ml projects/sentiment_analysis.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "sentiment analysis.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "uBw4cGjujhDl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import tensorflow as tf\n",
        "\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "I4EERDzPjjPA",
        "colab_type": "code",
        "outputId": "c5601fd4-17e3-48e4-d2b7-b331df4edf66",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 332
        }
      },
      "source": [
        "!wget --no-check-certificate \\\n",
        "    -O /tmp/sentiment.csv https://drive.google.com/uc?id=13ySLC_ue6Umt9RJYSeM2t-V0kCv-4C-P"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-05-23 08:04:39--  https://drive.google.com/uc?id=13ySLC_ue6Umt9RJYSeM2t-V0kCv-4C-P\n",
            "Resolving drive.google.com (drive.google.com)... 64.233.189.102, 64.233.189.138, 64.233.189.101, ...\n",
            "Connecting to drive.google.com (drive.google.com)|64.233.189.102|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Moved Temporarily\n",
            "Location: https://doc-08-ak-docs.googleusercontent.com/docs/securesc/ha0ro937gcuc7l7deffksulhg5h7mbp1/q11opdeipcebk8b557q7sdjlfaocmabt/1590221025000/11118900490791463723/*/13ySLC_ue6Umt9RJYSeM2t-V0kCv-4C-P [following]\n",
            "Warning: wildcards not supported in HTTP.\n",
            "--2020-05-23 08:04:39--  https://doc-08-ak-docs.googleusercontent.com/docs/securesc/ha0ro937gcuc7l7deffksulhg5h7mbp1/q11opdeipcebk8b557q7sdjlfaocmabt/1590221025000/11118900490791463723/*/13ySLC_ue6Umt9RJYSeM2t-V0kCv-4C-P\n",
            "Resolving doc-08-ak-docs.googleusercontent.com (doc-08-ak-docs.googleusercontent.com)... 74.125.23.132, 2404:6800:4008:c02::84\n",
            "Connecting to doc-08-ak-docs.googleusercontent.com (doc-08-ak-docs.googleusercontent.com)|74.125.23.132|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 127831 (125K) [text/csv]\n",
            "Saving to: ‘/tmp/sentiment.csv’\n",
            "\n",
            "\r/tmp/sentiment.csv    0%[                    ]       0  --.-KB/s               \r/tmp/sentiment.csv  100%[===================>] 124.83K  --.-KB/s    in 0.001s  \n",
            "\n",
            "2020-05-23 08:04:40 (101 MB/s) - ‘/tmp/sentiment.csv’ saved [127831/127831]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7dHFlM91jkeH",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "dataset = pd.read_csv('/tmp/sentiment.csv')\n",
        "\n",
        "sentences = dataset['text'].tolist()\n",
        "labels = dataset['sentiment'].tolist()\n",
        "\n",
        "# Separate out the sentences and labels into training and test sets\n",
        "training_size = int(len(sentences) * 0.8)\n",
        "\n",
        "training_sentences = sentences[0:training_size]\n",
        "testing_sentences = sentences[training_size:]\n",
        "training_labels = labels[0:training_size]\n",
        "testing_labels = labels[training_size:]\n",
        "\n",
        "# Make labels into numpy arrays for use with the network later\n",
        "training_labels_final = np.array(training_labels)\n",
        "testing_labels_final = np.array(testing_labels)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Wsjg_ccmjl2-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "vocab_size = 1000\n",
        "embedding_dim = 16\n",
        "max_length = 100\n",
        "trunc_type='post'\n",
        "padding_type='post'\n",
        "oov_tok = \"<OOV>\"\n",
        "\n",
        "\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "tokenizer = Tokenizer(num_words = vocab_size, oov_token=oov_tok)\n",
        "tokenizer.fit_on_texts(training_sentences)\n",
        "word_index = tokenizer.word_index\n",
        "sequences = tokenizer.texts_to_sequences(training_sentences)\n",
        "padded = pad_sequences(sequences,maxlen=max_length, padding=padding_type, \n",
        "                       truncating=trunc_type)\n",
        "\n",
        "testing_sequences = tokenizer.texts_to_sequences(testing_sentences)\n",
        "testing_padded = pad_sequences(testing_sequences,maxlen=max_length, \n",
        "                               padding=padding_type, truncating=trunc_type)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wzfeu_8DkB8v",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        },
        "outputId": "b28cff4a-5a10-4194-9019-bd7c4a836045"
      },
      "source": [
        "reverse_word_index = dict([(value, key) for (key, value) in word_index.items()])\n",
        "\n",
        "def decode_review(text):\n",
        "    return ' '.join([reverse_word_index.get(i, '0') for i in text])\n",
        "\n",
        "print(decode_review(padded[1]))\n",
        "print(training_sentences[1])"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "good case excellent value 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
            "Good case Excellent value.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_MLToltzkGcK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "3a8721f1-f7f9-4063-c007-a1291b68f1eb"
      },
      "source": [
        "model = tf.keras.Sequential([\n",
        "    tf.keras.layers.Embedding(vocab_size, embedding_dim, input_length=max_length),\n",
        "    tf.keras.layers.Flatten(),\n",
        "    tf.keras.layers.Dense(6, activation='relu'),\n",
        "    tf.keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])\n",
        "model.summary()"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_1 (Embedding)      (None, 100, 16)           16000     \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 1600)              0         \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 6)                 9606      \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 1)                 7         \n",
            "=================================================================\n",
            "Total params: 25,613\n",
            "Trainable params: 25,613\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TCfyHqyMkN6k",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 225
        },
        "outputId": "56cc5548-76bc-4d6a-8130-df8fe07658a6"
      },
      "source": [
        "num_epochs = 6\n",
        "history = model.fit(padded, training_labels_final, epochs=num_epochs, validation_data=(testing_padded, testing_labels_final))"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/6\n",
            "50/50 [==============================] - 0s 5ms/step - loss: 0.0032 - accuracy: 0.9994 - val_loss: 1.0006 - val_accuracy: 0.7494\n",
            "Epoch 2/6\n",
            "50/50 [==============================] - 0s 5ms/step - loss: 0.0032 - accuracy: 0.9994 - val_loss: 0.9697 - val_accuracy: 0.7494\n",
            "Epoch 3/6\n",
            "50/50 [==============================] - 0s 6ms/step - loss: 0.0031 - accuracy: 0.9994 - val_loss: 1.0033 - val_accuracy: 0.7444\n",
            "Epoch 4/6\n",
            "50/50 [==============================] - 0s 5ms/step - loss: 0.0028 - accuracy: 0.9994 - val_loss: 1.0469 - val_accuracy: 0.7469\n",
            "Epoch 5/6\n",
            "50/50 [==============================] - 0s 6ms/step - loss: 0.0025 - accuracy: 1.0000 - val_loss: 1.0117 - val_accuracy: 0.7444\n",
            "Epoch 6/6\n",
            "50/50 [==============================] - 0s 6ms/step - loss: 0.0024 - accuracy: 1.0000 - val_loss: 1.0535 - val_accuracy: 0.7469\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TQymGL4jkSJ1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "# def plot_graphs(history, string):\n",
        "#   plt.plot(history.history[string])\n",
        "#   plt.plot(history.history['val_'+string])\n",
        "#   plt.xlabel(\"Epochs\")\n",
        "#   plt.ylabel(string)\n",
        "#   plt.legend([string, 'val_'+string])\n",
        "#   plt.show()\n",
        "  \n",
        "# plot_graphs(history, \"accuracy\")\n",
        "# plot_graphs(history, \"loss\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yhTAjNo1khof",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 973
        },
        "outputId": "963a4072-e759-4b42-ae70-161e5766a28b"
      },
      "source": [
        "# Use the model to predict a review   \n",
        "some_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']\n",
        "\n",
        "print(some_reviews) \n",
        "\n",
        "# Create the sequences\n",
        "padding_type='post'\n",
        "sample_sequences = tokenizer.texts_to_sequences(some_reviews)\n",
        "sample_padded = pad_sequences(sample_sequences, padding=padding_type, maxlen=max_length)           \n",
        "\n",
        "print('\\nLets try implementing the model\\n')              \n",
        "\n",
        "classes = model.predict(sample_padded)\n",
        "\n",
        "# The closer the class is to 1, the more positive the review is deemed to be\n",
        "for x in range(len(some_reviews)):\n",
        "  print(fake_reviews[x])\n",
        "  i = np.round(classes[x],15)\n",
        "  print(i)\n",
        "  if(i<0.2):\n",
        "    print('It is a very bad responce')\n",
        "  elif(i<0.4):\n",
        "    print('It is not  a bad responce')\n",
        "  elif(i<0.6):\n",
        "    print('It is an everage responce')\n",
        "  elif(i<0.8):\n",
        "    print('It is a good responce')\n",
        "  else:\n",
        "    print('It is an excellent responce')\n",
        "  print('\\n')\n",
        "\n",
        "# Try adding reviews of your own\n",
        "# Add some negative words (such as \"not\") to the good reviews and see what happens\n",
        "# For example:\n",
        "# they gave us free chocolate cake and did not charge us"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['I love this phone', 'I hate spaghetti', 'Everything was cold', 'Everything was hot exactly as I wanted', 'Everything was green', 'the host seated us immediately', 'they gave us free chocolate cake', 'not sure about the wilted flowers on the table', 'only works when I stand on tippy toes', 'does not work when I stand on my head']\n",
            "\n",
            "Lets try implementing the model\n",
            "\n",
            "I love this phone\n",
            "[0.99999404]\n",
            "It is an excellent responce\n",
            "\n",
            "\n",
            "I hate spaghetti\n",
            "[3.0542866e-05]\n",
            "It is a very bad responce\n",
            "\n",
            "\n",
            "Everything was cold\n",
            "[0.00084935]\n",
            "It is a very bad responce\n",
            "\n",
            "\n",
            "Everything was hot exactly as I wanted\n",
            "[0.02002569]\n",
            "It is a very bad responce\n",
            "\n",
            "\n",
            "Everything was green\n",
            "[0.01272446]\n",
            "It is a very bad responce\n",
            "\n",
            "\n",
            "the host seated us immediately\n",
            "[0.9999609]\n",
            "It is an excellent responce\n",
            "\n",
            "\n",
            "they gave us free chocolate cake\n",
            "[0.9999843]\n",
            "It is an excellent responce\n",
            "\n",
            "\n",
            "not sure about the wilted flowers on the table\n",
            "[6.750078e-05]\n",
            "It is a very bad responce\n",
            "\n",
            "\n",
            "only works when I stand on tippy toes\n",
            "[0.9999628]\n",
            "It is an excellent responce\n",
            "\n",
            "\n",
            "does not work when I stand on my head\n",
            "[2.766072e-06]\n",
            "It is a very bad responce\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DHoflBuvm09K",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}