12. CNN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 12.2.2 CNN을 이용한 NLTK 영화 리뷰 분류"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\anaconda3\\lib\\site-packages\\numpy\\_distributor_init.py:30: UserWarning: loaded more than 1 DLL from .libs:\n",
      "d:\\anaconda3\\lib\\site-packages\\numpy\\.libs\\libopenblas.4SP5SUA7CBGXUEOC35YP2ASOICYYEQZZ.gfortran-win_amd64.dll\n",
      "d:\\anaconda3\\lib\\site-packages\\numpy\\.libs\\libopenblas64__v0.3.23-gcc_10_3_0.dll\n",
      "  warnings.warn(\"loaded more than 1 DLL from .libs:\"\n"
     ]
    }
   ],
   "source": [
    "from nltk.corpus import movie_reviews\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from keras.preprocessing.text import Tokenizer\n",
    "from keras.utils import pad_sequences\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# movie review data에서 file id를 가져옴\n",
    "fileids = movie_reviews.fileids()\n",
    "\n",
    "# file id를 이용해 raw text file을 가져옴\n",
    "reviews = [ movie_reviews.raw(fileid) for fileid in fileids ]\n",
    "categories = [ movie_reviews.categories(fileid)[0] for fileid in fileids ]\n",
    "\n",
    "np.random.seed(7)\n",
    "tf.random.set_seed(7)\n",
    "\n",
    "max_words = 10000  # 사용할 단어의 수\n",
    "maxlen = 500  # 문서의 단어 수를 제한\n",
    "\n",
    "# 빈도가 높은 10000개의 단어를 선택하도록 객체 생성\n",
    "tokenizer = Tokenizer(num_words = max_words, oov_token='UNK')\n",
    "tokenizer.fit_on_texts(reviews)  # 단어 인덱스 구축\n",
    "\n",
    "X = tokenizer.texts_to_sequences(reviews)  # 만들어진 단어 인덱스를 이용해 반환\n",
    "X = pad_sequences(X, maxlen=maxlen, padding='pre')\n",
    "\n",
    "# label을 0, 1의 값으로 변환\n",
    "label_dict = {'pos':0, 'neg':1}\n",
    "y = np.array([ label_dict[c] for c in categories ])\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " embedding (Embedding)       (None, 500, 64)           640000    \n",
      "                                                                 \n",
      " conv1d (Conv1D)             (None, 496, 128)          41088     \n",
      "                                                                 \n",
      " max_pooling1d (MaxPooling1D  (None, 248, 128)         0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv1d_1 (Conv1D)           (None, 244, 256)          164096    \n",
      "                                                                 \n",
      " max_pooling1d_1 (MaxPooling  (None, 122, 256)         0         \n",
      " 1D)                                                             \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 31232)             0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 64)                1998912   \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 1)                 65        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 2,844,161\n",
      "Trainable params: 2,844,161\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Conv1D, MaxPooling1D, Embedding, Flatten, Dropout\n",
    "from keras.optimizers import Adam\n",
    "\n",
    "model = Sequential([\n",
    "    Embedding(input_dim=max_words, output_dim=64, input_length=maxlen),\n",
    "    Conv1D(\n",
    "        filters=128,\n",
    "        kernel_size=5,  \n",
    "        padding='valid',\n",
    "        activation='relu',\n",
    "        strides=1\n",
    "    ),\n",
    "    MaxPooling1D(pool_size=2),\n",
    "    Conv1D(\n",
    "        filters=256,\n",
    "        kernel_size=5,\n",
    "        padding='valid',\n",
    "        activation='relu',\n",
    "        strides=1\n",
    "    ),\n",
    "    MaxPooling1D(pool_size=2),\n",
    "    Flatten(),\n",
    "    Dense(64, activation='relu'),\n",
    "    Dense(1, activation='sigmoid')  # binary logistic regression\n",
    "])\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "adam = Adam(learning_rate=1e-3)\n",
    "model.compile(\n",
    "    optimizer = adam,\n",
    "    loss = 'binary_crossentropy',\n",
    "    metrics = ['acc']\n",
    ")\n",
    "\n",
    "history = model.fit(\n",
    "    X_train, y_train,\n",
    "    epochs = 20,\n",
    "    batch_size = 256,\n",
    "    validation_split = 0.2,\n",
    "    verbose=0\n",
    ")\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('ggplot')\n",
    "\n",
    "fig = plt.figure(figsize = (10,6))\n",
    "ax1 = fig.add_subplot(1,2,1)\n",
    "ax1.plot(history.history['acc'], 'r:',label='train accuracy')\n",
    "ax1.plot(history.history['val_acc'], 'b:',label='valid accuracy')\n",
    "ax1.legend()\n",
    "ax1.set_title('Accuracy Graph')\n",
    "ax1.set_xlabel('Epochs')\n",
    "ax1.set_ylabel('Accuracy')\n",
    "\n",
    "ax2 = fig.add_subplot(1,2,2)\n",
    "ax2.plot(history.history['loss'], 'r-', label='train loss')\n",
    "ax2.plot(history.history['val_loss'], 'b-', label='valid loss')\n",
    "ax2.legend()\n",
    "ax2.set_title('Loss Graph')\n",
    "ax2.set_xlabel('Epochs')\n",
    "ax2.set_ylabel('Loss')\n",
    "\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13/13 [==============================] - 1s 46ms/step - loss: 1.2652 - acc: 0.8200\n",
      "Test accuracy : 0.820\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(X_test, y_test)\n",
    "print(\"Test accuracy : {:.3f}\".format(score[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.9.13"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}