tweaked sentiment and embedding model nlp.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "085a6c32-526c-48e4-a838-54bcd953dfd6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.preprocessing.text import Tokenizer\n",
    "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "18a19c34-3b7e-4a44-b6df-74c5ade8dcd3",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = pd.read_csv(\"D:\\Data Science\\Datasets\\combined_data.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "e38351b2-e7c3-4cda-a795-43ce46cbe0b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "sentences = dataset['text'].tolist()\n",
    "labels = dataset['sentiment'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "9de97a14-9415-45aa-a645-76dc796e71e0",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_size = int(len(sentences) *0.8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "e8a31059-f2b1-4faa-b455-8e756c457fc8",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_sentences = sentences[0:train_size]\n",
    "train_labels = labels[0:train_size]\n",
    "test_sentences = sentences[train_size:]\n",
    "test_labels = labels[train_size:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "3853c8de-d263-4051-bca3-6c5c58a08511",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_label_fin = np.array(train_labels)\n",
    "test_label_fin = np.array(test_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "36393f36-04eb-4e2c-8bfd-dc15aad2d09e",
   "metadata": {},
   "outputs": [],
   "source": [
    "vocab_size = 500\n",
    "embedding_dim = 16\n",
    "max_length = 50\n",
    "trunc_type = 'post'\n",
    "padding_type = 'post'\n",
    "oov_tok = \"<OOV>\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "4199eabe-247a-4421-826d-bcd1e5ff0ce3",
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer = Tokenizer(num_words = vocab_size, oov_token = oov_tok)\n",
    "tokenizer.fit_on_texts(train_sentences)\n",
    "\n",
    "word_index = tokenizer.word_index\n",
    "train_sequences = tokenizer.texts_to_sequences(train_sentences)\n",
    "train_padding = pad_sequences(train_sequences, maxlen = max_length, padding = padding_type,\n",
    "                              truncating = trunc_type)\n",
    "\n",
    "test_sequences = tokenizer.texts_to_sequences(test_sentences)\n",
    "test_padding = pad_sequences(test_sequences, maxlen = max_length, padding = padding_type,\n",
    "                             truncating = trunc_type)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "762753a3-5c01-4e63-9e60-d73212003f94",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = tf.keras.Sequential([\n",
    "    tf.keras.layers.Embedding(vocab_size, embedding_dim, input_length = max_length),\n",
    "    tf.keras.layers.GlobalAveragePooling1D(),\n",
    "    tf.keras.layers.Dense(6, activation = 'relu'),\n",
    "    tf.keras.layers.Dense(1, activation = 'sigmoid')\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "ddd3a5db-7ecc-46d0-b49e-6c1865fa0ac4",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss = 'binary_crossentropy', optimizer = 'adam', metrics = ['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "96b27bea-0d46-4c1b-b577-e8c8c7750214",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_3\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " embedding_3 (Embedding)     (None, 50, 16)            8000      \n",
      "                                                                 \n",
      " global_average_pooling1d_3  (None, 16)                0         \n",
      "  (GlobalAveragePooling1D)                                       \n",
      "                                                                 \n",
      " dense_6 (Dense)             (None, 6)                 102       \n",
      "                                                                 \n",
      " dense_7 (Dense)             (None, 1)                 7         \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 8109 (31.68 KB)\n",
      "Trainable params: 8109 (31.68 KB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "68940f35-5c0e-484e-a757-4fb0b68d815e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "50/50 [==============================] - 1s 3ms/step - loss: 0.6926 - accuracy: 0.5166 - val_loss: 0.6945 - val_accuracy: 0.4110\n",
      "Epoch 2/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6903 - accuracy: 0.5223 - val_loss: 0.6957 - val_accuracy: 0.4135\n",
      "Epoch 3/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6877 - accuracy: 0.5311 - val_loss: 0.6938 - val_accuracy: 0.4361\n",
      "Epoch 4/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6841 - accuracy: 0.5800 - val_loss: 0.6920 - val_accuracy: 0.4662\n",
      "Epoch 5/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6786 - accuracy: 0.5951 - val_loss: 0.6850 - val_accuracy: 0.5313\n",
      "Epoch 6/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6713 - accuracy: 0.6648 - val_loss: 0.6804 - val_accuracy: 0.5514\n",
      "Epoch 7/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6613 - accuracy: 0.6824 - val_loss: 0.6694 - val_accuracy: 0.6190\n",
      "Epoch 8/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6489 - accuracy: 0.7219 - val_loss: 0.6579 - val_accuracy: 0.6466\n",
      "Epoch 9/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6335 - accuracy: 0.7495 - val_loss: 0.6470 - val_accuracy: 0.6667\n",
      "Epoch 10/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.6132 - accuracy: 0.7608 - val_loss: 0.6199 - val_accuracy: 0.7594\n",
      "Epoch 11/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.5873 - accuracy: 0.7872 - val_loss: 0.6110 - val_accuracy: 0.7343\n",
      "Epoch 12/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.5623 - accuracy: 0.8067 - val_loss: 0.5903 - val_accuracy: 0.7694\n",
      "Epoch 13/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.5339 - accuracy: 0.8230 - val_loss: 0.5795 - val_accuracy: 0.7343\n",
      "Epoch 14/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.5105 - accuracy: 0.8230 - val_loss: 0.5568 - val_accuracy: 0.7719\n",
      "Epoch 15/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.4825 - accuracy: 0.8374 - val_loss: 0.5504 - val_accuracy: 0.7444\n",
      "Epoch 16/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.4584 - accuracy: 0.8500 - val_loss: 0.5298 - val_accuracy: 0.7970\n",
      "Epoch 17/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.4383 - accuracy: 0.8493 - val_loss: 0.5367 - val_accuracy: 0.7318\n",
      "Epoch 18/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.4180 - accuracy: 0.8512 - val_loss: 0.5167 - val_accuracy: 0.7619\n",
      "Epoch 19/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.3989 - accuracy: 0.8625 - val_loss: 0.5019 - val_accuracy: 0.7895\n",
      "Epoch 20/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.3819 - accuracy: 0.8675 - val_loss: 0.5052 - val_accuracy: 0.7519\n",
      "Epoch 21/30\n",
      "50/50 [==============================] - 0s 2ms/step - loss: 0.3667 - accuracy: 0.8682 - val_loss: 0.5130 - val_accuracy: 0.7343\n",
      "Epoch 22/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.3544 - accuracy: 0.8738 - val_loss: 0.4895 - val_accuracy: 0.7820\n",
      "Epoch 23/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.3431 - accuracy: 0.8757 - val_loss: 0.4927 - val_accuracy: 0.7644\n",
      "Epoch 24/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.3303 - accuracy: 0.8788 - val_loss: 0.4889 - val_accuracy: 0.7719\n",
      "Epoch 25/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.3211 - accuracy: 0.8839 - val_loss: 0.5028 - val_accuracy: 0.7293\n",
      "Epoch 26/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.3114 - accuracy: 0.8883 - val_loss: 0.5121 - val_accuracy: 0.7268\n",
      "Epoch 27/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.2999 - accuracy: 0.8889 - val_loss: 0.5140 - val_accuracy: 0.7243\n",
      "Epoch 28/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.2930 - accuracy: 0.8920 - val_loss: 0.4934 - val_accuracy: 0.7444\n",
      "Epoch 29/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.2850 - accuracy: 0.8939 - val_loss: 0.4974 - val_accuracy: 0.7368\n",
      "Epoch 30/30\n",
      "50/50 [==============================] - 0s 1ms/step - loss: 0.2808 - accuracy: 0.8945 - val_loss: 0.5011 - val_accuracy: 0.7343\n"
     ]
    }
   ],
   "source": [
    "num_epochs = 30\n",
    "history = model.fit(train_padding, train_label_fin, epochs = num_epochs, validation_data = (test_padding, test_label_fin))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "d2648dcd-a91d-453c-adc4-ae638798dd07",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_graphs(history, string):\n",
    "    plt.plot(history.history[string])\n",
    "    plt.plot(history.history['val_'+string])\n",
    "    plt.xlabel(\"Epoch\")\n",
    "    plt.ylabel(string)\n",
    "    plt.legend([string, 'val_'+string])\n",
    "    plt.show()\n",
    "\n",
    "plot_graphs(history, 'accuracy')\n",
    "plot_graphs(history, 'loss')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c95fa29-c4cd-4374-9d88-f14edc5d0e02",
   "metadata": {},
   "source": [
    "3 below codings are for vizing them in: https://projector.tensorflow.org/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "82e9629c-af06-4d54-822b-bc1810cb9ada",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(500, 16)\n"
     ]
    }
   ],
   "source": [
    "e = model.layers[0]\n",
    "weights = e.get_weights()[0]\n",
    "print(weights.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "c3e85fd4-09cd-4ecc-b0c4-dbdd0d0995b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import io\n",
    "reverse_word_index = dict([(value, key) for (key, value) in word_index.items()])\n",
    "\n",
    "out_v = io.open('vecs.tsv', 'w', encoding = 'utf-8')\n",
    "out_m = io.open('meta.tsv', 'w', encoding = 'utf-8')\n",
    "for word_num in range(1, vocab_size):\n",
    "    word = reverse_word_index[word_num]\n",
    "    embeddings = weights[word_num]\n",
    "    out_m.write(word + \"\\n\")\n",
    "    out_v.write('\\t'.join([str(x) for x in embeddings]) + '\\n')\n",
    "out_v.close()\n",
    "out_m.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "7b0c9d3d-faeb-45c6-bbe8-4b004bc293c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<a href='vecs.tsv' target='_blank'>vecs.tsv</a><br>"
      ],
      "text/plain": [
       "C:\\Users\\Harivenkat\\my_project\\hvenv\\Data science\\vecs.tsv"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<a href='meta.tsv' target='_blank'>meta.tsv</a><br>"
      ],
      "text/plain": [
       "C:\\Users\\Harivenkat\\my_project\\hvenv\\Data science\\meta.tsv"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython.display import FileLink\n",
    "import os\n",
    "\n",
    "vecs_file = 'vecs.tsv'\n",
    "meta_file = 'meta.tsv'\n",
    "\n",
    "if os.path.exists(vecs_file):\n",
    "    display(FileLink(vecs_file))\n",
    "else:\n",
    "    print(f\"{vecs_file} not found.\")\n",
    "\n",
    "if os.path.exists(meta_file):\n",
    "    display(FileLink(meta_file))\n",
    "else:\n",
    "    print(f\"{meta_file} not found.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d740015-c2c3-42f1-ab9b-c84584940fdb",
   "metadata": {},
   "source": [
    "Prediction............."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "ab767749-594f-4132-9d6b-40fc21810775",
   "metadata": {},
   "outputs": [],
   "source": [
    "fake_reviews = ['I love this phone', 'I hate spaghetti', \n",
    "                'Everything was cold',\n",
    "                'Everything was hot exactly as I wanted', \n",
    "                'Everything was green', \n",
    "                'the host seated us immediately',\n",
    "                'they gave us free chocolate cake', \n",
    "                'not sure about the wilted flowers on the table',\n",
    "                'only works when I stand on tippy toes', \n",
    "                'does not work when I stand on my head']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "02bf5647-c64f-4e97-acc1-2ffb75aa9b1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "padding_type = 'post'\n",
    "sample_sequences = tokenizer.texts_to_sequences(fake_reviews)\n",
    "fake_pad = pad_sequences(sample_sequences, padding = padding_type, maxlen = max_length)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "0313c063-8fd0-45c1-94f9-5d7fc63145d6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Value close to 1 - positive, \n",
      "Value close to 0 - negative, \n",
      "0.5 - ambiguous\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "I love this phone\n",
      "[0.89368695]\n",
      "\n",
      "\n",
      "I hate spaghetti\n",
      "[0.21749732]\n",
      "\n",
      "\n",
      "Everything was cold\n",
      "[0.6119154]\n",
      "\n",
      "\n",
      "Everything was hot exactly as I wanted\n",
      "[0.4538913]\n",
      "\n",
      "\n",
      "Everything was green\n",
      "[0.6119154]\n",
      "\n",
      "\n",
      "the host seated us immediately\n",
      "[0.59690744]\n",
      "\n",
      "\n",
      "they gave us free chocolate cake\n",
      "[0.8283504]\n",
      "\n",
      "\n",
      "not sure about the wilted flowers on the table\n",
      "[0.13515812]\n",
      "\n",
      "\n",
      "only works when I stand on tippy toes\n",
      "[0.81606734]\n",
      "\n",
      "\n",
      "does not work when I stand on my head\n",
      "[0.03357479]\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"Value close to 1 - positive, \\nValue close to 0 - negative, \\n0.5 - ambiguous\")\n",
    "\n",
    "classes = model.predict(fake_pad)\n",
    "\n",
    "for x in range(len(fake_reviews)):\n",
    "    print(fake_reviews[x])\n",
    "    print(classes[x])\n",
    "    print(\"\\n\")"
   ]
  }
 ],
 "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}