     ]
    }
   ],
   "source": [
    "for i in rounded_predictions:\n",
    "    print(i)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import itertools\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "cm = confusion_matrix(test_labels, rounded_predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, cm[i, j],\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[191  19]\n",
      " [ 10 200]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAEmCAYAAADBbUO1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdf391ff6d8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm_plot_labels = ['no_side_effects','had_side_effects']\n",
    "plot_confusion_matrix(cm, cm_plot_labels, title='Confusion Matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.4.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}