ANN_LSTM.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ANN-LSTM.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# **ANN**"
      ],
      "metadata": {
        "id": "giDYCA7k9Uwy"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Import Packages"
      ],
      "metadata": {
        "id": "A3E4kgJT6mg5"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "metadata": {
        "id": "46uE02cO1kJp"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "import seaborn as sn\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.naive_bayes import GaussianNB\n",
        "from sklearn.svm import LinearSVC\n",
        "\n",
        "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "from sklearn.model_selection import KFold\n",
        "from sklearn.metrics import confusion_matrix, f1_score"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Data Reading"
      ],
      "metadata": {
        "id": "r9zhKXau6t3L"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import files\n",
        "uploaded = files.upload()"
      ],
      "metadata": {
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 74
        },
        "id": "UEimHOUQ2CZb",
        "outputId": "6ff72f1c-350a-4a4f-c51e-db357e010411"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-d06bf1c0-5a88-48ac-94c9-d431d5fd4e0d\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-d06bf1c0-5a88-48ac-94c9-d431d5fd4e0d\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving spam_ham_dataset.csv to spam_ham_dataset (2).csv\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data = pd.read_csv(\"spam_ham_dataset.csv\", header=0)\n",
        "data = data.drop('Unnamed: 0', axis=1)\n",
        "#data = data.drop('label_num', axis=1)"
      ],
      "metadata": {
        "id": "jwKBvm8w1qak"
      },
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "rfRR28jG4sv3",
        "outputId": "44d89fd0-5225-40d7-d5d9-9ed91e7dbdcc"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  label                                               text  label_num\n",
              "0   ham  Subject: enron methanol ; meter # : 988291\\r\\n...          0\n",
              "1   ham  Subject: hpl nom for january 9 , 2001\\r\\n( see...          0\n",
              "2   ham  Subject: neon retreat\\r\\nho ho ho , we ' re ar...          0\n",
              "3  spam  Subject: photoshop , windows , office . cheap ...          1\n",
              "4   ham  Subject: re : indian springs\\r\\nthis deal is t...          0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-dadfb616-3660-4184-a224-27c08748976c\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>label</th>\n",
              "      <th>text</th>\n",
              "      <th>label_num</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ham</td>\n",
              "      <td>Subject: enron methanol ; meter # : 988291\\r\\n...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ham</td>\n",
              "      <td>Subject: hpl nom for january 9 , 2001\\r\\n( see...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ham</td>\n",
              "      <td>Subject: neon retreat\\r\\nho ho ho , we ' re ar...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>spam</td>\n",
              "      <td>Subject: photoshop , windows , office . cheap ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ham</td>\n",
              "      <td>Subject: re : indian springs\\r\\nthis deal is t...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-dadfb616-3660-4184-a224-27c08748976c')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-dadfb616-3660-4184-a224-27c08748976c button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-dadfb616-3660-4184-a224-27c08748976c');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "x = data['text']\n",
        "y = data['label_num']\n",
        "x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.3,\n",
        "                                         random_state=1) \n",
        "print(\"Train Data size:\", len(x_train))\n",
        "print(\"Test Data size\", len(x_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vOGYiAEz5Xh7",
        "outputId": "0f111b24-b2c4-42b7-b28b-fa0e454f5cff"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Data size: 3619\n",
            "Test Data size 1552\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.preprocessing.text import Tokenizer\n",
        "\n",
        "tokenizer = Tokenizer()\n",
        "tokenizer.fit_on_texts(x_train)\n",
        "\n",
        "word_index = tokenizer.word_index\n",
        "vocab_size = len(tokenizer.word_index) + 1000\n",
        "print(\"Vocabulary Size :\", vocab_size)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kJimJ4VS6BEi",
        "outputId": "c0ca59cd-93e7-427f-8a6a-06532eace97d"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Vocabulary Size : 47083\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "x_train = pad_sequences(tokenizer.texts_to_sequences(x_train),\n",
        "                        maxlen = 50)\n",
        "x_test = pad_sequences(tokenizer.texts_to_sequences(x_test),\n",
        "                       maxlen = 50)\n",
        "\n",
        "print(\"Training X Shape:\",x_train.shape)\n",
        "print(\"Testing X Shape:\",x_test.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tUO8wfXI501v",
        "outputId": "389a161f-882c-43e4-d0c8-9b59cae6b599"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training X Shape: (3619, 50)\n",
            "Testing X Shape: (1552, 50)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout\n",
        "ann = Sequential()\n",
        "ann.add(Dense(16, activation='relu', input_dim=50))\n",
        "# droput overfit'i önlemek için eklendi\n",
        "ann.add(Dropout(0.1))\n",
        "ann.add(Dense(1, activation='sigmoid'))"
      ],
      "metadata": {
        "id": "nWRWhfmA48TQ"
      },
      "execution_count": 58,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ann.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qd5n5S-B5FnE",
        "outputId": "717498c7-a429-4a5a-df26-0a774d46d896"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_4\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense_8 (Dense)             (None, 16)                816       \n",
            "                                                                 \n",
            " dropout_4 (Dropout)         (None, 16)                0         \n",
            "                                                                 \n",
            " dense_9 (Dense)             (None, 1)                 17        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 833\n",
            "Trainable params: 833\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "ann.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "kQH2Jg0H5Kxi"
      },
      "execution_count": 60,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ann.fit(x_train, y_train, batch_size=100, epochs=30)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rzTRo8ab5OeA",
        "outputId": "f33c5424-c8de-4f6d-c1c6-8ca7964afaae"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/30\n",
            "37/37 [==============================] - 0s 4ms/step - loss: 0.6179 - accuracy: 0.7458\n",
            "Epoch 2/30\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.5592 - accuracy: 0.7568\n",
            "Epoch 3/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.6571 - accuracy: 0.7447\n",
            "Epoch 4/30\n",
            "37/37 [==============================] - 0s 7ms/step - loss: 0.6978 - accuracy: 0.7262\n",
            "Epoch 5/30\n",
            "37/37 [==============================] - 0s 6ms/step - loss: 0.5798 - accuracy: 0.7519\n",
            "Epoch 6/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.6419 - accuracy: 0.7499\n",
            "Epoch 7/30\n",
            "37/37 [==============================] - 0s 6ms/step - loss: 0.5638 - accuracy: 0.7555\n",
            "Epoch 8/30\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.6027 - accuracy: 0.7472\n",
            "Epoch 9/30\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.6490 - accuracy: 0.7472\n",
            "Epoch 10/30\n",
            "37/37 [==============================] - 0s 10ms/step - loss: 0.6044 - accuracy: 0.7532\n",
            "Epoch 11/30\n",
            "37/37 [==============================] - 0s 7ms/step - loss: 0.6151 - accuracy: 0.7488\n",
            "Epoch 12/30\n",
            "37/37 [==============================] - 0s 6ms/step - loss: 0.5754 - accuracy: 0.7629\n",
            "Epoch 13/30\n",
            "37/37 [==============================] - 0s 6ms/step - loss: 0.6554 - accuracy: 0.7336\n",
            "Epoch 14/30\n",
            "37/37 [==============================] - 0s 4ms/step - loss: 0.6430 - accuracy: 0.7325\n",
            "Epoch 15/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.5763 - accuracy: 0.7541\n",
            "Epoch 16/30\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.6159 - accuracy: 0.7544\n",
            "Epoch 17/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.6705 - accuracy: 0.7392\n",
            "Epoch 18/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.5915 - accuracy: 0.7505\n",
            "Epoch 19/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.5869 - accuracy: 0.7563\n",
            "Epoch 20/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.6236 - accuracy: 0.7535\n",
            "Epoch 21/30\n",
            "37/37 [==============================] - 0s 6ms/step - loss: 0.5821 - accuracy: 0.7571\n",
            "Epoch 22/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.5746 - accuracy: 0.7568\n",
            "Epoch 23/30\n",
            "37/37 [==============================] - 0s 4ms/step - loss: 0.7164 - accuracy: 0.7386\n",
            "Epoch 24/30\n",
            "37/37 [==============================] - 0s 4ms/step - loss: 0.5487 - accuracy: 0.7651\n",
            "Epoch 25/30\n",
            "37/37 [==============================] - 0s 4ms/step - loss: 0.5655 - accuracy: 0.7671\n",
            "Epoch 26/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.5778 - accuracy: 0.7577\n",
            "Epoch 27/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.6928 - accuracy: 0.7375\n",
            "Epoch 28/30\n",
            "37/37 [==============================] - 0s 7ms/step - loss: 0.5738 - accuracy: 0.7640\n",
            "Epoch 29/30\n",
            "37/37 [==============================] - 0s 5ms/step - loss: 0.5884 - accuracy: 0.7458\n",
            "Epoch 30/30\n",
            "37/37 [==============================] - 0s 4ms/step - loss: 0.5353 - accuracy: 0.7654\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fc25e9a9250>"
            ]
          },
          "metadata": {},
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Test sonuçlarının tahmin edilmesi\n",
        "y_pred = ann.predict(x_test)\n",
        "y_pred = (y_pred > 0.5)"
      ],
      "metadata": {
        "id": "bZLsFjGY5rdw"
      },
      "execution_count": 63,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y_test1 = np.array(y_test)"
      ],
      "metadata": {
        "id": "siy7rGC26Uo8"
      },
      "execution_count": 64,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "cm = confusion_matrix(y_test, y_pred)"
      ],
      "metadata": {
        "id": "2nohSb5v6X2l"
      },
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "cm"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZMykEYJk6asr",
        "outputId": "2c96fd0c-c65f-4af0-f614-be24694be13e"
      },
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1023,   78],\n",
              "       [ 303,  148]])"
            ]
          },
          "metadata": {},
          "execution_count": 66
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Accuracy is {}%\".format(((cm[0][0] + cm[1][1])/1035)*100))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p036BMIh6hpZ",
        "outputId": "08abeba1-c6c8-4d26-edc2-4ca4a15ada95"
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy is 113.14009661835749%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# ***LSTM***"
      ],
      "metadata": {
        "id": "twRPMGcU9SC2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "MAX_SEQUENCE_LENGTH = 50\n",
        "n_lstm = 200\n",
        "drop_lstm =0.2\n",
        "\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Embedding, GlobalAveragePooling1D, Dense, Dropout, LSTM, Bidirectional, GRU"
      ],
      "metadata": {
        "id": "CM0Y7kjZ9b-x"
      },
      "execution_count": 82,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "embeding_dim = 16\n",
        "drop_value = 0.2\n",
        "n_dense = 24\n",
        "num_epochs = 30\n",
        "early_stop = EarlyStopping(monitor='val_loss', patience=3)\n",
        "\n",
        "#LSTM Spam detection architecture\n",
        "lstm = Sequential()\n",
        "lstm.add(Embedding(vocab_size, embeding_dim, input_length=MAX_SEQUENCE_LENGTH))\n",
        "lstm.add(LSTM(n_lstm, dropout=drop_lstm, return_sequences=True))\n",
        "lstm.add(LSTM(n_lstm, dropout=drop_lstm, return_sequences=True))\n",
        "lstm.add(Dense(1, activation='sigmoid'))"
      ],
      "metadata": {
        "id": "RAnJM0Y5GNU8"
      },
      "execution_count": 83,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lstm.compile(loss = 'binary_crossentropy', optimizer = 'adam', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "GA4FsxAlGS9m"
      },
      "execution_count": 84,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "early_stop = EarlyStopping(monitor='val_loss', patience=5)\n",
        "history = lstm.fit(x_train, y_train,validation_data=(x_test, y_test), epochs=num_epochs, verbose=2, callbacks=[early_stop])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XqtYWt7jGWhO",
        "outputId": "259d5ad4-c2eb-4c29-9726-2702bd7e61ba"
      },
      "execution_count": 85,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/30\n",
            "114/114 - 67s - loss: 0.4420 - accuracy: 0.7933 - val_loss: 0.2965 - val_accuracy: 0.8502 - 67s/epoch - 584ms/step\n",
            "Epoch 2/30\n",
            "114/114 - 40s - loss: 0.2045 - accuracy: 0.9114 - val_loss: 0.2528 - val_accuracy: 0.8879 - 40s/epoch - 350ms/step\n",
            "Epoch 3/30\n",
            "114/114 - 37s - loss: 0.1534 - accuracy: 0.9348 - val_loss: 0.2754 - val_accuracy: 0.8877 - 37s/epoch - 326ms/step\n",
            "Epoch 4/30\n",
            "114/114 - 41s - loss: 0.1275 - accuracy: 0.9417 - val_loss: 0.3333 - val_accuracy: 0.8902 - 41s/epoch - 362ms/step\n",
            "Epoch 5/30\n",
            "114/114 - 35s - loss: 0.1163 - accuracy: 0.9467 - val_loss: 0.3851 - val_accuracy: 0.8880 - 35s/epoch - 311ms/step\n",
            "Epoch 6/30\n",
            "114/114 - 37s - loss: 0.1212 - accuracy: 0.9456 - val_loss: 0.3412 - val_accuracy: 0.8795 - 37s/epoch - 321ms/step\n",
            "Epoch 7/30\n",
            "114/114 - 34s - loss: 0.1152 - accuracy: 0.9479 - val_loss: 0.3323 - val_accuracy: 0.8765 - 34s/epoch - 296ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "metrics = pd.DataFrame(history.history)\n",
        "\n",
        "metrics.rename(columns= {'loss': 'Training_Loss', 'accuracy': 'Training_Accuracy', \n",
        "                         'val_loss': 'Validation_Loss', 'val_accuracy': 'Validation_Accuracy'}, \n",
        "               inplace = True)\n",
        "\n",
        "def plot_graphs(var1, var2, string):\n",
        "    metrics[[var1, var2]].plot()\n",
        "    plt.title('LSTM Model: Training and Validation ' + string)\n",
        "    plt.xlabel ('Number of epochs')\n",
        "    plt.ylabel(string)\n",
        "    plt.legend([var1, var2])\n",
        "\n",
        "plot_graphs('Training_Loss', 'Validation_Loss', 'loss')\n",
        "plot_graphs('Training_Accuracy', 'Validation_Accuracy', 'accuracy')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "id": "LWoVtg9fGbf9",
        "outputId": "7a9a81df-b832-4b8e-98a5-1fc2686ab5cd"
      },
      "execution_count": 90,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Tahmin yapılması\n",
        "train_pred = lstm.predict(x_train, batch_size=256)\n",
        "lstm.reset_states()\n",
        "test_pred = lstm.predict(x_test, batch_size=256)"
      ],
      "metadata": {
        "id": "8SVItlC_GezJ"
      },
      "execution_count": 91,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "pred=np.concatenate((train_pred,test_pred),axis=0)"
      ],
      "metadata": {
        "id": "z1V6F5EWGfd2"
      },
      "execution_count": 94,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_score = lstm.evaluate(x_train, y_train, verbose=0)\n",
        "print(\"Accuracy is %{}\".format(train_score[1]*100))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZaVmQI9LGhrA",
        "outputId": "1742cafc-57ae-4f4d-c6e6-2fb0213d537b"
      },
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy is %94.94999647140503\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Multi-Class Perceptron Classification**"
      ],
      "metadata": {
        "id": "xdS9LzrwNz2k"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.neural_network import MLPClassifier\n",
        "X = data['text']\n",
        "y = data['label']\n",
        "clf = MLPClassifier(solver='lbfgs', alpha=1e-5,\n",
        "                    hidden_layer_sizes=(5, 2), random_state=1)\n",
        "\n",
        "clf.fit(x_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0KWQt-rqN6nY",
        "outputId": "aa962bab-78fc-448b-d68d-fc5a43474857"
      },
      "execution_count": 99,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:549: ConvergenceWarning: lbfgs failed to converge (status=2):\n",
            "ABNORMAL_TERMINATION_IN_LNSRCH.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "  self.n_iter_ = _check_optimize_result(\"lbfgs\", opt_res, self.max_iter)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "MLPClassifier(alpha=1e-05, hidden_layer_sizes=(5, 2), random_state=1,\n",
              "              solver='lbfgs')"
            ]
          },
          "metadata": {},
          "execution_count": 99
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_mlp = clf.predict(x_test)"
      ],
      "metadata": {
        "id": "gvr1iQahPAD9"
      },
      "execution_count": 101,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "print(\"Accuracy is \",accuracy_score(y_test, test_mlp))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nwTfD5DtPFNG",
        "outputId": "50357431-af7a-4460-f2d2-dd50eaf1d171"
      },
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy is  0.7094072164948454\n"
          ]
        }
      ]
    }
  ]
}