Reading Comprehension Final Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Simple Question And Answering System \n",
    "#### Greg Murray, Srinivasan Vasudevan \n",
    "dataset: bAbi dataset \n",
    "Date: DEC 5th 2018"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from keras.models import Sequential, Model\n",
    "from keras.layers.embeddings import Embedding\n",
    "from keras.layers import Input, Activation, Dense, Permute, Dropout\n",
    "from keras.layers import add, dot, concatenate\n",
    "from keras.layers import LSTM\n",
    "from keras.utils.data_utils import get_file\n",
    "from keras.preprocessing.sequence import pad_sequences\n",
    "from functools import reduce\n",
    "import tarfile\n",
    "import numpy as np\n",
    "import re\n",
    "import os, fnmatch,glob,pathlib\n",
    "from keras.layers import Bidirectional\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "TEXT_DATA_DIR = '/Users/gregmurray/Documents/Deep_Learning/Question_Answering/tasks_1-20_v1-2/en-valid-10k'\n",
    "path = TEXT_DATA_DIR\n",
    "def tokenize(sent):\n",
    "    '''Return the tokens of a sentence including punctuation.\n",
    "    >>> tokenize('Bob dropped the apple. Where is the apple?')\n",
    "    ['Bob', 'dropped', 'the', 'apple', '.', 'Where', 'is', 'the', 'apple', '?']\n",
    "    '''\n",
    "    return [x.strip() for x in re.split('(\\W+)?', sent) if x.strip()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def parse_stories(lines, only_supporting=False):\n",
    "    '''Parse stories provided in the bAbi tasks format\n",
    "    If only_supporting is true, only the sentences\n",
    "    that support the answer are kept.\n",
    "    '''\n",
    "    data = []\n",
    "    story = []\n",
    "    for line in lines:\n",
    "        nid, line = line.split(' ', 1)\n",
    "        nid = int(nid)\n",
    "        if nid == 1:\n",
    "            story = []\n",
    "        if '\\t' in line:\n",
    "            q, a, supporting = line.split('\\t')\n",
    "            q = tokenize(q)\n",
    "            if only_supporting:\n",
    "                # Only select the related substory\n",
    "                supporting = map(int, supporting.split())\n",
    "                substory = [story[i - 1] for i in supporting]\n",
    "            else:\n",
    "                # Provide all the substories\n",
    "                substory = [x for x in story if x]\n",
    "            data.append((substory, q, a))\n",
    "            story.append('')\n",
    "        else:\n",
    "            sent = tokenize(line)\n",
    "            story.append(sent)\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def get_stories(fi, only_supporting=False, max_length=None):\n",
    "    '''Read in file and convert to single story'''\n",
    "    print(fi)\n",
    "    f = open(fi)\n",
    "    data = parse_stories(f.readlines(), only_supporting=only_supporting)\n",
    "    flatten = lambda data: reduce(lambda x, y: x + y, data)\n",
    "    data = [(flatten(story), q, answer) for story, q, answer in data\n",
    "            if not max_length or len(flatten(story)) < max_length]\n",
    "    f.close\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def vectorize_stories(data):\n",
    "    '''Break out story to context, question, and answer\n",
    "    and pad context and question\n",
    "    '''\n",
    "    inputs, queries, answers = [], [], []\n",
    "    for story, query, answer in data:\n",
    "        inputs.append([word_idx[w] for w in story])\n",
    "        queries.append([word_idx[w] for w in query])\n",
    "        answers.append(word_idx[answer])\n",
    "    return (pad_sequences(inputs, maxlen=story_maxlen),\n",
    "            pad_sequences(queries, maxlen=query_maxlen),\n",
    "            np.array(answers))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/gregmurray/Documents/Deep_Learning/Question_Answering/tasks_1-20_v1-2/en-valid-10k/qa1_train.txt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gregmurray/Library/Enthought/Canopy/edm/envs/User/lib/python3.5/re.py:203: FutureWarning: split() requires a non-empty pattern match.\n",
      "  return _compile(pattern, flags).split(string, maxsplit)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/gregmurray/Documents/Deep_Learning/Question_Answering/tasks_1-20_v1-2/en-valid-10k/qa1_test.txt\n",
      "-\n",
      "Vocab size: 22 unique words\n",
      "Story max length: 68 words\n",
      "Query max length: 4 words\n",
      "Number of training stories: 9000\n",
      "Number of test stories: 1000\n",
      "-\n",
      "Here's what a \"story\" tuple looks like (input, query, answer):\n",
      "(['Mary', 'moved', 'to', 'the', 'bathroom', '.', 'John', 'went', 'to', 'the', 'hallway', '.'], ['Where', 'is', 'Mary', '?'], 'bathroom')\n",
      "-\n",
      "Vectorizing the word sequences...\n",
      "-\n",
      "inputs: integer tensor of shape (samples, max_length)\n",
      "inputs_train shape: (9000, 68)\n",
      "inputs_test shape: (1000, 68)\n",
      "-\n",
      "queries: integer tensor of shape (samples, max_length)\n",
      "queries_train shape: (9000, 4)\n",
      "queries_test shape: (1000, 4)\n",
      "-\n",
      "answers: binary (1 or 0) tensor of shape (samples, vocab_size)\n",
      "answers_train shape: (9000,)\n",
      "answers_test shape: (1000,)\n",
      "-\n",
      "Compiling...\n"
     ]
    }
   ],
   "source": [
    "# we will only use qa1 data\n",
    "trainpattern='qa1_train.txt'\n",
    "testpattern='qa1_test.txt'\n",
    "\n",
    "# Read in files\n",
    "trainfiles=fnmatch.filter(os.listdir(TEXT_DATA_DIR), trainpattern)\n",
    "testfiles=fnmatch.filter(os.listdir(TEXT_DATA_DIR), testpattern)\n",
    "for f in trainfiles:\n",
    "    train_stories = get_stories(os.path.join(TEXT_DATA_DIR, f))\n",
    "for f in testfiles:\n",
    "    test_stories = get_stories(os.path.join(TEXT_DATA_DIR, f))\n",
    "\n",
    "# Build the vocabulary from our data\n",
    "vocab = set()\n",
    "for story, q, answer in train_stories + test_stories:\n",
    "    vocab |= set(story + q + [answer])\n",
    "vocab = sorted(vocab)\n",
    "\n",
    "# Reserve 0 for masking via pad_sequences\n",
    "vocab_size = len(vocab) + 1\n",
    "story_maxlen = max(map(len, (x for x, _, _ in train_stories + test_stories)))\n",
    "query_maxlen = max(map(len, (x for _, x, _ in train_stories + test_stories)))\n",
    "\n",
    "# Print details of our data\n",
    "print('-')\n",
    "print('Vocab size:', vocab_size, 'unique words')\n",
    "print('Story max length:', story_maxlen, 'words')\n",
    "print('Query max length:', query_maxlen, 'words')\n",
    "print('Number of training stories:', len(train_stories))\n",
    "print('Number of test stories:', len(test_stories))\n",
    "print('-')\n",
    "print('Here\\'s what a \"story\" tuple looks like (input, query, answer):')\n",
    "print(train_stories[0])\n",
    "print('-')\n",
    "print('Vectorizing the word sequences...')\n",
    "\n",
    "word_idx = dict((c, i + 1) for i, c in enumerate(vocab))\n",
    "inputs_train, queries_train, answers_train = vectorize_stories(train_stories)\n",
    "inputs_test, queries_test, answers_test = vectorize_stories(test_stories)\n",
    "\n",
    "print('-')\n",
    "print('inputs: integer tensor of shape (samples, max_length)')\n",
    "print('inputs_train shape:', inputs_train.shape)\n",
    "print('inputs_test shape:', inputs_test.shape)\n",
    "print('-')\n",
    "print('queries: integer tensor of shape (samples, max_length)')\n",
    "print('queries_train shape:', queries_train.shape)\n",
    "print('queries_test shape:', queries_test.shape)\n",
    "print('-')\n",
    "print('answers: binary (1 or 0) tensor of shape (samples, vocab_size)')\n",
    "print('answers_train shape:', answers_train.shape)\n",
    "print('answers_test shape:', answers_test.shape)\n",
    "print('-')\n",
    "print('Compiling...')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Build Model with BIDIR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 9000 samples, validate on 1000 samples\n",
      "Epoch 1/140\n",
      "9000/9000 [==============================] - 5s 520us/step - loss: 2.0419 - acc: 0.1640 - val_loss: 1.8773 - val_acc: 0.1890\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\n",
      "Epoch 2/140\n",
      "9000/9000 [==============================] - 3s 352us/step - loss: 1.7799 - acc: 0.2678 - val_loss: 1.5726 - val_acc: 0.4100\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\n",
      "Epoch 3/140\n",
      "9000/9000 [==============================] - 3s 359us/step - loss: 1.5897 - acc: 0.3961 - val_loss: 1.5184 - val_acc: 0.4380\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\n",
      "Epoch 4/140\n",
      "9000/9000 [==============================] - 4s 396us/step - loss: 1.5245 - acc: 0.4376 - val_loss: 1.4531 - val_acc: 0.4720\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\n",
      "Epoch 5/140\n",
      "8512/9000 [===========================>..] - ETA: 0s - loss: 1.4775 - acc: 0.4656\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b"
     ]
    }
   ],
   "source": [
    "from keras import regularizers\n",
    "from keras.layers import BatchNormalization\n",
    "# placeholders\n",
    "input_sequence = Input((story_maxlen,))\n",
    "question = Input((query_maxlen,))\n",
    "\n",
    "# encoders\n",
    "# embed the input sequence into a sequence of vectors\n",
    "input_encoder_m = Sequential()\n",
    "input_encoder_m.add(Embedding(input_dim=vocab_size,\n",
    "                              output_dim=64))\n",
    "input_encoder_m.add(Dropout(0.3))\n",
    "# output: (samples, story_maxlen, embedding_dim)\n",
    "\n",
    "# embed the input into a sequence of vectors of size query_maxlen\n",
    "input_encoder_c = Sequential()\n",
    "input_encoder_c.add(Embedding(input_dim=vocab_size,\n",
    "                              output_dim=query_maxlen))\n",
    "input_encoder_c.add(Dropout(0.3))\n",
    "# output: (samples, story_maxlen, query_maxlen)\n",
    "\n",
    "# embed the question into a sequence of vectors\n",
    "question_encoder = Sequential()\n",
    "question_encoder.add(Embedding(input_dim=vocab_size,\n",
    "                               output_dim=64,\n",
    "                               input_length=query_maxlen))\n",
    "question_encoder.add(Dropout(0.3))\n",
    "# output: (samples, query_maxlen, embedding_dim)\n",
    "\n",
    "# encode input sequence and questions (which are indices)\n",
    "# to sequences of dense vectors\n",
    "input_encoded_m = input_encoder_m(input_sequence)\n",
    "input_encoded_c = input_encoder_c(input_sequence)\n",
    "question_encoded = question_encoder(question)\n",
    "\n",
    "# compute a 'match' between the first input vector sequence\n",
    "# and the question vector sequence\n",
    "# shape: `(samples, story_maxlen, query_maxlen)`\n",
    "match = dot([input_encoded_m, question_encoded], axes=(2, 2))\n",
    "match = Activation('softmax')(match)\n",
    "\n",
    "# add the match matrix with the second input vector sequence\n",
    "response = add([match, input_encoded_c])  # (samples, story_maxlen, query_maxlen)\n",
    "response = Permute((2, 1))(response)  # (samples, query_maxlen, story_maxlen)\n",
    "\n",
    "# concatenate the match matrix with the question vector sequence\n",
    "answer = concatenate([response, question_encoded])\n",
    "\n",
    "# BiDirectional layer\n",
    "answer = Bidirectional(LSTM(96, dropout=.2))(answer)\n",
    "\n",
    "#answer = Dense(vocab_size, kernel_regularizer=regularizers.l1(.0001), activation='relu')(answer)  # (samples, vocab_size)\n",
    "\n",
    "answer = Dense(vocab_size, kernel_regularizer=regularizers.l1_l2(.0001))(answer)\n",
    "# output: probability that each word in vocab is the answer\n",
    "\n",
    "answer = Activation('softmax')(answer)\n",
    "\n",
    "# build the final model\n",
    "model = Model([input_sequence, question], answer)\n",
    "model.compile(optimizer='adam', loss='sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "\n",
    "\n",
    "# train\n",
    "history = model.fit([inputs_train, queries_train], answers_train,\n",
    "          batch_size=32,\n",
    "          epochs=140,\n",
    "          validation_data=([inputs_test, queries_test], answers_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Note: we tested various other optimizers, hyper parameters, etc. but settled on this as our final model. To avoid confusion, we have only included our final model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluation\n",
      "1000/1000 [==============================] - 0s 122us/step\n",
      "Test loss / test accuracy = 0.3044 / 0.9240\n"
     ]
    }
   ],
   "source": [
    "# Evaluate performance on testing data\n",
    "print('Evaluation')\n",
    "loss, acc = model.evaluate([inputs_test, queries_test], answers_test,\n",
    "                           batch_size=32)\n",
    "print('Test loss / test accuracy = {:.4f} / {:.4f}'.format(loss, acc))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import ast\n",
    "\n",
    "def populateEncodedWordMap(embeddings_index,train_stories,inputs_train,queries_train):\n",
    "    '''To be used to display sample answers.\n",
    "    Build mapping of encoded numbers to words\n",
    "    '''\n",
    "    for idx in range(len(train_stories)):\n",
    "        line =train_stories[idx]\n",
    "        ak = len(line[0])\n",
    "        it = inputs_train[idx]\n",
    "        at= len(it)-ak\n",
    "        aw = line[0]\n",
    "        for i in range(len(aw)):\n",
    "            embeddings_index[it[at+i]]=aw[i]\n",
    "        qk = len(line[1])\n",
    "        qtt = queries_train[idx]\n",
    "        qt= len(qtt)-qk\n",
    "        qw= line[1]\n",
    "        for i in range(len(qw)):\n",
    "            embeddings_index[qtt[qt+i]]=qw[i]\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Populate mapping\n",
    "embeddings_index = {}\n",
    "populateEncodedWordMap(embeddings_index,train_stories,inputs_train,queries_train)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Example predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Take one input, question, answer sequence from test data\n",
    "testdata=test_stories[0]\n",
    "test_answer=testdata[0] # Context\n",
    "test_q=testdata[1] # Question\n",
    "test_ans=testdata[2] # Answer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Printing Sample Passage from the test observation: \n",
      "['John', 'travelled', 'to', 'the', 'hallway', '.', 'Mary', 'journeyed', 'to', 'the', 'bathroom', '.']\n",
      " \n",
      "Printing Sample Query from the test observation:  ['Where', 'is', 'John', '?']\n",
      " \n",
      "Printing Sample Answer from the test observation:  hallway\n",
      " \n",
      "Now predicting...\n",
      " \n",
      "Printing the predicted Answer is: hallway\n"
     ]
    }
   ],
   "source": [
    "print(\"Printing Sample Passage from the test observation: \")\n",
    "print(test_answer)\n",
    "print(\" \")\n",
    "print(\"Printing Sample Query from the test observation: \",test_q)\n",
    "print(\" \")\n",
    "print(\"Printing Sample Answer from the test observation: \",embeddings_index[answers_test[0]])\n",
    "print(\" \")\n",
    "print(\"Now predicting...\")\n",
    "answers_test1=model.predict([inputs_test,queries_test])\n",
    "print(\" \")\n",
    "print(\"Printing the predicted Answer is:\",embeddings_index[answers_test1[0].argmax(axis=-1)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Take another sample\n",
    "testdata=test_stories[100]\n",
    "test_answer=testdata[0]\n",
    "test_q=testdata[1]\n",
    "test_ans=testdata[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Printing Sample Passage from the test observation: \n",
      "['Mary', 'travelled', 'to', 'the', 'garden', '.', 'Mary', 'moved', 'to', 'the', 'bedroom', '.']\n",
      " \n",
      "Printing Sample Query from the test observation:  ['Where', 'is', 'Mary', '?']\n",
      " \n",
      "Printing Sample Answer from the test observation:  bedroom\n",
      " \n",
      "Now predicting...\n",
      " \n",
      "Printing the predicted Answer is: bedroom\n"
     ]
    }
   ],
   "source": [
    "print(\"Printing Sample Passage from the test observation: \")\n",
    "print(test_answer)\n",
    "print(\" \")\n",
    "print(\"Printing Sample Query from the test observation: \",test_q)\n",
    "print(\" \")\n",
    "print(\"Printing Sample Answer from the test observation: \",embeddings_index[answers_test[100]])\n",
    "print(\" \")\n",
    "print(\"Now predicting...\")\n",
    "answers_test1=model.predict([inputs_test,queries_test])\n",
    "print(\" \")\n",
    "print(\"Printing the predicted Answer is:\",embeddings_index[answers_test1[100].argmax(axis=-1)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Printing Sample Passage from the test observation: \n",
      "['Sandra', 'went', 'to', 'the', 'bedroom', '.', 'Sandra', 'travelled', 'to', 'the', 'office', '.', 'Daniel', 'went', 'back', 'to', 'the', 'garden', '.', 'Mary', 'travelled', 'to', 'the', 'bedroom', '.', 'John', 'journeyed', 'to', 'the', 'garden', '.', 'Daniel', 'went', 'back', 'to', 'the', 'hallway', '.', 'Mary', 'went', 'back', 'to', 'the', 'garden', '.', 'Mary', 'journeyed', 'to', 'the', 'bedroom', '.', 'Sandra', 'went', 'back', 'to', 'the', 'kitchen', '.', 'Sandra', 'went', 'to', 'the', 'garden', '.']\n",
      " \n",
      "Printing Sample Query from the test observation:  ['Where', 'is', 'Daniel', '?']\n",
      " \n",
      "Printing Sample Answer from the test observation:  hallway\n",
      " \n",
      "Now predicting...\n",
      " \n",
      "Printing the predicted Answer is: kitchen\n"
     ]
    }
   ],
   "source": [
    "# Take another sample\n",
    "testdata=test_stories[754]\n",
    "test_answer=testdata[0]\n",
    "test_q=testdata[1]\n",
    "test_ans=testdata[2]\n",
    "print(\"Printing Sample Passage from the test observation: \")\n",
    "print(test_answer)\n",
    "print(\" \")\n",
    "print(\"Printing Sample Query from the test observation: \",test_q)\n",
    "print(\" \")\n",
    "print(\"Printing Sample Answer from the test observation: \",embeddings_index[answers_test[754]])\n",
    "print(\" \")\n",
    "print(\"Now predicting...\")\n",
    "answers_test1=model.predict([inputs_test,queries_test])\n",
    "print(\" \")\n",
    "print(\"Printing the predicted Answer is:\",embeddings_index[answers_test1[754].argmax(axis=-1)])\n",
    "        #print(test_stories[i],embeddings_index[answers_test[i]],embeddings_index[answers_test1[i].argmax(axis=-1)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## We can also generate our own examples:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\kyle\\anaconda3\\lib\\re.py:212: FutureWarning: split() requires a non-empty pattern match.\n",
      "  return _compile(pattern, flags).split(string, maxsplit)\n"
     ]
    }
   ],
   "source": [
    "user_context=\"Mary went to the office. John went to the bathroom. Daniel went back to the kitchen. \\\n",
    "Mary travelled to the garden. Sandra journeyed to the bedroom. Daniel moved to the office. \\\n",
    "John went to the hallway.\"\n",
    "user_q=\"Where is Mary?\"\n",
    "\n",
    "user_story = (tokenize(user_context), tokenize(user_q), '?')\n",
    "sample_context, sample_q, sample_a = vectorize_stories([user_story])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Printing Sample Context: \n",
      "Mary went to the office. John went to the bathroom. Daniel went back to the kitchen. Mary travelled to the garden. Sandra journeyed to the bedroom. Daniel moved to the office. John went to the hallway.\n",
      " \n",
      "Printing Sample Query:  Where is Mary?\n",
      " \n",
      "Now predicting...\n",
      " \n",
      "Printing the predicted 1st observation's answer  garden\n"
     ]
    }
   ],
   "source": [
    "print(\"Printing Sample Context: \")\n",
    "print(user_context)\n",
    "print(\" \")\n",
    "print(\"Printing Sample Query: \",user_q)\n",
    "print(\" \")\n",
    "\n",
    "print(\"Now predicting...\")\n",
    "pred=model.predict([sample_context,sample_q])\n",
    "print(\" \")\n",
    "print(\"Printing the predicted 1st observation's answer \",embeddings_index[pred[0].argmax(axis=-1)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot training vs. validation loss\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['figure.figsize'] = 15,8\n",
    "\n",
    "acc = history.history['acc']\n",
    "val_acc = history.history['val_acc']\n",
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "\n",
    "epochs = range(1, len(acc) + 1)\n",
    "epochs = range(1, len(loss) + 1)\n",
    "\n",
    "plt.plot(epochs, loss, 'bo', label='Training loss')\n",
    "plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
    "plt.title('Training and validation loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot training vs. validation accuracy\n",
    "\n",
    "plt.clf()   # clear figure\n",
    "acc = history.history['acc']\n",
    "val_acc = history.history['val_acc']\n",
    "\n",
    "plt.plot(epochs, acc, 'bo', label='Training acc')\n",
    "plt.plot(epochs, val_acc, 'b', label='Validation acc')\n",
    "plt.title('Training and validation accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Due to randomization, sometimes plots show slight overfitting and sometimes they don't. We tested multiple regularization parameters and found these to be best. In most cases, overfitting was not an issue and validation accuracy was not significantly different from training accuracy."
   ]
  }
 ],
 "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": "3.5.2"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}