Text Classification methods in NLP using Deep Learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Text Classification Methods in NLP using Deep Learning\n",
    "#### Using pre-trained word embeddings \n",
    "- implementations are all based in Keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import logging, sys\n",
    "logging.disable(sys.maxsize)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_path = keras.utils.get_file(\n",
    "    \"news20.tar.gz\",\n",
    "    \"http://www.cs.cmu.edu/afs/cs.cmu.edu/project/theo-20/www/data/news20.tar.gz\",\n",
    "    untar=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pathlib\n",
    "\n",
    "data_dir = pathlib.Path(data_path).parent / \"20_newsgroup\"\n",
    "dirnames = os.listdir(data_dir)\n",
    "print(\"Number of directories:\", len(dirnames))\n",
    "print(\"Directory names:\", dirnames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "samples = []\n",
    "labels = []\n",
    "class_names = []\n",
    "class_index = 0\n",
    "for dirname in sorted(os.listdir(data_dir)):\n",
    "    class_names.append(dirname)\n",
    "    dirpath = data_dir / dirname\n",
    "    fnames = os.listdir(dirpath)\n",
    "    print(\"Processing %s, %d files found\" % (dirname, len(fnames)))\n",
    "    for fname in fnames:\n",
    "        fpath = dirpath / fname\n",
    "        f = open(fpath, encoding=\"latin-1\")\n",
    "        content = f.read()\n",
    "        lines = content.split(\"\\n\")\n",
    "        lines = lines[10:]\n",
    "        content = \"\\n\".join(lines)\n",
    "        samples.append(content)\n",
    "        labels.append(class_index)\n",
    "    class_index += 1\n",
    "\n",
    "print(\"Classes:\", class_names)\n",
    "print(\"Number of samples:\", len(samples))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Shuffle the data\n",
    "seed = 1337\n",
    "rng = np.random.RandomState(seed)\n",
    "rng.shuffle(samples)\n",
    "rng = np.random.RandomState(seed)\n",
    "rng.shuffle(labels)\n",
    "\n",
    "# Extract a training & validation split\n",
    "validation_split = 0.2\n",
    "num_validation_samples = int(validation_split * len(samples))\n",
    "train_samples = samples[:-num_validation_samples]\n",
    "val_samples = samples[-num_validation_samples:]\n",
    "train_labels = labels[:-num_validation_samples]\n",
    "val_labels = labels[-num_validation_samples:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#creating vocabulary index \n",
    "from tensorflow.keras.layers.experimental.preprocessing import TextVectorization\n",
    "\n",
    "vectorizer = TextVectorization(max_tokens=20000, output_sequence_length=200)\n",
    "text_ds = tf.data.Dataset.from_tensor_slices(train_samples).batch(128)\n",
    "vectorizer.adapt(text_ds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['', '[UNK]', 'the', 'to', 'of']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectorizer.get_vocabulary()[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "voc = vectorizer.get_vocabulary()\n",
    "word_index = dict(zip(voc, range(len(voc))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#load pretrained word embeddings \n",
    "!wget http://nlp.stanford.edu/data/glove.6B.zip\n",
    "!unzip -q glove.6B.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 400000 word vectors.\n"
     ]
    }
   ],
   "source": [
    "#making a dict mapping words (strings) to their NumPy vector representation\n",
    "path_to_glove_file = os.path.join(\n",
    "    os.path.expanduser(\"~\"), \".keras/datasets/glove.6B.100d.txt\"\n",
    ")\n",
    "\n",
    "embeddings_index = {}\n",
    "with open(path_to_glove_file, encoding=\"utf8\") as f:\n",
    "    for line in f:\n",
    "        word, coefs = line.split(maxsplit=1)\n",
    "        coefs = np.fromstring(coefs, \"f\", sep=\" \")\n",
    "        embeddings_index[word] = coefs\n",
    "\n",
    "print(\"Found %s word vectors.\" % len(embeddings_index))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Converted 18019 words (1981 misses)\n"
     ]
    }
   ],
   "source": [
    "num_tokens = len(voc) + 2\n",
    "embedding_dim = 100\n",
    "hits = 0\n",
    "misses = 0\n",
    "\n",
    "# Prepare embedding matrix\n",
    "embedding_matrix = np.zeros((num_tokens, embedding_dim))\n",
    "for word, i in word_index.items():\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",
    "        # This includes the representation for \"padding\" and \"OOV\"\n",
    "        embedding_matrix[i] = embedding_vector\n",
    "        hits += 1\n",
    "    else:\n",
    "        misses += 1\n",
    "print(\"Converted %d words (%d misses)\" % (hits, misses))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#loading the pre-trained word embeddings matrix into an Embedding layer\n",
    "from tensorflow.keras.layers import Embedding\n",
    "\n",
    "embedding_layer = Embedding(\n",
    "    num_tokens,\n",
    "    embedding_dim,\n",
    "    embeddings_initializer=keras.initializers.Constant(embedding_matrix),\n",
    "    trainable=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_3\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_2 (InputLayer)         [(None, None)]            0         \n",
      "_________________________________________________________________\n",
      "embedding (Embedding)        (None, None, 100)         2000200   \n",
      "_________________________________________________________________\n",
      "conv1d (Conv1D)              (None, None, 128)         64128     \n",
      "_________________________________________________________________\n",
      "max_pooling1d (MaxPooling1D) (None, None, 128)         0         \n",
      "_________________________________________________________________\n",
      "conv1d_1 (Conv1D)            (None, None, 128)         82048     \n",
      "_________________________________________________________________\n",
      "max_pooling1d_1 (MaxPooling1 (None, None, 128)         0         \n",
      "_________________________________________________________________\n",
      "conv1d_2 (Conv1D)            (None, None, 128)         82048     \n",
      "_________________________________________________________________\n",
      "global_max_pooling1d (Global (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 128)               16512     \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_6 (Dense)              (None, 20)                2580      \n",
      "=================================================================\n",
      "Total params: 2,247,516\n",
      "Trainable params: 247,316\n",
      "Non-trainable params: 2,000,200\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#building a simple 1D convnet with global max pooling and a classifier at the end\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "int_sequences_input = keras.Input(shape=(None,), dtype=\"int64\")\n",
    "embedded_sequences = embedding_layer(int_sequences_input)\n",
    "x = layers.Conv1D(128, 5, activation=\"relu\")(embedded_sequences)\n",
    "x = layers.MaxPooling1D(5)(x)\n",
    "x = layers.Conv1D(128, 5, activation=\"relu\")(x)\n",
    "x = layers.MaxPooling1D(5)(x)\n",
    "x = layers.Conv1D(128, 5, activation=\"relu\")(x)\n",
    "x = layers.GlobalMaxPooling1D()(x)\n",
    "x = layers.Dense(128, activation=\"relu\")(x)\n",
    "x = layers.Dropout(0.5)(x)\n",
    "preds = layers.Dense(len(class_names), activation=\"softmax\")(x)\n",
    "model = keras.Model(int_sequences_input, preds)\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "cp=ModelCheckpoint('model_cnn.hdf5',monitor='val_acc',verbose=1,save_best_only=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#model training \n",
    "#convert our list-of-strings data to NumPy arrays of integer indices. The arrays are right-padded\n",
    "x_train = vectorizer(np.array([[s] for s in train_samples])).numpy()\n",
    "x_val = vectorizer(np.array([[s] for s in val_samples])).numpy()\n",
    "\n",
    "y_train = np.array(train_labels)\n",
    "y_val = np.array(val_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#we use sparse_categorical_crossentropy since our labels are integers and we use softmax classification \n",
    "model.compile(\n",
    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"rmsprop\", metrics=[\"acc\"]\n",
    ")\n",
    "history= model.fit(x_train, y_train, batch_size=128, epochs=20, validation_data=(x_val, y_val), callbacks=[cp])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_samples = vectorizer(np.array([[s] for s in samples])).numpy()\n",
    "preds = model.predict(full_samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "#clustering \n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.metrics.cluster import normalized_mutual_info_score as nmi_score \n",
    "import statistics "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "km = KMeans(n_clusters = 20, init=\"random\", n_init=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 3, 17, 11, ..., 10,  5,  0])"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "km.fit_predict(preds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8578260737102869\n"
     ]
    }
   ],
   "source": [
    "print(nmi_score(labels,km.labels_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig1 = plt.figure()\n",
    "plt.plot(history.history['loss'],'r',linewidth=3.0)\n",
    "plt.plot(history.history['val_loss'],'b',linewidth=3.0)\n",
    "plt.legend(['Training loss', 'Validation Loss'],fontsize=18)\n",
    "plt.xlabel('Epochs ',fontsize=16)\n",
    "plt.ylabel('Loss',fontsize=16)\n",
    "plt.title('Loss Curves :CNN',fontsize=16)\n",
    "fig1.savefig('loss_cnn.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig2=plt.figure()\n",
    "plt.plot(history.history['acc'],'r',linewidth=3.0)\n",
    "plt.plot(history.history['val_acc'],'b',linewidth=3.0)\n",
    "plt.legend(['Training Accuracy', 'Validation Accuracy'],fontsize=18)\n",
    "plt.xlabel('Epochs ',fontsize=16)\n",
    "plt.ylabel('Accuracy',fontsize=16)\n",
    "plt.title('Accuracy Curves : CNN',fontsize=16)\n",
    "fig2.savefig('accuracy_cnn.png')\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}