IMDB movie reviews sentiment classification - CNN Model - Keras.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.datasets import imdb\n",
    "from keras.preprocessing import sequence\n",
    "\n",
    "from keras.models import Sequential\n",
    "from keras.layers.embeddings import Embedding\n",
    "from keras.layers import Conv1D, LeakyReLU, MaxPooling1D, Dropout, Activation, Flatten, Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loading the dataset\n",
    "top_words = 5000\n",
    "(X_train, y_train), (X_test, y_test) = imdb.load_data(num_words=top_words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preprocessing sestences\n",
    "max_review_len = 500\n",
    "X_train = sequence.pad_sequences(X_train, maxlen=max_review_len)\n",
    "X_test = sequence.pad_sequences(X_test, maxlen=max_review_len)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "VOCAB_SIZE = 5000\n",
    "EMBEDDING_SIZE = 32\n",
    "EPOCHS = 3\n",
    "BATCH_SIZE = 64\n",
    "KEEP_PROB = 0.75"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "embedding_1 (Embedding)      (None, 500, 32)           160000    \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 500, 32)           0         \n",
      "_________________________________________________________________\n",
      "conv1d_1 (Conv1D)            (None, 499, 32)           2080      \n",
      "_________________________________________________________________\n",
      "leaky_re_lu_1 (LeakyReLU)    (None, 499, 32)           0         \n",
      "_________________________________________________________________\n",
      "max_pooling1d_1 (MaxPooling1 (None, 249, 32)           0         \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 249, 32)           0         \n",
      "_________________________________________________________________\n",
      "conv1d_2 (Conv1D)            (None, 248, 32)           2080      \n",
      "_________________________________________________________________\n",
      "leaky_re_lu_2 (LeakyReLU)    (None, 248, 32)           0         \n",
      "_________________________________________________________________\n",
      "max_pooling1d_2 (MaxPooling1 (None, 124, 32)           0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 3968)              0         \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 3968)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 256)               1016064   \n",
      "_________________________________________________________________\n",
      "dropout_4 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1)                 257       \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 1)                 0         \n",
      "=================================================================\n",
      "Total params: 1,180,481\n",
      "Trainable params: 1,180,481\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Building the model\n",
    "\n",
    "model = Sequential()\n",
    "\n",
    "model.add(Embedding(VOCAB_SIZE, EMBEDDING_SIZE, input_length=max_review_len))\n",
    "model.add(Dropout(1 - KEEP_PROB))\n",
    "\n",
    "model.add(Conv1D(32, 2, strides=1, activation='linear', padding='valid'))\n",
    "model.add(LeakyReLU(alpha=0.1))\n",
    "model.add(MaxPooling1D(pool_size=2))\n",
    "model.add(Dropout(1 - KEEP_PROB))\n",
    "\n",
    "model.add(Conv1D(32, 2, strides=1, activation='linear', padding='valid'))\n",
    "model.add(LeakyReLU(alpha=0.1))\n",
    "model.add(MaxPooling1D(pool_size=2))\n",
    "model.add(Flatten())\n",
    "model.add(Dropout(1 - KEEP_PROB))\n",
    "\n",
    "model.add(Dense(256))\n",
    "model.add(Dropout(1 - KEEP_PROB))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dense(1))\n",
    "model.add(Activation('sigmoid'))\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/3\n",
      "25000/25000 [==============================] - 59s 2ms/step - loss: 0.4656 - acc: 0.7409\n",
      "Epoch 2/3\n",
      "25000/25000 [==============================] - 61s 2ms/step - loss: 0.2530 - acc: 0.8990\n",
      "Epoch 3/3\n",
      "25000/25000 [==============================] - 61s 2ms/step - loss: 0.2096 - acc: 0.9179\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1879ff07b38>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train, epochs=EPOCHS, batch_size=BATCH_SIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 87.93%\n"
     ]
    }
   ],
   "source": [
    "scores = model.evaluate(X_test, y_test, verbose=0)\n",
    "print(\"Accuracy: %.2f%%\" % (scores[1]*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}