09-imdb-mlp-cnn.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# IMDB classification with MLPs and CNNs\n",
    "\n",
    "The dataset is originally from http://ai.stanford.edu/~amaas/data/sentiment/\n",
    "\n",
    "Code based on \n",
    "https://github.com/fchollet/keras/blob/master/examples/pretrained_word_embeddings.py\n",
    "\n",
    "Here's our plan:\n",
    "\n",
    "* load the dataset\n",
    "* Keep only the most frequently occuring 20000 words\n",
    "* Pad or truncate all reviews to fixed length (300)\n",
    "* Look up an embedding for each word\n",
    "* Build our networks: a simple MLP and then a 1D CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "\n",
    "import os\n",
    "import os.path\n",
    "import sys\n",
    "import numpy as np\n",
    "import cPickle as pickle\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.datasets import imdb\n",
    "from keras.preprocessing.text import Tokenizer\n",
    "from keras.preprocessing.sequence import pad_sequences\n",
    "from keras.utils import to_categorical\n",
    "from keras.layers import Dense, Input, Flatten, Dropout, Activation\n",
    "from keras.layers import Conv1D, MaxPooling1D, Embedding, LSTM\n",
    "from keras.models import Sequential, Model\n",
    "from keras.preprocessing.text import Tokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from display import visualize_keras_model, plot_training_curves"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "sns.set_style('white')\n",
    "sns.set_context(\"notebook\", font_scale=1.5, rc={\"lines.linewidth\": 2.5})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading data...\n",
      "25000 train sequences\n",
      "25000 test sequences\n"
     ]
    }
   ],
   "source": [
    "# To work around bug in latest version of the dataset in Keras,\n",
    "# load older version manually, downloaded from \n",
    "# https://s3.amazonaws.com/text-datasets/imdb_full.pkl\n",
    "print('Loading data...')\n",
    "path = os.path.expanduser('~/.keras/datasets/imdb_full.pkl')\n",
    "f = open(path, 'rb')\n",
    "(x_train, y_train), (x_test, y_test) = pickle.load(f)\n",
    "print(len(x_train), 'train sequences')\n",
    "print(len(x_test), 'test sequences')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "word2index = imdb.get_word_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# we want the other direction\n",
    "index2word = dict([(i,w) for (w,i) in word2index.items()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "def totext(review):\n",
    "    return ' '.join(index2word[i] for i in review)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "GLOVE_DIR = os.path.expanduser('~/data/glove')\n",
    "MAX_SEQUENCE_LENGTH = 300\n",
    "MAX_NB_WORDS = 20000\n",
    "EMBEDDING_DIM = 100\n",
    "VALIDATION_SPLIT = 0.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Indexing word vectors.\n",
      "Found 400000 word vectors.\n"
     ]
    }
   ],
   "source": [
    "# first, build index mapping words in the embeddings set\n",
    "# to their embedding vector\n",
    "\n",
    "print('Indexing word vectors.')\n",
    "\n",
    "embeddings_index = {} # word -> coefs\n",
    "# We'll use the 100-dimensional version\n",
    "with open(os.path.join(GLOVE_DIR, 'glove.6B.100d.txt')) as f:\n",
    "    for line in f:\n",
    "        values = line.split()\n",
    "        word = values[0]\n",
    "        coefs = np.asarray(values[1:], dtype='float32')\n",
    "        embeddings_index[word] = coefs\n",
    "\n",
    "print('Found %s word vectors.' % len(embeddings_index))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Shuffle training and test data\n",
    "# (copying from imdb.py because we had to load data manually)\n",
    "seed = 113\n",
    "np.random.seed(seed)\n",
    "np.random.shuffle(x_train)\n",
    "np.random.seed(seed) # same shuffle for labels!\n",
    "np.random.shuffle(y_train)\n",
    "\n",
    "np.random.seed(seed * 2)\n",
    "np.random.shuffle(x_test)\n",
    "np.random.seed(seed * 2)\n",
    "np.random.shuffle(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pad sequences\n",
      "x_train shape: (25000, 300)\n",
      "x_test shape: (25000, 300)\n"
     ]
    }
   ],
   "source": [
    "print('Pad sequences')\n",
    "x_train = pad_sequences(x_train, maxlen=MAX_SEQUENCE_LENGTH)\n",
    "x_test = pad_sequences(x_test, maxlen=MAX_SEQUENCE_LENGTH)\n",
    "print('x_train shape:', x_train.shape)\n",
    "print('x_test shape:', x_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Recall: words are indexed in descending order of frequency. Remove the \n",
    "# less frequent ones -- replace with constant value\n",
    "x_train[x_train >= MAX_NB_WORDS] = MAX_NB_WORDS-1\n",
    "x_test[x_test >= MAX_NB_WORDS] = MAX_NB_WORDS-1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Preparing embedding matrix.\n"
     ]
    }
   ],
   "source": [
    "print('Preparing embedding matrix.')\n",
    "\n",
    "# prepare embedding matrix\n",
    "num_words = min(MAX_NB_WORDS, len(word2index))\n",
    "embedding_matrix = np.zeros((num_words, EMBEDDING_DIM))\n",
    "for word, i in word2index.items():\n",
    "    if i >= MAX_NB_WORDS:\n",
    "        continue\n",
    "    embedding_vector = embeddings_index.get(word)\n",
    "    if embedding_vector is not None:\n",
    "        # words not found in embedding index will be all-zeros.\n",
    "        embedding_matrix[i] = embedding_vector\n",
    "\n",
    "# load pre-trained word embeddings into an Embedding layer\n",
    "# note that we set trainable = False so as to keep the embeddings fixed\n",
    "embedding_layer = Embedding(num_words,\n",
    "                            EMBEDDING_DIM,\n",
    "                            weights=[embedding_matrix],\n",
    "                            input_length=MAX_SEQUENCE_LENGTH,\n",
    "                            trainable=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# MLP\n",
    "\n",
    "Let's try a simple two-layer dense network, based on the embeddings..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# train a regular MLP\n",
    "sequence_input = Input(shape=(MAX_SEQUENCE_LENGTH,), dtype='float32')\n",
    "embedded_sequences = embedding_layer(sequence_input)\n",
    "x = Flatten()(embedded_sequences)\n",
    "x = Dense(256, activation='relu')(x)\n",
    "x = Dropout(0.5)(x)\n",
    "x = Dense(256, activation='relu')(x)\n",
    "preds = Dense(1, activation='sigmoid')(x)\n",
    "\n",
    "model_mlp = Model(sequence_input, preds)\n",
    "model_mlp.compile(loss='binary_crossentropy',\n",
    "              optimizer='rmsprop',\n",
    "              metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_19 (InputLayer)        (None, 300)               0         \n",
      "_________________________________________________________________\n",
      "embedding_1 (Embedding)      (None, 300, 100)          2000000   \n",
      "_________________________________________________________________\n",
      "flatten_7 (Flatten)          (None, 30000)             0         \n",
      "_________________________________________________________________\n",
      "dense_27 (Dense)             (None, 256)               7680256   \n",
      "_________________________________________________________________\n",
      "dropout_6 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_28 (Dense)             (None, 256)               65792     \n",
      "_________________________________________________________________\n",
      "dense_29 (Dense)             (None, 1)                 257       \n",
      "=================================================================\n",
      "Total params: 9,746,305.0\n",
      "Trainable params: 7,746,305.0\n",
      "Non-trainable params: 2,000,000.0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model_mlp.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training model.\n",
      "Train on 25000 samples, validate on 25000 samples\n",
      "Epoch 1/25\n",
      "25000/25000 [==============================] - 46s - loss: 0.6791 - acc: 0.5932 - val_loss: 0.6450 - val_acc: 0.6370\n",
      "Epoch 2/25\n",
      "25000/25000 [==============================] - 46s - loss: 0.6162 - acc: 0.6568 - val_loss: 0.6020 - val_acc: 0.6690\n",
      "Epoch 3/25\n",
      "25000/25000 [==============================] - 47s - loss: 0.5699 - acc: 0.6976 - val_loss: 0.5575 - val_acc: 0.7260\n",
      "Epoch 4/25\n",
      "25000/25000 [==============================] - 47s - loss: 0.5334 - acc: 0.7247 - val_loss: 0.7564 - val_acc: 0.6234\n",
      "Epoch 5/25\n",
      "25000/25000 [==============================] - 47s - loss: 0.4937 - acc: 0.7499 - val_loss: 0.5565 - val_acc: 0.7216\n",
      "Epoch 6/25\n",
      "25000/25000 [==============================] - 46s - loss: 0.4539 - acc: 0.7766 - val_loss: 0.5664 - val_acc: 0.7245\n",
      "Epoch 7/25\n",
      "25000/25000 [==============================] - 42s - loss: 0.4225 - acc: 0.7924 - val_loss: 0.6253 - val_acc: 0.7342\n",
      "Epoch 8/25\n",
      "25000/25000 [==============================] - 45s - loss: 0.3956 - acc: 0.8080 - val_loss: 0.6632 - val_acc: 0.7139\n",
      "Epoch 9/25\n",
      "25000/25000 [==============================] - 47s - loss: 0.3637 - acc: 0.8274 - val_loss: 0.9327 - val_acc: 0.6280\n",
      "Epoch 10/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.3328 - acc: 0.8417 - val_loss: 0.5827 - val_acc: 0.7404\n",
      "Epoch 11/25\n",
      "25000/25000 [==============================] - 45s - loss: 0.3097 - acc: 0.8548 - val_loss: 0.6028 - val_acc: 0.7301\n",
      "Epoch 12/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.2847 - acc: 0.8709 - val_loss: 0.8806 - val_acc: 0.7048\n",
      "Epoch 13/25\n",
      "25000/25000 [==============================] - 45s - loss: 0.2590 - acc: 0.8859 - val_loss: 0.7205 - val_acc: 0.7378\n",
      "Epoch 14/25\n",
      "25000/25000 [==============================] - 46s - loss: 0.2338 - acc: 0.8959 - val_loss: 0.8090 - val_acc: 0.7148\n",
      "Epoch 15/25\n",
      "25000/25000 [==============================] - 46s - loss: 0.2245 - acc: 0.9040 - val_loss: 0.6873 - val_acc: 0.7218\n",
      "Epoch 16/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.2023 - acc: 0.9128 - val_loss: 0.8681 - val_acc: 0.7143\n",
      "Epoch 17/25\n",
      "25000/25000 [==============================] - 43s - loss: 0.1858 - acc: 0.9189 - val_loss: 0.7540 - val_acc: 0.7313\n",
      "Epoch 18/25\n",
      "25000/25000 [==============================] - 41s - loss: 0.1715 - acc: 0.9276 - val_loss: 0.9520 - val_acc: 0.7395\n",
      "Epoch 19/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.1627 - acc: 0.9314 - val_loss: 0.8516 - val_acc: 0.7267\n",
      "Epoch 20/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.1523 - acc: 0.9374 - val_loss: 0.9227 - val_acc: 0.7353\n",
      "Epoch 21/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.1435 - acc: 0.9417 - val_loss: 0.8689 - val_acc: 0.7311\n",
      "Epoch 22/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.1323 - acc: 0.9449 - val_loss: 0.8892 - val_acc: 0.7320\n",
      "Epoch 23/25\n",
      "25000/25000 [==============================] - 43s - loss: 0.1238 - acc: 0.9514 - val_loss: 1.0707 - val_acc: 0.7171\n",
      "Epoch 24/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.1220 - acc: 0.9520 - val_loss: 0.9777 - val_acc: 0.7136\n",
      "Epoch 25/25\n",
      "25000/25000 [==============================] - 44s - loss: 0.1133 - acc: 0.9540 - val_loss: 0.8902 - val_acc: 0.7292\n"
     ]
    }
   ],
   "source": [
    "print('Training model.')\n",
    "history = model_mlp.fit(x_train, y_train, validation_data=(x_test, y_test),\n",
    "          epochs=25, batch_size=128)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtkAAAD/CAYAAAAkJ8qcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x144206710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_training_curves(history.history);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "24960/25000 [============================>.] - ETA: 0sTest score: 0.890198716736\n",
      "Test accuracy: 0.729199999962\n"
     ]
    }
   ],
   "source": [
    "score, acc = model_mlp.evaluate(x_test, y_test,\n",
    "                            batch_size=128)\n",
    "print('Test score:', score)\n",
    "print('Test accuracy:', acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Overfitting! In practice, would use the weights after ~5 epochs. Get 70% accuracy. Let's see if we can do better with a CNN..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1D CNN\n",
    "\n",
    "It's not a crazy assumption that adjacent words are more important than far away ones, and that the same groups of words in different positions mean roughly the same thing, so we can use our CNN tricks. Sequences are 1D, so we'll use a 1-D sliding window of weights."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# train a 1D convnet with global maxpooling\n",
    "sequence_input = Input(shape=(MAX_SEQUENCE_LENGTH,), dtype='int32')\n",
    "embedded_sequences = embedding_layer(sequence_input)\n",
    "x = Conv1D(128, 5, activation='relu')(embedded_sequences)\n",
    "x = MaxPooling1D(5)(x)\n",
    "x = Conv1D(128, 5, activation='relu')(x)\n",
    "x = MaxPooling1D(5)(x)\n",
    "x = Conv1D(128, 5, activation='relu')(x)\n",
    "x = MaxPooling1D(7)(x)  # modified from example since our seq len is 300 \n",
    "x = Flatten()(x)\n",
    "x = Dense(128, activation='relu')(x)\n",
    "preds = Dense(1, activation='sigmoid')(x)\n",
    "\n",
    "model = Model(sequence_input, preds)\n",
    "model.compile(loss='binary_crossentropy',\n",
    "              optimizer='rmsprop',\n",
    "              metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_9 (InputLayer)         (None, 300)               0         \n",
      "_________________________________________________________________\n",
      "embedding_1 (Embedding)      (None, 300, 100)          2000000   \n",
      "_________________________________________________________________\n",
      "conv1d_25 (Conv1D)           (None, 296, 128)          64128     \n",
      "_________________________________________________________________\n",
      "max_pooling1d_22 (MaxPooling (None, 59, 128)           0         \n",
      "_________________________________________________________________\n",
      "conv1d_26 (Conv1D)           (None, 55, 128)           82048     \n",
      "_________________________________________________________________\n",
      "max_pooling1d_23 (MaxPooling (None, 11, 128)           0         \n",
      "_________________________________________________________________\n",
      "conv1d_27 (Conv1D)           (None, 7, 128)            82048     \n",
      "_________________________________________________________________\n",
      "max_pooling1d_24 (MaxPooling (None, 1, 128)            0         \n",
      "_________________________________________________________________\n",
      "flatten_3 (Flatten)          (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 128)               16512     \n",
      "_________________________________________________________________\n",
      "dense_6 (Dense)              (None, 1)                 129       \n",
      "=================================================================\n",
      "Total params: 2,244,865.0\n",
      "Trainable params: 244,865.0\n",
      "Non-trainable params: 2,000,000.0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training model.\n",
      "Train on 25000 samples, validate on 25000 samples\n",
      "Epoch 1/10\n",
      "25000/25000 [==============================] - 78s - loss: 0.5911 - acc: 0.6814 - val_loss: 0.5472 - val_acc: 0.7362\n",
      "Epoch 2/10\n",
      "25000/25000 [==============================] - 78s - loss: 0.4484 - acc: 0.7970 - val_loss: 0.3987 - val_acc: 0.8168\n",
      "Epoch 3/10\n",
      "25000/25000 [==============================] - 78s - loss: 0.3992 - acc: 0.8235 - val_loss: 0.6086 - val_acc: 0.7018\n",
      "Epoch 4/10\n",
      "25000/25000 [==============================] - 76s - loss: 0.3614 - acc: 0.8424 - val_loss: 0.3474 - val_acc: 0.8494\n",
      "Epoch 5/10\n",
      "25000/25000 [==============================] - 77s - loss: 0.3245 - acc: 0.8599 - val_loss: 0.4183 - val_acc: 0.8120\n",
      "Epoch 6/10\n",
      "25000/25000 [==============================] - 77s - loss: 0.2889 - acc: 0.8784 - val_loss: 0.5135 - val_acc: 0.7538\n",
      "Epoch 7/10\n",
      "25000/25000 [==============================] - 80s - loss: 0.2570 - acc: 0.8942 - val_loss: 0.5595 - val_acc: 0.7968\n",
      "Epoch 8/10\n",
      "25000/25000 [==============================] - 80s - loss: 0.2226 - acc: 0.9121 - val_loss: 0.4079 - val_acc: 0.8461\n",
      "Epoch 9/10\n",
      "25000/25000 [==============================] - 78s - loss: 0.1919 - acc: 0.9251 - val_loss: 0.3701 - val_acc: 0.8431\n",
      "Epoch 10/10\n",
      "25000/25000 [==============================] - 79s - loss: 0.1672 - acc: 0.9372 - val_loss: 0.4250 - val_acc: 0.8436\n"
     ]
    }
   ],
   "source": [
    "print('Training model.')\n",
    "history = model.fit(x_train, y_train, validation_data=(x_test, y_test),\n",
    "          epochs=10, batch_size=128)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtMAAAD/CAYAAAAzBVpVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlnmZOWhrNZgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1369a8810>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_training_curves(history.history);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25000/25000 [==============================] - 25s    \n",
      "Test score: 0.425013681984\n",
      "Test accuracy: 0.843600000019\n"
     ]
    }
   ],
   "source": [
    "score, acc = model.evaluate(x_test, y_test,\n",
    "                            batch_size=128)\n",
    "print('Test score:', score)\n",
    "print('Test accuracy:', acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Still overfitting on the training data, but looks like we have about 84% validation accuracy. Not too bad for a first attempt."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LSTM\n",
    "Just for fun -- dataset isn't really big enough. Using fchollet's parameters.."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "print('Build model...')\n",
    "batch_size = 32\n",
    "maxlen = 80\n",
    "\n",
    "x_train_short = pad_sequences(x_train, maxlen=maxlen)\n",
    "x_test_short = pad_sequences(x_train, maxlen=maxlen)\n",
    "\n",
    "model_lstm = Sequential()\n",
    "model_lstm.add(Embedding(MAX_NB_WORDS, 128))\n",
    "model_lstm.add(LSTM(128, dropout=0.2, recurrent_dropout=0.2))\n",
    "model_lstm.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "# try using different optimizers and different optimizer configs\n",
    "model_lstm.compile(loss='binary_crossentropy',\n",
    "              optimizer='adam',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "print('Train...')\n",
    "model_lstm.fit(x_train_short, y_train, batch_size=batch_size, epochs=15,\n",
    "          validation_data=(x_test_short, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25000/25000 [==============================] - 34s    \n",
      "Test score: 3.83098610489\n",
      "Test accuracy: 0.5068\n"
     ]
    }
   ],
   "source": [
    "score, acc = model_lstm.evaluate(x_test_short, y_test,\n",
    "                            batch_size=batch_size)\n",
    "print('Test score:', score)\n",
    "print('Test accuracy:', acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Perhaps due to different data prep because of the imdb.py bug, this didn't work at all... That's part of the fun of neural networks. To try:\n",
    "\n",
    "* completely remove low frequency words from training data before training network. That way the 80 words we do keep will have more signal. (Can check how much it matters).\n",
    "* try the example unchanged, using older version of TF/Keras."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [default]",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}