SuicideIdeation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "SuicideIdeation",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyM0nToO6xe7H5Zq/EWjI3jh",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/shvms/Suicide_pred/blob/master/SuicideIdeation.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_cfIe1h6VcU3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import re\n",
        "import json\n",
        "import csv\n",
        "import numpy\n",
        "from matplotlib import pyplot as plt\n",
        "import numpy as np"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gYnzHP2R_VC4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# helper function\n",
        "# to condense three or more repetitive letters into one\n",
        "def condense(text, rep_size=3):\n",
        "  \"\"\"\n",
        "  text: string data to filter\n",
        "  rep_size: >= repetitive letters\n",
        "  \"\"\"\n",
        "  return re.sub('(.)\\\\1{'+ str(rep_size-1) +',}', r'\\1', text)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6SSdp44anJH-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# helper function to plot graphs\n",
        "def plot_graphs(history, metric):\n",
        "  plt.plot(history.history[f'{metric}'])\n",
        "  plt.plot(history.history[f'val_{metric}'])\n",
        "  plt.xlabel('#Epochs')\n",
        "  plt.ylabel(metric)\n",
        "  plt.legend([metric, f'val_{metric}'])\n",
        "  plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EnV8-ki6nwuK",
        "colab_type": "code",
        "outputId": "74ec5255-6046-4aaa-c327-bcc2dd966912",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 760
        }
      },
      "source": [
        "!pip install tensorflow===1.15.0"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting tensorflow===1.15.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/3f/98/5a99af92fb911d7a88a0005ad55005f35b4c1ba8d75fba02df726cd936e6/tensorflow-1.15.0-cp36-cp36m-manylinux2010_x86_64.whl (412.3MB)\n",
            "\u001b[K     |████████████████████████████████| 412.3MB 27kB/s \n",
            "\u001b[?25hRequirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (1.1.0)\n",
            "Requirement already satisfied: keras-preprocessing>=1.0.5 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (1.1.0)\n",
            "Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (1.12.0)\n",
            "Requirement already satisfied: wrapt>=1.11.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (1.11.2)\n",
            "Requirement already satisfied: grpcio>=1.8.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (1.15.0)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (3.1.0)\n",
            "Requirement already satisfied: numpy<2.0,>=1.16.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (1.17.5)\n",
            "Requirement already satisfied: keras-applications>=1.0.8 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (1.0.8)\n",
            "Requirement already satisfied: absl-py>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (0.9.0)\n",
            "Requirement already satisfied: google-pasta>=0.1.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (0.1.8)\n",
            "Collecting tensorflow-estimator==1.15.1\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/de/62/2ee9cd74c9fa2fa450877847ba560b260f5d0fb70ee0595203082dafcc9d/tensorflow_estimator-1.15.1-py2.py3-none-any.whl (503kB)\n",
            "\u001b[K     |████████████████████████████████| 512kB 45.3MB/s \n",
            "\u001b[?25hCollecting tensorboard<1.16.0,>=1.15.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/1e/e9/d3d747a97f7188f48aa5eda486907f3b345cd409f0a0850468ba867db246/tensorboard-1.15.0-py3-none-any.whl (3.8MB)\n",
            "\u001b[K     |████████████████████████████████| 3.8MB 50.5MB/s \n",
            "\u001b[?25hRequirement already satisfied: wheel>=0.26 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (0.34.2)\n",
            "Requirement already satisfied: protobuf>=3.6.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (3.10.0)\n",
            "Requirement already satisfied: astor>=0.6.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (0.8.1)\n",
            "Requirement already satisfied: gast==0.2.2 in /usr/local/lib/python3.6/dist-packages (from tensorflow===1.15.0) (0.2.2)\n",
            "Requirement already satisfied: h5py in /usr/local/lib/python3.6/dist-packages (from keras-applications>=1.0.8->tensorflow===1.15.0) (2.8.0)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.6/dist-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow===1.15.0) (3.1.1)\n",
            "Requirement already satisfied: setuptools>=41.0.0 in /usr/local/lib/python3.6/dist-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow===1.15.0) (45.1.0)\n",
            "Requirement already satisfied: werkzeug>=0.11.15 in /usr/local/lib/python3.6/dist-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow===1.15.0) (0.16.1)\n",
            "Installing collected packages: tensorflow-estimator, tensorboard, tensorflow\n",
            "  Found existing installation: tensorflow-estimator 2.0.1\n",
            "    Uninstalling tensorflow-estimator-2.0.1:\n",
            "      Successfully uninstalled tensorflow-estimator-2.0.1\n",
            "  Found existing installation: tensorboard 2.0.2\n",
            "    Uninstalling tensorboard-2.0.2:\n",
            "      Successfully uninstalled tensorboard-2.0.2\n",
            "  Found existing installation: tensorflow 2.0.0\n",
            "    Uninstalling tensorflow-2.0.0:\n",
            "      Successfully uninstalled tensorflow-2.0.0\n",
            "Successfully installed tensorboard-1.15.0 tensorflow-1.15.0 tensorflow-estimator-1.15.1\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "tensorboard",
                  "tensorflow",
                  "tensorflow_core",
                  "tensorflow_estimator"
                ]
              }
            }
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y8JRDZz6XYfC",
        "colab_type": "code",
        "outputId": "af8486a3-3b67-4c98-9947-217c7c8b0ad6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 79
        }
      },
      "source": [
        "from sklearn.metrics import classification_report\n",
        "import tensorflow as tf\n",
        "import keras\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<p style=\"color: red;\">\n",
              "The default version of TensorFlow in Colab will soon switch to TensorFlow 2.x.<br>\n",
              "We recommend you <a href=\"https://www.tensorflow.org/guide/migrate\" target=\"_blank\">upgrade</a> now \n",
              "or ensure your notebook will continue to use TensorFlow 1.x via the <code>%tensorflow_version 1.x</code> magic:\n",
              "<a href=\"https://colab.research.google.com/notebooks/tensorflow_version.ipynb\" target=\"_blank\">more info</a>.</p>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pJnnndEqnqCw",
        "colab_type": "code",
        "outputId": "1c64fc37-5c75-44c3-82ab-5dbceb2621b4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "print(tf.__version__)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1.15.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bZboKhZ5lUpu",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "tf.compat.v1.disable_eager_execution()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "REyoxZ8BaD1j",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "with open('/tmp/stopwords.txt', 'r', encoding='utf-8') as file:\n",
        "  stopwords = file.readlines()\n",
        "stopwords = [line.strip() for line in stopwords]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SWUfWfQqarym",
        "colab_type": "code",
        "outputId": "b5eba532-5606-48a9-8f30-c7432cbb88c5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        }
      },
      "source": [
        "print(f\"Total stopwords: {len(stopwords)}\")\n",
        "print(stopwords[:20])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Total stopwords: 823\n",
            "['x', 'y', 'your', 'yours', 'yourself', 'yourselves', 'you', 'yond', 'yonder', 'yon', 'ye', 'yet', 'z', 'zillion', 'j', 'u', 'umpteen', 'usually', 'us', 'username']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Tb4A9lmAayfz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# hyperparameters\n",
        "vocab_size = 1000\n",
        "embedding_dim = 300\n",
        "max_length = 150\n",
        "trunc_type = 'post'\n",
        "padding_type = 'post'\n",
        "oov_tok = '<oov>'\n",
        "training_portion = .8"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OcYApDQvbiWj",
        "colab_type": "code",
        "outputId": "df703b4a-fbd3-4352-d731-65799280752b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 67
        }
      },
      "source": [
        "sentences, labels = list(), list()\n",
        "with open('/tmp/finalcsv.csv', 'r', encoding='utf-8') as csvfile:\n",
        "  reader = csv.DictReader(csvfile)\n",
        "  for row in reader:\n",
        "    sentences.append(row['posts/post'])\n",
        "    labels.append(row['posts/intent'])\n",
        "\n",
        "labels = list(map(int, labels))\n",
        "\n",
        "print(f\"Total dataset: {len(sentences)}\")\n",
        "print(f\"Total suicidal intent posts: {sum(labels)}\")\n",
        "print(f\"Total non-suicidal intent posts: {len(labels)-sum(labels)}\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Total dataset: 695\n",
            "Total suicidal intent posts: 341\n",
            "Total non-suicidal intent posts: 354\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ik5yWfG5b1uw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_sentences, train_labels = list(), list()\n",
        "test_sentences, test_labels = list(), list()\n",
        "\n",
        "train_size = int(len(sentences) * training_portion)\n",
        "\n",
        "train_sentences = sentences[:train_size]\n",
        "train_labels = labels[:train_size]\n",
        "\n",
        "test_sentences = sentences[train_size:]\n",
        "test_labels = labels[train_size:]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_N7pQwDthjXV",
        "colab_type": "text"
      },
      "source": [
        "# Remove Repeated Characters\n",
        "Characters repeated thrice or more can be removed, like *dieeeee* becomes *die*."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EnTRJbLGh7WF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_sentences = [condense(sentence) for sentence in train_sentences]\n",
        "test_sentences = [condense(sentence) for sentence in test_sentences]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OD6FCjHhDS2x",
        "colab_type": "text"
      },
      "source": [
        "# Final Text Preprocessing\n",
        "### Removing special characters & digits along with mixed words"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bWZldU_lDU03",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "pat = r'([A-Za-z]+[\\d@]+[\\w@]*|[\\d@]+[A-Za-z]+[\\w@]*)'\n",
        "train_sentences = [re.sub(pat, '', sentence) for sentence in train_sentences]             # removing words like he12eoo00\n",
        "train_sentences = [re.sub(r'<!--?.*?-->', '', sentence) for sentence in train_sentences]  # removing html tags if available\n",
        "train_sentences = [re.sub(r'[\\W|\\d]+', ' ', sentence) for sentence in train_sentences]   # removing "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4v5FLABsidLa",
        "colab_type": "text"
      },
      "source": [
        "# Preprocessing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GPNLeJAViZAj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "tokenizer = Tokenizer(num_words=vocab_size, oov_token=oov_tok)\n",
        "\n",
        "tokenizer.fit_on_texts(train_sentences)\n",
        "word_index = tokenizer.word_index\n",
        "\n",
        "train_sequences = tokenizer.texts_to_sequences(train_sentences)\n",
        "train_padded_sequences = pad_sequences(train_sequences, maxlen=max_length,\n",
        "  padding=padding_type, truncating=trunc_type)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "blgjG3iU2ou3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "with open('/tmp/word_index.json', 'w') as file:\n",
        "  json.dump(word_index, file)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1-dtfFvB3H9L",
        "colab_type": "code",
        "outputId": "42156ec9-4d63-4147-c38e-f78da8179361",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        }
      },
      "source": [
        "print(train_sequences[0])\n",
        "print(len(train_sequences[0]))\n",
        "print(train_padded_sequences[0])\n",
        "print(len(train_padded_sequences[0]))\n",
        "print(type(train_padded_sequences[0]))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[2, 184, 1, 11, 452, 36, 76, 281, 68, 36, 19, 619, 2, 24, 197, 2, 184, 2, 24, 197, 7, 208, 50, 2, 24, 668, 19, 72, 23, 7, 1, 1, 60, 6, 1, 2, 54, 119, 46, 7, 342, 9, 7, 1, 17, 8, 75, 85, 46, 19, 83, 694, 240, 7, 91, 2, 24, 343, 2, 184, 64, 3, 299, 1, 13, 5, 832, 6, 93, 59, 883, 17, 2, 77, 52, 478, 1, 350, 11, 49, 2, 24, 78, 145, 35, 1, 8, 45]\n",
            "88\n",
            "[  2 184   1  11 452  36  76 281  68  36  19 619   2  24 197   2 184   2\n",
            "  24 197   7 208  50   2  24 668  19  72  23   7   1   1  60   6   1   2\n",
            "  54 119  46   7 342   9   7   1  17   8  75  85  46  19  83 694 240   7\n",
            "  91   2  24 343   2 184  64   3 299   1  13   5 832   6  93  59 883  17\n",
            "   2  77  52 478   1 350  11  49   2  24  78 145  35   1   8  45   0   0\n",
            "   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
            "   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
            "   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
            "   0   0   0   0   0   0]\n",
            "150\n",
            "<class 'numpy.ndarray'>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nVTeCu9x_okT",
        "colab_type": "code",
        "outputId": "74c96a77-2906-436d-8d9a-97f2f337866e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "train_sentences[0]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'i cry everydayIs that normal Do more depressed people do this Whenever I m alone I cry I m alone a lot As i m writing this there s a tear rolling from my cheek I ve always been a bit of a crybaby but it has never been this much Multiple times a day I m crying I cry myself to sleep everynight In the beginning my family would notice but I think they stopped bothering Either that or I m really good at covering it up '"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "81A2nCcbjyvr",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "test_sequences = tokenizer.texts_to_sequences(test_sentences)\n",
        "test_padded_sequences = pad_sequences(test_sequences, maxlen=max_length,\n",
        "  padding=padding_type, truncating=trunc_type)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kudFtXDZkTOG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_labels = np.array(train_labels)\n",
        "test_labels = np.array(test_labels)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ODRL8ZyAklGR",
        "colab_type": "text"
      },
      "source": [
        "# 1. Word2Vec Only Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z2tnsg7Dlzgn",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# hyperparameters\n",
        "embedding_dim_1 = 300\n",
        "dense1 = 24\n",
        "dense2 = 1"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BtZ_ntiRlMKb",
        "colab_type": "code",
        "outputId": "d6c65188-b05f-4520-9820-294d40aa46ce",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        }
      },
      "source": [
        "# setting up\n",
        "model1 = keras.models.Sequential([\n",
        "  keras.layers.Embedding(vocab_size, embedding_dim_1, input_length=max_length),\n",
        "  keras.layers.GlobalAveragePooling1D(),\n",
        "  keras.layers.Dense(dense1, activation='relu'),\n",
        "  keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "model1.compile(loss='binary_crossentropy', optimizer='rmsprop', metrics=['accuracy'])\n",
        "print(model1.summary())"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:66: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:541: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4432: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:793: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3657: The name tf.log is deprecated. Please use tf.math.log instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/nn_impl.py:183: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_1 (Embedding)      (None, 120, 300)          300000    \n",
            "_________________________________________________________________\n",
            "global_average_pooling1d_1 ( (None, 300)               0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 24)                7224      \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 1)                 25        \n",
            "=================================================================\n",
            "Total params: 307,249\n",
            "Trainable params: 307,249\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1F3Qz8BVmlbM",
        "colab_type": "code",
        "outputId": "cb253e18-8d0f-413c-d00d-eb27606633d8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 658
        }
      },
      "source": [
        "# training\n",
        "num_epochs = 10\n",
        "history1 = model1.fit(train_padded_sequences, train_labels, epochs=num_epochs,\n",
        "                    validation_data=(test_padded_sequences, test_labels))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1033: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1020: The name tf.assign is deprecated. Please use tf.compat.v1.assign instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3005: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.\n",
            "\n",
            "Train on 556 samples, validate on 139 samples\n",
            "Epoch 1/10\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:190: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:197: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:207: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:216: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:223: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.\n",
            "\n",
            "556/556 [==============================] - 4s 6ms/step - loss: 0.6752 - acc: 0.6385 - val_loss: 0.6763 - val_acc: 0.6763\n",
            "Epoch 2/10\n",
            "556/556 [==============================] - 0s 138us/step - loss: 0.6335 - acc: 0.7446 - val_loss: 0.6514 - val_acc: 0.6978\n",
            "Epoch 3/10\n",
            "556/556 [==============================] - 0s 143us/step - loss: 0.5866 - acc: 0.7716 - val_loss: 0.6147 - val_acc: 0.7410\n",
            "Epoch 4/10\n",
            "556/556 [==============================] - 0s 140us/step - loss: 0.5412 - acc: 0.7788 - val_loss: 0.5924 - val_acc: 0.7266\n",
            "Epoch 5/10\n",
            "556/556 [==============================] - 0s 147us/step - loss: 0.4950 - acc: 0.8129 - val_loss: 0.5453 - val_acc: 0.7554\n",
            "Epoch 6/10\n",
            "556/556 [==============================] - 0s 142us/step - loss: 0.4533 - acc: 0.8327 - val_loss: 0.5348 - val_acc: 0.7194\n",
            "Epoch 7/10\n",
            "556/556 [==============================] - 0s 129us/step - loss: 0.4174 - acc: 0.8507 - val_loss: 0.5113 - val_acc: 0.7410\n",
            "Epoch 8/10\n",
            "556/556 [==============================] - 0s 163us/step - loss: 0.3866 - acc: 0.8579 - val_loss: 0.5136 - val_acc: 0.7266\n",
            "Epoch 9/10\n",
            "556/556 [==============================] - 0s 144us/step - loss: 0.3580 - acc: 0.8705 - val_loss: 0.4794 - val_acc: 0.7410\n",
            "Epoch 10/10\n",
            "556/556 [==============================] - 0s 124us/step - loss: 0.3308 - acc: 0.8867 - val_loss: 0.5286 - val_acc: 0.7194\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L7iGfF8JnBVG",
        "colab_type": "code",
        "outputId": "8aad8146-33d9-4217-fa7f-be08fc44f54a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        }
      },
      "source": [
        "# Metrics analysis\n",
        "plot_graphs(history1, 'acc')\n",
        "plot_graphs(history1, 'loss')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEGCAYAAAB/+QKOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEGCAYAAAB/+QKOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njMmQBYQxxpgM/T/09Fmpr21uiwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "d2YsunlEsRkx",
        "colab_type": "text"
      },
      "source": [
        "# 2. LSTM"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4tGYUFxEskRh",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# hyperparametes\n",
        "embedding_dim_2 = 64\n",
        "memory_units = 64\n",
        "dense_units = 32\n",
        "dropout_rate = 0.2"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GWid8EzZr5LO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class clbk(keras.callbacks.Callback):\n",
        "  def on_epoch_end(self, epoch, logs={}):\n",
        "    if (logs.get('val_acc') >= 0.8):\n",
        "      print(\"Achieved.\")\n",
        "      self.model.stop_training=True"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aY-o5qJDsZnQ",
        "colab_type": "code",
        "outputId": "0946d5d7-b7ad-4658-e9e5-14c51cfd8853",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 336
        }
      },
      "source": [
        "model2 = keras.Sequential([\n",
        "  keras.layers.Embedding(vocab_size, embedding_dim_2, input_length=max_length),\n",
        "  keras.layers.LSTM(memory_units),\n",
        "  keras.layers.Dense(dense_units, activation='relu'),\n",
        "  keras.layers.Dropout(dropout_rate),\n",
        "  keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "model2.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "print(model2.summary())"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_2\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_2 (Embedding)      (None, 150, 64)           64000     \n",
            "_________________________________________________________________\n",
            "lstm_2 (LSTM)                (None, 64)                33024     \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 32)                2080      \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 32)                0         \n",
            "_________________________________________________________________\n",
            "dense_4 (Dense)              (None, 1)                 33        \n",
            "=================================================================\n",
            "Total params: 99,137\n",
            "Trainable params: 99,137\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "giuzW11Mx0H0",
        "colab_type": "code",
        "outputId": "b80994d1-abcd-4068-ffb3-0ca62c908370",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 319
        }
      },
      "source": [
        "# training\n",
        "num_epochs = 30\n",
        "history2 = model2.fit(train_padded_sequences, train_labels, epochs=num_epochs,\n",
        "                    validation_data=(test_padded_sequences, test_labels), batch_size=64, callbacks=[clbk()])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 556 samples, validate on 139 samples\n",
            "Epoch 1/30\n",
            "556/556 [==============================] - 3s 6ms/step - loss: 0.6905 - acc: 0.5162 - val_loss: 0.6943 - val_acc: 0.4748\n",
            "Epoch 2/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6816 - acc: 0.6277 - val_loss: 0.6991 - val_acc: 0.5468\n",
            "Epoch 3/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6696 - acc: 0.6331 - val_loss: 0.7079 - val_acc: 0.5683\n",
            "Epoch 4/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6531 - acc: 0.6655 - val_loss: 0.6904 - val_acc: 0.5683\n",
            "Epoch 5/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.5960 - acc: 0.7014 - val_loss: 0.6616 - val_acc: 0.5971\n",
            "Epoch 6/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.5707 - acc: 0.7464 - val_loss: 0.6655 - val_acc: 0.6259\n",
            "Epoch 7/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.5158 - acc: 0.7500 - val_loss: 0.6095 - val_acc: 0.6475\n",
            "Epoch 8/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.4454 - acc: 0.8112 - val_loss: 0.4803 - val_acc: 0.8058\n",
            "Achieved.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ou8j9CFL1FRn",
        "colab_type": "code",
        "outputId": "cf94e78f-ea80-418f-ba10-227419c7e56a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        }
      },
      "source": [
        "# Metrics analysis\n",
        "plot_graphs(history2, 'acc')\n",
        "plot_graphs(history2, 'loss')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEGCAYAAAB/+QKOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJwgzMyvlBGFmZqWcIMzMrJQThJmZlfo/2/6gZS6ngmoAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEGCAYAAABo25JHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qNi9s5D-zpYO",
        "colab_type": "text"
      },
      "source": [
        "# Saving model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dpjekhtwzon3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import pickle"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XMpbR364zuUp",
        "colab_type": "code",
        "outputId": "9d1a0cad-5c44-42d4-ff59-a645a68b842f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "with open('lstm.model.json', 'w') as json_file:\n",
        "  json_file.write(model2.to_json())\n",
        "\n",
        "model2.save_weights('model.lstm.h5')\n",
        "print('Saved model to disk')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Saved model to disk\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "im-q-jehz3np",
        "colab_type": "code",
        "outputId": "665dea99-9f21-41de-bfb3-3435b3d57e73",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 340
        }
      },
      "source": [
        "converter = tf.lite.TFLiteConverter.from_keras_model_file('/tmp/model.lstm.h5')\n",
        "tfmodel = converter.convert()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-128-ec7cfe2c9ffa>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mconverter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlite\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTFLiteConverter\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_keras_model_file\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'/tmp/model.lstm.h5'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mtfmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mconverter\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconvert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow_core/lite/python/lite.py\u001b[0m in \u001b[0;36mfrom_keras_model_file\u001b[0;34m(cls, model_file, input_arrays, input_shapes, output_arrays, custom_objects)\u001b[0m\n\u001b[1;32m    818\u001b[0m     \u001b[0m_keras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclear_session\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    819\u001b[0m     \u001b[0m_keras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_learning_phase\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 820\u001b[0;31m     \u001b[0mkeras_model\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_keras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodels\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_file\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcustom_objects\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    821\u001b[0m     \u001b[0msess\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_keras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_session\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    822\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/keras/saving/save.py\u001b[0m in \u001b[0;36mload_model\u001b[0;34m(filepath, custom_objects, compile)\u001b[0m\n\u001b[1;32m    141\u001b[0m   if (h5py is not None and (\n\u001b[1;32m    142\u001b[0m       isinstance(filepath, h5py.File) or h5py.is_hdf5(filepath))):\n\u001b[0;32m--> 143\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mhdf5_format\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_model_from_hdf5\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcustom_objects\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcompile\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    144\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    145\u001b[0m   \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msix\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstring_types\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/keras/saving/hdf5_format.py\u001b[0m in \u001b[0;36mload_model_from_hdf5\u001b[0;34m(filepath, custom_objects, compile)\u001b[0m\n\u001b[1;32m    157\u001b[0m     \u001b[0mmodel_config\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mattrs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'model_config'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    158\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mmodel_config\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 159\u001b[0;31m       \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'No model found in config file.'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    160\u001b[0m     \u001b[0mmodel_config\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_config\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'utf-8'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    161\u001b[0m     model = model_config_lib.model_from_config(model_config,\n",
            "\u001b[0;31mValueError\u001b[0m: No model found in config file."
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qi_xFS7D2cZX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DwOr-nhM1qCv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "o0GIibwr1Pbs",
        "colab_type": "text"
      },
      "source": [
        "# C-LSTM"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7pZFCLzg1S8X",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# hyperparameters\n",
        "memory_units = 128\n",
        "dense_units = 32\n",
        "dropout_rate = 0.2"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WlEIA-vm1hEL",
        "colab_type": "code",
        "outputId": "582270cb-c47d-4048-ee33-56ac023c13ac",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        }
      },
      "source": [
        "model3 = keras.models.Sequential([\n",
        "  keras.layers.Conv1D(64, (5), activation='relu', input_shape=(max_length, 1)),\n",
        "  keras.layers.MaxPooling1D(2),\n",
        "  keras.layers.LSTM(int(memory_units/2)),\n",
        "  keras.layers.Dropout(0.2),\n",
        "  keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "model3.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "print(model3.summary())"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4267: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.\n",
            "\n",
            "Model: \"sequential_7\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv1d_1 (Conv1D)            (None, 116, 64)           384       \n",
            "_________________________________________________________________\n",
            "max_pooling1d_1 (MaxPooling1 (None, 58, 64)            0         \n",
            "_________________________________________________________________\n",
            "lstm_6 (LSTM)                (None, 32)                12416     \n",
            "_________________________________________________________________\n",
            "dropout_6 (Dropout)          (None, 32)                0         \n",
            "_________________________________________________________________\n",
            "dense_13 (Dense)             (None, 1)                 33        \n",
            "=================================================================\n",
            "Total params: 12,833\n",
            "Trainable params: 12,833\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "655t2VznSDxJ",
        "colab_type": "code",
        "outputId": "9c059501-e510-4e16-8019-c4f4ac08c74f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        }
      },
      "source": [
        "print(train_padded_sequences.shape)\n",
        "print(test_padded_sequences.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(556, 120)\n",
            "(139, 120)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fOimNRCdNpK4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_padded_sequences_2 = train_padded_sequences.reshape(556, 120, 1)\n",
        "test_padded_sequences_2 = test_padded_sequences.reshape(139, 120, 1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V7GOCs2F4kBQ",
        "colab_type": "code",
        "outputId": "08ecc6c9-98f8-4b26-dfbc-ef125e54999f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# training\n",
        "num_epochs = 30\n",
        "history3 = model3.fit(train_padded_sequences_2, train_labels, epochs=num_epochs,\n",
        "                    validation_data=(test_padded_sequences_2, test_labels), batch_size=64, callbacks=[clbk()])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 556 samples, validate on 139 samples\n",
            "Epoch 1/30\n",
            "556/556 [==============================] - 8s 15ms/step - loss: 0.7487 - acc: 0.5719 - val_loss: 0.7383 - val_acc: 0.5180\n",
            "Epoch 2/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6817 - acc: 0.6169 - val_loss: 0.7376 - val_acc: 0.4964\n",
            "Epoch 3/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6783 - acc: 0.6223 - val_loss: 0.7239 - val_acc: 0.4820\n",
            "Epoch 4/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6660 - acc: 0.6115 - val_loss: 0.7161 - val_acc: 0.4676\n",
            "Epoch 5/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6417 - acc: 0.6277 - val_loss: 0.7199 - val_acc: 0.4748\n",
            "Epoch 6/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6464 - acc: 0.6313 - val_loss: 0.7158 - val_acc: 0.5252\n",
            "Epoch 7/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6358 - acc: 0.6295 - val_loss: 0.7302 - val_acc: 0.5180\n",
            "Epoch 8/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6394 - acc: 0.6349 - val_loss: 0.7314 - val_acc: 0.5324\n",
            "Epoch 9/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6266 - acc: 0.6529 - val_loss: 0.7024 - val_acc: 0.5540\n",
            "Epoch 10/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6280 - acc: 0.6367 - val_loss: 0.7110 - val_acc: 0.5324\n",
            "Epoch 11/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6286 - acc: 0.6493 - val_loss: 0.7220 - val_acc: 0.5324\n",
            "Epoch 12/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6108 - acc: 0.6691 - val_loss: 0.7083 - val_acc: 0.5180\n",
            "Epoch 13/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6193 - acc: 0.6385 - val_loss: 0.7230 - val_acc: 0.5036\n",
            "Epoch 14/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6028 - acc: 0.6853 - val_loss: 0.7364 - val_acc: 0.5396\n",
            "Epoch 15/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.6021 - acc: 0.7014 - val_loss: 0.7101 - val_acc: 0.5180\n",
            "Epoch 16/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5939 - acc: 0.7014 - val_loss: 0.7329 - val_acc: 0.5468\n",
            "Epoch 17/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5981 - acc: 0.6637 - val_loss: 0.7359 - val_acc: 0.4892\n",
            "Epoch 18/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5738 - acc: 0.6888 - val_loss: 0.7530 - val_acc: 0.4892\n",
            "Epoch 19/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5874 - acc: 0.6978 - val_loss: 0.8004 - val_acc: 0.5324\n",
            "Epoch 20/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5764 - acc: 0.6906 - val_loss: 0.7681 - val_acc: 0.5396\n",
            "Epoch 21/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5849 - acc: 0.6924 - val_loss: 0.7480 - val_acc: 0.5108\n",
            "Epoch 22/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5953 - acc: 0.6835 - val_loss: 0.7083 - val_acc: 0.5108\n",
            "Epoch 23/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5884 - acc: 0.6817 - val_loss: 0.7169 - val_acc: 0.5468\n",
            "Epoch 24/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5711 - acc: 0.6888 - val_loss: 0.7262 - val_acc: 0.5324\n",
            "Epoch 25/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5663 - acc: 0.6906 - val_loss: 0.7721 - val_acc: 0.4964\n",
            "Epoch 26/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5584 - acc: 0.7140 - val_loss: 0.6964 - val_acc: 0.5540\n",
            "Epoch 27/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5718 - acc: 0.7068 - val_loss: 0.7409 - val_acc: 0.5468\n",
            "Epoch 28/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5702 - acc: 0.6978 - val_loss: 0.7869 - val_acc: 0.5036\n",
            "Epoch 29/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5729 - acc: 0.6960 - val_loss: 0.7404 - val_acc: 0.5252\n",
            "Epoch 30/30\n",
            "556/556 [==============================] - 1s 2ms/step - loss: 0.5546 - acc: 0.7320 - val_loss: 0.7222 - val_acc: 0.5108\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GMm62jsD4m_R",
        "colab_type": "code",
        "outputId": "0bd59f0a-a916-4353-ecd8-84a41674794f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 914
        }
      },
      "source": [
        "# Metrics analysis\n",
        "plot_graphs(history3, 'acc')\n",
        "plot_graphs(history3, 'loss')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEGCAYAAACQO2mwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEGCAYAAACHGfl5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n+wAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-221-61c3e98f2d2e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mplot_graphs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhistory3\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'acc'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mplot_graphs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhistory3\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'loss'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mpredicted_labels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel3\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest_padded_sequences\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      4\u001b[0m \u001b[0mpredicted_labels_bool\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpredicted_labels\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mpredict\u001b[0;34m(self, x, batch_size, verbose, steps, callbacks, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1378\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1379\u001b[0m         \u001b[0;31m# Case 2: Symbolic tensors or Numpy array-like.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1380\u001b[0;31m         \u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_standardize_user_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1381\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstateful\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1382\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/keras/engine/training.py\u001b[0m in \u001b[0;36m_standardize_user_data\u001b[0;34m(self, x, y, sample_weight, class_weight, check_array_lengths, batch_size)\u001b[0m\n\u001b[1;32m    755\u001b[0m             \u001b[0mfeed_input_shapes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    756\u001b[0m             \u001b[0mcheck_batch_axis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m  \u001b[0;31m# Don't enforce the batch size.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 757\u001b[0;31m             exception_prefix='input')\n\u001b[0m\u001b[1;32m    758\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    759\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0my\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/keras/engine/training_utils.py\u001b[0m in \u001b[0;36mstandardize_input_data\u001b[0;34m(data, names, shapes, check_batch_axis, exception_prefix)\u001b[0m\n\u001b[1;32m    129\u001b[0m                         \u001b[0;34m': expected '\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mnames\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m' to have '\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    130\u001b[0m                         \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m' dimensions, but got array '\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 131\u001b[0;31m                         'with shape ' + str(data_shape))\n\u001b[0m\u001b[1;32m    132\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mcheck_batch_axis\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    133\u001b[0m                     \u001b[0mdata_shape\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata_shape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: Error when checking input: expected conv1d_5_input to have 3 dimensions, but got array with shape (139, 120)"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tEW-wHfAuegp",
        "colab_type": "code",
        "outputId": "060f9fc2-19f1-42f0-bfba-6f770e503fac",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "vocab_size = 1000\n",
        "max_length2 = 150\n",
        "\n",
        "train_sentences2 = [' '.join([word for word in sentence.split(' ') if word.strip() not in stopwords]) for sentence in train_sentences]\n",
        "test_sentences2 = [' '.join([word for word in sentence.split(' ') if word.strip() not in stopwords]) for sentence in test_sentences]\n",
        "\n",
        "tokenizer2 = Tokenizer(num_words=vocab_size, oov_token=oov_tok)\n",
        "\n",
        "tokenizer2.fit_on_texts(train_sentences2)\n",
        "word_index = tokenizer2.word_index\n",
        "print(len(word_index))\n",
        "\n",
        "train_sequences2 = tokenizer.texts_to_sequences(train_sentences2)\n",
        "train_padded_sequences2 = pad_sequences(train_sequences2, maxlen=max_length2,\n",
        "  padding=padding_type, truncating=trunc_type)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "7928\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0XNneP1XuptX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "test_sequences2 = tokenizer.texts_to_sequences(test_sentences2)\n",
        "test_padded_sequences2 = pad_sequences(test_sequences2, maxlen=max_length2,\n",
        "  padding=padding_type, truncating=trunc_type)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fzxJcp9quU-C",
        "colab_type": "code",
        "outputId": "c52eacc3-7851-4bac-9f9d-0d82374c1ace",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        }
      },
      "source": [
        "# hyperparametes\n",
        "embedding_dim_2 = 64\n",
        "memory_units = 64\n",
        "dense_units = 32\n",
        "dropout_rate = 0.2\n",
        "\n",
        "model4 = keras.Sequential([\n",
        "  keras.layers.Embedding(vocab_size, embedding_dim_2, input_length=max_length2),\n",
        "  keras.layers.Dropout(dropout_rate),\n",
        "  keras.layers.LSTM(memory_units),\n",
        "  keras.layers.Dropout(dropout_rate),\n",
        "  keras.layers.Dense(dense_units, activation='relu'),\n",
        "  keras.layers.Dropout(dropout_rate),\n",
        "  keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "model4.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "print(model4.summary())"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_4\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_4 (Embedding)      (None, 150, 64)           64000     \n",
            "_________________________________________________________________\n",
            "dropout_6 (Dropout)          (None, 150, 64)           0         \n",
            "_________________________________________________________________\n",
            "lstm_4 (LSTM)                (None, 64)                33024     \n",
            "_________________________________________________________________\n",
            "dropout_7 (Dropout)          (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "dense_7 (Dense)              (None, 32)                2080      \n",
            "_________________________________________________________________\n",
            "dropout_8 (Dropout)          (None, 32)                0         \n",
            "_________________________________________________________________\n",
            "dense_8 (Dense)              (None, 1)                 33        \n",
            "=================================================================\n",
            "Total params: 99,137\n",
            "Trainable params: 99,137\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XRwsDgh-vAtx",
        "colab_type": "code",
        "outputId": "ce1d132c-0fe7-4bfb-faa5-085d0dae5880",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# training\n",
        "session=tf.Session()\n",
        "with session.as_default():\n",
        "    with session.graph.as_default():\n",
        "      num_epochs = 30\n",
        "      history4 = model4.fit(train_padded_sequences2, train_labels, epochs=num_epochs,\n",
        "                    validation_data=(test_padded_sequences2, test_labels), batch_size=64, callbacks=[clbk()])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 556 samples, validate on 139 samples\n",
            "Epoch 1/30\n",
            "556/556 [==============================] - 4s 7ms/step - loss: 0.6920 - acc: 0.5216 - val_loss: 0.6961 - val_acc: 0.4748\n",
            "Epoch 2/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.6933 - acc: 0.5216 - val_loss: 0.6993 - val_acc: 0.4748\n",
            "Epoch 3/30\n",
            "556/556 [==============================] - 3s 6ms/step - loss: 0.6915 - acc: 0.5216 - val_loss: 0.6968 - val_acc: 0.4748\n",
            "Epoch 4/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6912 - acc: 0.5360 - val_loss: 0.6963 - val_acc: 0.4748\n",
            "Epoch 5/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6888 - acc: 0.5522 - val_loss: 0.6976 - val_acc: 0.4676\n",
            "Epoch 6/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.6869 - acc: 0.5809 - val_loss: 0.6994 - val_acc: 0.4676\n",
            "Epoch 7/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6811 - acc: 0.6097 - val_loss: 0.7000 - val_acc: 0.4532\n",
            "Epoch 8/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6619 - acc: 0.6349 - val_loss: 0.6329 - val_acc: 0.6547\n",
            "Epoch 9/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.5979 - acc: 0.6978 - val_loss: 0.9699 - val_acc: 0.5396\n",
            "Epoch 10/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6241 - acc: 0.7176 - val_loss: 0.6223 - val_acc: 0.6906\n",
            "Epoch 11/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6070 - acc: 0.7104 - val_loss: 0.6165 - val_acc: 0.6978\n",
            "Epoch 12/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.6003 - acc: 0.6637 - val_loss: 0.6170 - val_acc: 0.7194\n",
            "Epoch 13/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.5994 - acc: 0.6727 - val_loss: 0.7475 - val_acc: 0.6906\n",
            "Epoch 14/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.7063 - acc: 0.7158 - val_loss: 0.9048 - val_acc: 0.5827\n",
            "Epoch 15/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6724 - acc: 0.6547 - val_loss: 0.6836 - val_acc: 0.5324\n",
            "Epoch 16/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6547 - acc: 0.5881 - val_loss: 0.6671 - val_acc: 0.5396\n",
            "Epoch 17/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.6518 - acc: 0.5953 - val_loss: 0.6914 - val_acc: 0.5180\n",
            "Epoch 18/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.6568 - acc: 0.6187 - val_loss: 0.6995 - val_acc: 0.5180\n",
            "Epoch 19/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6505 - acc: 0.5827 - val_loss: 0.6817 - val_acc: 0.5180\n",
            "Epoch 20/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.6533 - acc: 0.5971 - val_loss: 0.6864 - val_acc: 0.5180\n",
            "Epoch 21/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6517 - acc: 0.5899 - val_loss: 0.6896 - val_acc: 0.5180\n",
            "Epoch 22/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.6589 - acc: 0.5881 - val_loss: 0.6823 - val_acc: 0.5180\n",
            "Epoch 23/30\n",
            "556/556 [==============================] - 2s 4ms/step - loss: 0.6347 - acc: 0.6187 - val_loss: 0.6831 - val_acc: 0.5180\n",
            "Epoch 24/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6582 - acc: 0.5809 - val_loss: 0.6805 - val_acc: 0.5180\n",
            "Epoch 25/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6554 - acc: 0.6115 - val_loss: 0.6824 - val_acc: 0.5180\n",
            "Epoch 26/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6574 - acc: 0.6025 - val_loss: 0.6840 - val_acc: 0.5180\n",
            "Epoch 27/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6637 - acc: 0.6025 - val_loss: 0.6846 - val_acc: 0.5180\n",
            "Epoch 28/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6581 - acc: 0.6007 - val_loss: 0.6883 - val_acc: 0.5180\n",
            "Epoch 29/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6530 - acc: 0.5971 - val_loss: 0.6804 - val_acc: 0.5180\n",
            "Epoch 30/30\n",
            "556/556 [==============================] - 3s 5ms/step - loss: 0.6483 - acc: 0.6007 - val_loss: 0.6751 - val_acc: 0.5324\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ng088UJuvBSx",
        "colab_type": "code",
        "outputId": "d272a1f6-4a75-4cb7-de55-cd46596bcf60",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 178
        }
      },
      "source": [
        "# Metrics analysis\n",
        "plot_graphs(history2, 'acc')\n",
        "plot_graphs(history2, 'loss')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-47-d09c4c00a6ea>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mplot_graphs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhistory2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'acc'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mplot_graphs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhistory2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'loss'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'plot_graphs' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uOrD4LSxvIWr",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}