Sentiment_Analysis_using_NLP.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "25a0cd07",
   "metadata": {},
   "source": [
    "# Text Sentiment Analysis of IMDB Movie reviews using NLP (Word2Vec and RNN) for MYM Intern Assesment"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6ca3e39",
   "metadata": {},
   "source": [
    "## Assesment Objectives: \n",
    "### - Preprocessing the data\n",
    "### - Converting Text(words) to Vectors using word2vec \n",
    "### - Using the word representations given by word2vec to feed a RNN and training the model\n",
    "### - Evaluating the model and plotting the performance graphs\n",
    "### - Improving the model by Transfer Learning\n",
    "### - Comparing Accuracy of Baseline model, The model and Improved model.\n",
    "### - Testing the model (predicting the model with new review)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "84781652",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gensim==4.2.0\r\n"
     ]
    }
   ],
   "source": [
    "!pip freeze | grep gensim ##Checking the version of Gensim - Word2Vec"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12e2dbaa",
   "metadata": {},
   "source": [
    "## The Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b25eb77",
   "metadata": {},
   "source": [
    "### Starting with 20% of the sentences from TensorFlow Datasets of IMDB reviews to check the RAM compatibility of the PC to train the model faster by splitting the datasets as X_train, y_train, X_test and y_test.\n",
    "### Then preprocessing the textual data to create input features for a natural language processing (NLP) model.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "2079b965",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import tensorflow_datasets as tfds\n",
    "from tensorflow.keras.preprocessing.text import text_to_word_sequence\n",
    "\n",
    "def load_data(percentage_of_sentences=None):\n",
    "    train_data, test_data = tfds.load(name=\"imdb_reviews\", split=[\"train\", \"test\"], batch_size=-1, as_supervised=True)\n",
    "\n",
    "    train_sentences, y_train = tfds.as_numpy(train_data)\n",
    "    test_sentences, y_test = tfds.as_numpy(test_data)\n",
    "    \n",
    "    # Take only a given percentage of the entire data\n",
    "    if percentage_of_sentences is not None:\n",
    "      assert(percentage_of_sentences> 0 and percentage_of_sentences<=100)\n",
    "        \n",
    "      len_train = int(percentage_of_sentences/100*len(train_sentences))\n",
    "      train_sentences, y_train = train_sentences[:len_train], y_train[:len_train]\n",
    "  \n",
    "      len_test = int(percentage_of_sentences/100*len(test_sentences))\n",
    "      test_sentences, y_test = test_sentences[:len_test], y_test[:len_test]\n",
    "    \n",
    "    X_train = [text_to_word_sequence(_.decode(\"utf-8\")) for _ in train_sentences]\n",
    "    X_test = [text_to_word_sequence(_.decode(\"utf-8\")) for _ in test_sentences]\n",
    "    \n",
    "    return X_train, y_train, X_test, y_test\n",
    "\n",
    "X_train, y_train, X_test, y_test = load_data(percentage_of_sentences=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "4352850a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1, 0, ..., 1, 0, 0])"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2d707253",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "fdffccea",
   "metadata": {},
   "source": [
    "## First, training a word2vec model (with the arguments that we want) on your training sentence. Store it into the `word2vec` variable. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "f5c2e1b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "from gensim.models import Word2Vec\n",
    "\n",
    "word2vec = Word2Vec(sentences=X_train, vector_size=60, min_count=10, window=10)\n",
    "word2vec.save(\"word2vec.model\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81a82d8e",
   "metadata": {},
   "source": [
    "## Embedding the training and test sentences."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "62a835d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
    "import numpy as np\n",
    "\n",
    "# Function to convert a sentence (list of words) into a matrix representing the words in the embedding space\n",
    "def embed_sentence(word2vec, sentence):\n",
    "    embedded_sentence = []\n",
    "    for word in sentence:\n",
    "        if word in word2vec.wv:\n",
    "            embedded_sentence.append(word2vec.wv[word])\n",
    "        \n",
    "    return np.array(embedded_sentence)\n",
    "\n",
    "# Function that converts a list of sentences into a list of matrices\n",
    "def embedding(word2vec, sentences):\n",
    "    embed = []\n",
    "    \n",
    "    for sentence in sentences:\n",
    "        embedded_sentence = embed_sentence(word2vec, sentence)\n",
    "        embed.append(embedded_sentence)\n",
    "        \n",
    "    return embed\n",
    "\n",
    "# Embed the training and test sentences\n",
    "X_train_embed = embedding(word2vec, X_train)\n",
    "X_test_embed = embedding(word2vec, X_test)\n",
    "\n",
    "\n",
    "# Pad the training and test embedded sentences\n",
    "X_train_pad = pad_sequences(X_train_embed, dtype='float32', padding='post', maxlen=200)\n",
    "X_test_pad = pad_sequences(X_test_embed, dtype='float32', padding='post', maxlen=200)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea1e76af",
   "metadata": {},
   "source": [
    "### It's a good practice to check check the following for `X_train_pad` and `X_test_pad`:\n",
    "#### - they are numpy arrays\n",
    "#### - they are 3-dimensional\n",
    "#### - the last dimension is of the size of your word2vec embedding space (you can get it with `word2vec.wv.vector_size`\\\\\n",
    "#### - the first dimension is of the size of your `X_train` and `X_test`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "d4770855",
   "metadata": {},
   "outputs": [],
   "source": [
    "for X in [X_train_pad, X_test_pad]:\n",
    "    assert type(X) == np.ndarray\n",
    "    assert X.shape[-1] == word2vec.wv.vector_size\n",
    "\n",
    "\n",
    "assert X_train_pad.shape[0] == len(X_train)\n",
    "assert X_test_pad.shape[0] == len(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6418b6d1",
   "metadata": {},
   "source": [
    "## Baseline Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "3b477d35",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of labels in train set {0: 2474, 1: 2526}\n",
      "Baseline accuracy:  0.499\n"
     ]
    }
   ],
   "source": [
    "# It is always good to have a very simple model to test your own model against\n",
    "# Baseline accuracy can be to predict the label that is the most present in `y_train`.\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "unique, counts = np.unique(y_train, return_counts=True)\n",
    "counts = dict(zip(unique, counts))\n",
    "print('Number of labels in train set', counts)\n",
    "\n",
    "y_pred = 0 if counts[0] > counts[1] else 1\n",
    "\n",
    "print('Baseline accuracy: ', accuracy_score(y_test, [y_pred]*len(y_test)))\n",
    "\n",
    "baseline_acc = accuracy_score(y_test, [y_pred]*len(y_test))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7a80da2",
   "metadata": {},
   "source": [
    "## The Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "523cd8a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras import Sequential\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "# writing a RNN model with Masking, LSTM and Dense layers.\n",
    "\n",
    "def init_model():\n",
    "    model = Sequential()\n",
    "    model.add(layers.Masking())\n",
    "    model.add(layers.LSTM(20, activation='tanh'))\n",
    "    model.add(layers.Dense(15, activation='relu'))\n",
    "    model.add(layers.Dense(1, activation='sigmoid'))\n",
    "\n",
    "    model.compile(loss='binary_crossentropy', #compiling the model with rmsprop optimizer\n",
    "                  optimizer='rmsprop',\n",
    "                  metrics=['accuracy'])\n",
    "    \n",
    "    return model\n",
    "\n",
    "model = init_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "5317ce64",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "110/110 [==============================] - 8s 38ms/step - loss: 0.6787 - accuracy: 0.5734 - val_loss: 0.6713 - val_accuracy: 0.5813\n",
      "Epoch 2/100\n",
      "110/110 [==============================] - 4s 33ms/step - loss: 0.6151 - accuracy: 0.6729 - val_loss: 0.6193 - val_accuracy: 0.6520\n",
      "Epoch 3/100\n",
      "110/110 [==============================] - 4s 33ms/step - loss: 0.5585 - accuracy: 0.7160 - val_loss: 0.5868 - val_accuracy: 0.7000\n",
      "Epoch 4/100\n",
      "110/110 [==============================] - 4s 32ms/step - loss: 0.5229 - accuracy: 0.7449 - val_loss: 0.5764 - val_accuracy: 0.7113\n",
      "Epoch 5/100\n",
      "110/110 [==============================] - 4s 33ms/step - loss: 0.4947 - accuracy: 0.7623 - val_loss: 0.5630 - val_accuracy: 0.7220\n",
      "Epoch 6/100\n",
      "110/110 [==============================] - 4s 32ms/step - loss: 0.4740 - accuracy: 0.7786 - val_loss: 0.5570 - val_accuracy: 0.7200\n",
      "Epoch 7/100\n",
      "110/110 [==============================] - 4s 32ms/step - loss: 0.4569 - accuracy: 0.7911 - val_loss: 0.5202 - val_accuracy: 0.7607\n",
      "Epoch 8/100\n",
      "110/110 [==============================] - 4s 33ms/step - loss: 0.4430 - accuracy: 0.8009 - val_loss: 0.5688 - val_accuracy: 0.7153\n",
      "Epoch 9/100\n",
      "110/110 [==============================] - 4s 32ms/step - loss: 0.4278 - accuracy: 0.8154 - val_loss: 0.5160 - val_accuracy: 0.7640\n",
      "Epoch 10/100\n",
      "110/110 [==============================] - 4s 32ms/step - loss: 0.4107 - accuracy: 0.8203 - val_loss: 0.5312 - val_accuracy: 0.7527\n",
      "Epoch 11/100\n",
      "110/110 [==============================] - 4s 33ms/step - loss: 0.3973 - accuracy: 0.8274 - val_loss: 0.5955 - val_accuracy: 0.7140\n",
      "Epoch 12/100\n",
      "110/110 [==============================] - 4s 33ms/step - loss: 0.3867 - accuracy: 0.8340 - val_loss: 0.5898 - val_accuracy: 0.7367\n",
      "Epoch 13/100\n",
      "110/110 [==============================] - 4s 32ms/step - loss: 0.3772 - accuracy: 0.8389 - val_loss: 0.5303 - val_accuracy: 0.7647\n",
      "Epoch 14/100\n",
      "110/110 [==============================] - 4s 33ms/step - loss: 0.3615 - accuracy: 0.8457 - val_loss: 0.5863 - val_accuracy: 0.7480\n"
     ]
    }
   ],
   "source": [
    "#  Fiting the model on embedded and padded data with the early stopping criterion.\n",
    "\n",
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "\n",
    "es = EarlyStopping(patience=5, restore_best_weights=True)\n",
    "\n",
    "history = model.fit(X_train_pad, y_train, \n",
    "          batch_size = 32,\n",
    "          epochs=100,\n",
    "          validation_split=0.3,\n",
    "          callbacks=[es]\n",
    "         )\n",
    "the_model_acc = history.history['accuracy'][-1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "26e4350b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The accuracy evaluated on the test set is of 76.200%\n"
     ]
    }
   ],
   "source": [
    "# Evaluating the model on the test set.\n",
    "\n",
    "result = model.evaluate(X_test_pad, y_test, verbose=0)\n",
    "\n",
    "print(f'The accuracy evaluated on the test set is of {result[1]*100:.3f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "8d663411",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Plot the accuracy and loss curves\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(history.history['accuracy'], label='Training accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation accuracy')\n",
    "plt.title('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(history.history['loss'], label='Training loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation loss')\n",
    "plt.title('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9c43269",
   "metadata": {},
   "source": [
    "## Trained Word2Vec - Transfer Learning\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92b6edab",
   "metadata": {},
   "source": [
    "### The accuracy of the above the baseline model, might be quite low. By improving the quality of the embedding we can Improve accuracy of the model."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e80388e",
   "metadata": {},
   "source": [
    "### Let's improve the quality of our embedding, instead of just loading a larger corpus, let's benefit from the embedding that others have learned. Because, the quality of an embedding, i.e. the proximity of the words, can be derived from different tasks. This is exactly what transfer learning is."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "97d4ebeb",
   "metadata": {},
   "source": [
    "### Listing all the different models available in the word2vec using gensim api."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "b07b77ad",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['fasttext-wiki-news-subwords-300', 'conceptnet-numberbatch-17-06-300', 'word2vec-ruscorpora-300', 'word2vec-google-news-300', 'glove-wiki-gigaword-50', 'glove-wiki-gigaword-100', 'glove-wiki-gigaword-200', 'glove-wiki-gigaword-300', 'glove-twitter-25', 'glove-twitter-50', 'glove-twitter-100', 'glove-twitter-200', '__testing_word2vec-matrix-synopsis']\n"
     ]
    }
   ],
   "source": [
    "import gensim.downloader as api\n",
    "print(list(api.info()['models'].keys()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "bd25af6a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Let's load one of the pre-trained word2vec embedding spaces. \n",
    "\n",
    "word2vec_transfer = api.load(\"glove-wiki-gigaword-100\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "062fa47f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "400000\n",
      "100\n"
     ]
    }
   ],
   "source": [
    "print(len(word2vec_transfer.key_to_index))\n",
    "print(len(word2vec_transfer['dog']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "bc0a78fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to convert a sentence (list of words) into a matrix representing the words in the embedding space\n",
    "def embed_sentence_with_TF(word2vec, sentence):\n",
    "    embedded_sentence = []\n",
    "    for word in sentence:\n",
    "        if word in word2vec:\n",
    "            embedded_sentence.append(word2vec[word])\n",
    "        \n",
    "    return np.array(embedded_sentence)\n",
    "\n",
    "# Function that converts a list of sentences into a list of matrices\n",
    "def embedding(word2vec, sentences):\n",
    "    embed = []\n",
    "    \n",
    "    for sentence in sentences:\n",
    "        embedded_sentence = embed_sentence_with_TF(word2vec, sentence)\n",
    "        embed.append(embedded_sentence)\n",
    "        \n",
    "    return embed\n",
    "\n",
    "# Embed the training and test sentences\n",
    "X_train_embed_2 = embedding(word2vec_transfer, X_train)\n",
    "X_test_embed_2 = embedding(word2vec_transfer, X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "9270ed0e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pad the training and test embedded sentences\n",
    "X_train_pad_2 = pad_sequences(X_train_embed_2, dtype='float32', padding='post', maxlen=200)\n",
    "X_test_pad_2 = pad_sequences(X_test_embed_2, dtype='float32', padding='post', maxlen=200)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "cdee2b55",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "110/110 [==============================] - 7s 41ms/step - loss: 0.6757 - accuracy: 0.5751 - val_loss: 0.7198 - val_accuracy: 0.5120\n",
      "Epoch 2/30\n",
      "110/110 [==============================] - 4s 34ms/step - loss: 0.6235 - accuracy: 0.6537 - val_loss: 0.6080 - val_accuracy: 0.6733\n",
      "Epoch 3/30\n",
      "110/110 [==============================] - 4s 34ms/step - loss: 0.5765 - accuracy: 0.7043 - val_loss: 0.5740 - val_accuracy: 0.7160\n",
      "Epoch 4/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.5460 - accuracy: 0.7291 - val_loss: 0.5546 - val_accuracy: 0.7260\n",
      "Epoch 5/30\n",
      "110/110 [==============================] - 4s 34ms/step - loss: 0.5145 - accuracy: 0.7531 - val_loss: 0.6563 - val_accuracy: 0.6760\n",
      "Epoch 6/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.4897 - accuracy: 0.7734 - val_loss: 0.5021 - val_accuracy: 0.7713\n",
      "Epoch 7/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.4691 - accuracy: 0.7817 - val_loss: 0.5435 - val_accuracy: 0.7407\n",
      "Epoch 8/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.4414 - accuracy: 0.7940 - val_loss: 0.6042 - val_accuracy: 0.6833\n",
      "Epoch 9/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.4280 - accuracy: 0.8060 - val_loss: 0.5480 - val_accuracy: 0.7407\n",
      "Epoch 10/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.4076 - accuracy: 0.8211 - val_loss: 0.4953 - val_accuracy: 0.7860\n",
      "Epoch 11/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.3866 - accuracy: 0.8246 - val_loss: 0.5699 - val_accuracy: 0.7707\n",
      "Epoch 12/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.3736 - accuracy: 0.8320 - val_loss: 0.5021 - val_accuracy: 0.7667\n",
      "Epoch 13/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.3565 - accuracy: 0.8420 - val_loss: 0.4778 - val_accuracy: 0.7947\n",
      "Epoch 14/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.3434 - accuracy: 0.8491 - val_loss: 0.4844 - val_accuracy: 0.7800\n",
      "Epoch 15/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.3312 - accuracy: 0.8589 - val_loss: 0.5314 - val_accuracy: 0.7673\n",
      "Epoch 16/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.3152 - accuracy: 0.8686 - val_loss: 0.4895 - val_accuracy: 0.7933\n",
      "Epoch 17/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.3062 - accuracy: 0.8711 - val_loss: 0.6577 - val_accuracy: 0.7420\n",
      "Epoch 18/30\n",
      "110/110 [==============================] - 4s 35ms/step - loss: 0.2922 - accuracy: 0.8766 - val_loss: 0.5764 - val_accuracy: 0.7693\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "import tensorflow as tf\n",
    "\n",
    "es = EarlyStopping(patience=5, restore_best_weights=True)\n",
    "\n",
    "model = init_model()\n",
    "\n",
    "history = model.fit(X_train_pad_2, y_train, \n",
    "          batch_size = 32,\n",
    "          epochs=30,\n",
    "          validation_split=0.3,\n",
    "          callbacks=[es]\n",
    "         )\n",
    "model.save('my_model.h5')\n",
    "improved_model_acc = history.history['accuracy'][-1]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "d8297abb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The accuracy evaluated on the test set is of 80.800%\n"
     ]
    }
   ],
   "source": [
    "result = model.evaluate(X_test_pad_2, y_test, verbose=0)\n",
    "\n",
    "print(f'The accuracy evaluated on the test set is of {result[1]*100:.3f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "070d098d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the accuracy and loss curves\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(history.history['accuracy'], label='Training accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation accuracy')\n",
    "plt.title('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(history.history['loss'], label='Training loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation loss')\n",
    "plt.title('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81d63cb7",
   "metadata": {},
   "source": [
    "### There is a significant improvement in the accuracy after Transfer learning."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "379655e1",
   "metadata": {},
   "source": [
    "## Comparing Accuracy of Baseline model, The model and Improved model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "44b32cc6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting the Graph\n",
    "plt.plot([baseline_acc, the_model_acc, improved_model_acc], marker='o')\n",
    "plt.xticks([0, 1, 2], ['Baseline Model', 'The Model', 'The Improved Model'])\n",
    "plt.ylabel('Accuracy')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "043e1753",
   "metadata": {},
   "source": [
    "## Predicting the model for new review."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "645e44d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras import Sequential\n",
    "from tensorflow.keras.layers import Masking, LSTM, Dense\n",
    "\n",
    "# Load the pre-trained Word2Vec model\n",
    "word2vec_transfer = api.load(\"glove-wiki-gigaword-100\")\n",
    "\n",
    "# Define the function to embed a sentence with the pre-trained Word2Vec model\n",
    "def embed_sentence_with_TF(word2vec, sentence):\n",
    "    embedded_sentence = []\n",
    "    for word in sentence:\n",
    "        if word in word2vec:\n",
    "            embedded_sentence.append(word2vec[word])\n",
    "    return np.array(embedded_sentence)\n",
    "\n",
    "# Define the function to preprocess a new movie review\n",
    "def preprocess_review(review):\n",
    "    # Tokenize the review\n",
    "    review = text_to_word_sequence(review)\n",
    "    # Embed the review with the pre-trained Word2Vec model\n",
    "    review_embedded = embed_sentence_with_TF(word2vec_transfer, review)\n",
    "    # Pad the embedded review\n",
    "    review_padded = pad_sequences([review_embedded], dtype='float32', padding='post', maxlen=200)\n",
    "    return review_padded\n",
    "\n",
    "# Load the trained model\n",
    "model = Sequential()\n",
    "model.add(Masking())\n",
    "model.add(LSTM(20, activation='tanh'))\n",
    "model.add(Dense(15, activation='relu'))\n",
    "model.add(Dense(1, activation='sigmoid'))\n",
    "model.compile(loss='binary_crossentropy', optimizer='rmsprop', metrics=['accuracy'])\n",
    "model = load_model('my_model.h5')\n",
    "def predict_sentiment(review):\n",
    "    # Preprocess the review\n",
    "    review_padded = preprocess_review(review)\n",
    "    # Predict the sentiment\n",
    "    sentiment = model.predict(review_padded)[0][0]\n",
    "    return sentiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "faf5685a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 70ms/step\n",
      "Positive review\n"
     ]
    }
   ],
   "source": [
    "review = \"The movie is good.\"\n",
    "sentiment = predict_sentiment(review)\n",
    "if sentiment > 0.5:\n",
    "    print(\"Positive review\")\n",
    "else:\n",
    "    print(\"Negative review\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "1949d6e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "!git add Sentiment_Analysis_using_NLP.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "9de36bef",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[master fc27b7b] Sentiment Analysis using NLP\r\n",
      " 1 file changed, 47 insertions(+), 51 deletions(-)\r\n"
     ]
    }
   ],
   "source": [
    "!git commit -m 'Sentiment Analysis using NLP'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "c7bb06bc",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "On branch master\r\n",
      "Your branch is ahead of 'origin/master' by 1 commit.\r\n",
      "  (use \"git push\" to publish your local commits)\r\n",
      "\r\n",
      "Untracked files:\r\n",
      "  (use \"git add <file>...\" to include in what will be committed)\r\n",
      "\t\u001b[31m.DS_Store\u001b[m\r\n",
      "\t\u001b[31mmy_model.h5\u001b[m\r\n",
      "\t\u001b[31mmy_model/\u001b[m\r\n",
      "\t\u001b[31mmy_model_weights.h5\u001b[m\r\n",
      "\t\u001b[31msaved_model.pb\u001b[m\r\n",
      "\t\u001b[31mword2vec.model\u001b[m\r\n",
      "\t\u001b[31m~/\u001b[m\r\n",
      "\r\n",
      "nothing added to commit but untracked files present (use \"git add\" to track)\r\n"
     ]
    }
   ],
   "source": [
    "!git status"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "0c45598a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Already on 'master'\r\n",
      "Your branch is ahead of 'origin/master' by 1 commit.\r\n",
      "  (use \"git push\" to publish your local commits)\r\n"
     ]
    }
   ],
   "source": [
    "!git checkout master"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "8019355a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Enumerating objects: 5, done.\n",
      "Counting objects: 100% (5/5), done.\n",
      "Delta compression using up to 8 threads\n",
      "Compressing objects: 100% (3/3), done.\n",
      "Writing objects: 100% (3/3), 1.26 KiB | 214.00 KiB/s, done.\n",
      "Total 3 (delta 2), reused 0 (delta 0), pack-reused 0\n",
      "remote: Resolving deltas: 100% (2/2), completed with 2 local objects.\u001b[K\n",
      "remote: This repository moved. Please use the new location:\u001b[K\n",
      "remote:   git@github.com:pavankumarhm/Sentiment-Analysis-for-MYM-Intern-Assesment.git\u001b[K\n",
      "To github.com:pavankumarhm/Sentiment-Analysis-Assesment.git\n",
      "   161d352..fc27b7b  master -> master\n"
     ]
    }
   ],
   "source": [
    "!git push"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "44f6aaed",
   "metadata": {},
   "outputs": [],
   "source": [
    "!git status\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "1a3612a0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "From github.com:pavankumarhm/Sentiment-Analysis-Assesment\n",
      " * branch            master     -> FETCH_HEAD\n",
      "Already up to date.\n"
     ]
    }
   ],
   "source": [
    "!git pull origin master"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "da9ae35d",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "error: pathspec 'my-feature' did not match any file(s) known to git\r\n"
     ]
    }
   ],
   "source": [
    "!git checkout my-feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "498dd1ec",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Already up to date.\r\n"
     ]
    }
   ],
   "source": [
    "!git merge master"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "dc166b4f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "branch 'master' set up to track 'origin/master'.\r\n",
      "Everything up-to-date\r\n"
     ]
    }
   ],
   "source": [
    "!git push -u origin master"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d5f2bc35",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.6"
  },
  "toc": {
   "base_numbering": "1",
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": true,
   "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": 5
}