SO_notebook.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Stack Overflow Topic Classification using RNN\n",
        "\n",
        "Author: Jash Dubal"
      ],
      "metadata": {
        "id": "8wYaaltYp4z6"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        " ## Imports"
      ],
      "metadata": {
        "id": "fW69GajWqU-s"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "j7EIvVsVo257",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2d9887fd-e3c7-45ff-ee0a-d114d92ba4b0"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.0.0+cu118\n"
          ]
        }
      ],
      "source": [
        "import torch\n",
        "import pandas as pd\n",
        "from torchtext import data\n",
        "import warnings as wrn\n",
        "wrn.filterwarnings('ignore')\n",
        "print(torch.__version__)\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "%matplotlib inline"
      ],
      "metadata": {
        "id": "HMkTqKxop3gk"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Load the data"
      ],
      "metadata": {
        "id": "cy5xtv2zrBnP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Load Stack Overflow data into a pandas dataframe\n",
        "# Ensure dataset/SO.csv exists\n",
        "try:\n",
        "  df = pd.read_csv('dataset/SO.csv')\n",
        "  print('df loaded successfully!')\n",
        "except Exception as e:\n",
        "  print('dataset/SO.csv not found')"
      ],
      "metadata": {
        "id": "pj0bNm9lqGmS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "058860f7-7bda-4ad9-fee5-62f1bb4dbcf6"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "df loaded successfully!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Show first rows of dataframe\n",
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "TZrjPYBcrRG5",
        "outputId": "99c9c61e-eafa-44b5-f217-87475f8476b7"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                               Title  Label\n",
              "0  Different rlike behavior in Spark 1.6 and Spar...  spark\n",
              "1  Getting a column as concatenated column from a...  spark\n",
              "2  Write data using JDBC connection to Azure SQL ...  spark\n",
              "3  Get value from external client database for a ...  spark\n",
              "4  How to setup Apache Spark to use local hard di...  spark"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-dbf93f1f-b6de-46e0-ba97-87537135b461\">\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>Title</th>\n",
              "      <th>Label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Different rlike behavior in Spark 1.6 and Spar...</td>\n",
              "      <td>spark</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Getting a column as concatenated column from a...</td>\n",
              "      <td>spark</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Write data using JDBC connection to Azure SQL ...</td>\n",
              "      <td>spark</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Get value from external client database for a ...</td>\n",
              "      <td>spark</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>How to setup Apache Spark to use local hard di...</td>\n",
              "      <td>spark</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-dbf93f1f-b6de-46e0-ba97-87537135b461')\"\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-dbf93f1f-b6de-46e0-ba97-87537135b461 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-dbf93f1f-b6de-46e0-ba97-87537135b461');\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": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mmjUlfZurfvb",
        "outputId": "b17f91d5-06f8-41cf-eac1-de29f4418571"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 150000 entries, 0 to 149999\n",
            "Data columns (total 2 columns):\n",
            " #   Column  Non-Null Count   Dtype \n",
            "---  ------  --------------   ----- \n",
            " 0   Title   150000 non-null  object\n",
            " 1   Label   150000 non-null  object\n",
            "dtypes: object(2)\n",
            "memory usage: 2.3+ MB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# This cell generates a bar plot to visualize the distribution of categories in the dataset.\n",
        "# The purpose of this plot is to understand the class balance and any potential biases in the dataset.\n",
        "\n",
        "import seaborn as sns\n",
        "\n",
        "# Set the figure size\n",
        "plt.figure(figsize=(8, 6))\n",
        "\n",
        "# Generate a count plot using Seaborn\n",
        "sns.countplot(data=df, x='Label')\n",
        "\n",
        "# Set the title and labels for the plot\n",
        "plt.title('Distribution of Categories')\n",
        "plt.xlabel('Category')\n",
        "plt.ylabel('Frequency')\n",
        "\n",
        "# Display the plot\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "QyeP6lP1risJ",
        "outputId": "744aed16-a447-432d-b58d-9211fcdeb7ee"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Dataframe target distribution\n",
        "df['Label'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DJsYrV85rkYr",
        "outputId": "6a8800a8-6035-4b71-c7b0-b052ec20eaba"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "spark       50000\n",
              "ml          50000\n",
              "security    50000\n",
              "Name: Label, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Preprocess natural language text\n",
        "Steps:\n",
        "- Remove non-alphabetical characters\n",
        "- Convert to lowercase\n",
        "- Remove stopwords\n",
        "- Lemmatize\n",
        "- Tokenize"
      ],
      "metadata": {
        "id": "s1FMuKf6ron6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import re\n",
        "import pandas as pd\n",
        "import nltk\n",
        "from nltk.corpus import stopwords, wordnet\n",
        "from nltk.stem import WordNetLemmatizer\n",
        "from nltk.tokenize import word_tokenize\n",
        "from sklearn.base import BaseEstimator, TransformerMixin\n",
        "nltk.download('stopwords')\n",
        "nltk.download('punkt')\n",
        "nltk.download('wordnet')\n",
        "\n",
        "\n",
        "nltk.download('averaged_perceptron_tagger')\n",
        "\n",
        "class TextPreprocessor(BaseEstimator, TransformerMixin):\n",
        "    \"\"\"\n",
        "    A text preprocessor that removes non-alphabetic characters, converts text to lowercase,\n",
        "    tokenizes text, removes stopwords, and lemmatizes text with POS tagging.\n",
        "    \"\"\"\n",
        "    \n",
        "    def __init__(self):\n",
        "        \"\"\"\n",
        "        Initializes a new TextPreprocessor object with a WordNetLemmatizer and a set of stopwords.\n",
        "        \"\"\"\n",
        "        self.lemmatizer = WordNetLemmatizer()\n",
        "        self.stop_words = set(stopwords.words('english'))\n",
        "        \n",
        "    def get_wordnet_pos(self, word):\n",
        "        \"\"\"\n",
        "        Maps the POS tag to the first character that the lemmatize() function accepts.\n",
        "        \n",
        "        Parameters:\n",
        "        - word (str): a word to be lemmatized\n",
        "        \n",
        "        Returns:\n",
        "        - wordnet tag (str): a wordnet tag for the word to be lemmatized\n",
        "        \"\"\"\n",
        "        tag = nltk.pos_tag([word])[0][1][0].upper()\n",
        "        tag_dict = {\"J\": wordnet.ADJ,\n",
        "                    \"N\": wordnet.NOUN,\n",
        "                    \"V\": wordnet.VERB,\n",
        "                    \"R\": wordnet.ADV}\n",
        "\n",
        "        return tag_dict.get(tag, wordnet.NOUN)\n",
        "    \n",
        "    def _preprocess(self, text):\n",
        "        \"\"\"\n",
        "        Applies a series of text preprocessing steps to a string of text.\n",
        "        \n",
        "        Parameters:\n",
        "        - text (str): a string of text to be preprocessed\n",
        "        \n",
        "        Returns:\n",
        "        - tokens (list): a list of preprocessed and lemmatized tokens\n",
        "        \"\"\"\n",
        "        # Remove non-alphabetic characters\n",
        "        text = re.sub(r'[^a-zA-Z\\s]', '', text)\n",
        "        # Convert to lowercase\n",
        "        text = text.lower()\n",
        "        # Tokenize text\n",
        "        tokens = word_tokenize(text)\n",
        "        # Remove stopwords\n",
        "        tokens = [token for token in tokens if token not in self.stop_words]\n",
        "        # Lemmatize text with POS tagging\n",
        "        tokens = [self.lemmatizer.lemmatize(token, self.get_wordnet_pos(token)) for token in tokens]\n",
        "        # Return tokenized text\n",
        "        return tokens\n",
        "\n",
        "    def fit(self, X, y=None):\n",
        "        \"\"\"\n",
        "        Fits the TextPreprocessor to the input data X.\n",
        "        \n",
        "        Parameters:\n",
        "        - X (pandas.Series): a series of text data\n",
        "        \n",
        "        Returns:\n",
        "        - self (TextPreprocessor): returns the TextPreprocessor object\n",
        "        \"\"\"\n",
        "        return self\n",
        "    \n",
        "    def transform(self, X, y=None):\n",
        "        \"\"\"\n",
        "        Applies the TextPreprocessor to the input data X and returns the preprocessed data.\n",
        "        \n",
        "        Parameters:\n",
        "        - X (pandas.Series): a series of text data\n",
        "        \n",
        "        Returns:\n",
        "        - X_processed (pandas.Series): the preprocessed data as a pandas series\n",
        "        \"\"\"\n",
        "        X_processed = X.apply(self._preprocess)\n",
        "        return X_processed\n",
        "\n",
        "preprocessor = TextPreprocessor()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nMRts4j1rmvt",
        "outputId": "213c641a-1b03-4ca5-faef-f9a1722260c4"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Package stopwords is already up-to-date!\n",
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Package punkt is already up-to-date!\n",
            "[nltk_data] Downloading package wordnet to /root/nltk_data...\n",
            "[nltk_data] Downloading package averaged_perceptron_tagger to\n",
            "[nltk_data]     /root/nltk_data...\n",
            "[nltk_data]   Package averaged_perceptron_tagger is already up-to-\n",
            "[nltk_data]       date!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Create a new column cleaning the title of the post\n",
        "df['Clean_Title'] = preprocessor.transform(df['Title'])"
      ],
      "metadata": {
        "id": "Bdb-_tEsrtTz"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "YtHgWI5zr9tV",
        "outputId": "c803e896-32ff-4ff6-9376-e5dbf9e1bac5"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                               Title  Label  \\\n",
              "0  Different rlike behavior in Spark 1.6 and Spar...  spark   \n",
              "1  Getting a column as concatenated column from a...  spark   \n",
              "2  Write data using JDBC connection to Azure SQL ...  spark   \n",
              "3  Get value from external client database for a ...  spark   \n",
              "4  How to setup Apache Spark to use local hard di...  spark   \n",
              "\n",
              "                                         Clean_Title  \n",
              "0         [different, rlike, behavior, spark, spark]  \n",
              "1  [get, column, concatenate, column, reference, ...  \n",
              "2  [write, data, use, jdbc, connection, azure, sq...  \n",
              "3  [get, value, external, client, database, colum...  \n",
              "4  [setup, apache, spark, use, local, hard, disk,...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6750f443-787a-416f-8690-5b5ece3c69b9\">\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>Title</th>\n",
              "      <th>Label</th>\n",
              "      <th>Clean_Title</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Different rlike behavior in Spark 1.6 and Spar...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[different, rlike, behavior, spark, spark]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Getting a column as concatenated column from a...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[get, column, concatenate, column, reference, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Write data using JDBC connection to Azure SQL ...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[write, data, use, jdbc, connection, azure, sq...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Get value from external client database for a ...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[get, value, external, client, database, colum...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>How to setup Apache Spark to use local hard di...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[setup, apache, spark, use, local, hard, disk,...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6750f443-787a-416f-8690-5b5ece3c69b9')\"\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-6750f443-787a-416f-8690-5b5ece3c69b9 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-6750f443-787a-416f-8690-5b5ece3c69b9');\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": 17
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Use GloVe word embeddings to convert list of words to vectors."
      ],
      "metadata": {
        "id": "0n4rGhfRsukm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "from torchtext.vocab import GloVe\n",
        "\n",
        "# Load pre-trained GloVe embeddings with 100-dimensional vectors\n",
        "glove = GloVe(name='6B', dim=100)\n",
        "\n",
        "def vectorize_text(tokens, embeddings):\n",
        "    \"\"\"\n",
        "    Convert a list of tokens to a PyTorch tensor of embeddings.\n",
        "\n",
        "    Args:\n",
        "    tokens: list of strings representing tokens\n",
        "    embeddings: pre-trained word embeddings to use for vectorization\n",
        "\n",
        "    Returns:\n",
        "    a PyTorch tensor of size (num_tokens, embedding_dim) containing the embeddings of the tokens\n",
        "    \"\"\"\n",
        "    if not tokens:\n",
        "        # Return an empty tensor if no tokens are present\n",
        "        return torch.tensor([])\n",
        "    # Stack the embeddings of each token to form the final tensor\n",
        "    return torch.stack([embeddings[token] for token in tokens])\n",
        "\n",
        "# Add a new column 'Vectorized_Title' to the DataFrame, containing the vectorized form of 'Clean_Title'\n",
        "df['Vectorized_Title'] = df['Clean_Title'].apply(lambda x: vectorize_text(x, glove))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p8E0HPivssUt",
        "outputId": "de9c2aa6-46be-4960-978e-fc5a30e3b130"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            ".vector_cache/glove.6B.zip: 862MB [02:41, 5.34MB/s]                           \n",
            "100%|█████████▉| 399999/400000 [00:18<00:00, 21466.11it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 333
        },
        "id": "kz1g2Ce1s1ZE",
        "outputId": "3386deac-4dc1-4479-d3f1-ed6b18167f46"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                               Title  Label  \\\n",
              "0  Different rlike behavior in Spark 1.6 and Spar...  spark   \n",
              "1  Getting a column as concatenated column from a...  spark   \n",
              "2  Write data using JDBC connection to Azure SQL ...  spark   \n",
              "3  Get value from external client database for a ...  spark   \n",
              "4  How to setup Apache Spark to use local hard di...  spark   \n",
              "\n",
              "                                         Clean_Title  \\\n",
              "0         [different, rlike, behavior, spark, spark]   \n",
              "1  [get, column, concatenate, column, reference, ...   \n",
              "2  [write, data, use, jdbc, connection, azure, sq...   \n",
              "3  [get, value, external, client, database, colum...   \n",
              "4  [setup, apache, spark, use, local, hard, disk,...   \n",
              "\n",
              "                                    Vectorized_Title  \n",
              "0  [[tensor(-0.3419), tensor(0.6998), tensor(-0.0...  \n",
              "1  [[tensor(0.1443), tensor(0.4395), tensor(0.583...  \n",
              "2  [[tensor(-0.5106), tensor(0.4867), tensor(-0.0...  \n",
              "3  [[tensor(0.1443), tensor(0.4395), tensor(0.583...  \n",
              "4  [[tensor(-0.2554), tensor(0.2010), tensor(0.49...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1895d245-7cfa-4199-a61c-ff4098c99f89\">\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>Title</th>\n",
              "      <th>Label</th>\n",
              "      <th>Clean_Title</th>\n",
              "      <th>Vectorized_Title</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Different rlike behavior in Spark 1.6 and Spar...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[different, rlike, behavior, spark, spark]</td>\n",
              "      <td>[[tensor(-0.3419), tensor(0.6998), tensor(-0.0...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Getting a column as concatenated column from a...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[get, column, concatenate, column, reference, ...</td>\n",
              "      <td>[[tensor(0.1443), tensor(0.4395), tensor(0.583...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Write data using JDBC connection to Azure SQL ...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[write, data, use, jdbc, connection, azure, sq...</td>\n",
              "      <td>[[tensor(-0.5106), tensor(0.4867), tensor(-0.0...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Get value from external client database for a ...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[get, value, external, client, database, colum...</td>\n",
              "      <td>[[tensor(0.1443), tensor(0.4395), tensor(0.583...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>How to setup Apache Spark to use local hard di...</td>\n",
              "      <td>spark</td>\n",
              "      <td>[setup, apache, spark, use, local, hard, disk,...</td>\n",
              "      <td>[[tensor(-0.2554), tensor(0.2010), tensor(0.49...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1895d245-7cfa-4199-a61c-ff4098c99f89')\"\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-1895d245-7cfa-4199-a61c-ff4098c99f89 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-1895d245-7cfa-4199-a61c-ff4098c99f89');\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": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Check the shape of the first vector to ensure it is a 100-dimensional vector according to the GloVe embeddings\n",
        "df['Vectorized_Title'][0].shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4RAQC81Gs3xV",
        "outputId": "e24a2a7e-4322-47cf-ab6a-6402d7c672d8"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([5, 100])"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Filter out all rows that have empty tensors\n",
        "df = df[df[\"Vectorized_Title\"].apply(lambda x: len(x) > 0)]"
      ],
      "metadata": {
        "id": "5qmHhji1s9bS"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Splitting\n",
        "Split data into train, validation, test set"
      ],
      "metadata": {
        "id": "o8zK8ptHuRDI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# This cell splits the dataset into train, validation, and test sets using an 80/20 split for train/test and \n",
        "# further splitting the train set with an 80/20 ratio for train/validation.\n",
        "# The split is stratified by the 'Label' column to maintain the class distribution in each split.\n",
        "\n",
        "# Import the required library\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# Split the dataset into train and test sets with an 80/20 split, stratified by the 'Label' column\n",
        "train_df, test_df = train_test_split(df, test_size=0.2, random_state=42, stratify=df['Label'])\n",
        "\n",
        "# Split the train dataset into train and validation sets with an 80/20 split, stratified by the 'Label' column\n",
        "train_df, val_df = train_test_split(train_df, test_size=0.25, random_state=42, stratify=train_df['Label'])"
      ],
      "metadata": {
        "id": "c1Mn1cncuPAl"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Ensure that the split has even distribution of classes"
      ],
      "metadata": {
        "id": "N17XzINyucyo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# This cell generates bar plots to visualize the distribution of labels in the training, validation, and test sets.\n",
        "# The purpose of these plots is to confirm that the stratified split has maintained the class distribution in each split.\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "def plot_label_distribution(df, title, subplot_loc):\n",
        "    # Function that creates a bar plot for the label distribution of a given DataFrame\n",
        "    plt.subplot(subplot_loc)\n",
        "    label_counts = df['Label'].value_counts()\n",
        "    sns.barplot(x=label_counts.index, y=label_counts.values)\n",
        "    plt.title(title)\n",
        "    plt.xlabel('Label')\n",
        "    plt.ylabel('Count')\n",
        "\n",
        "# Create a 1x3 grid of subplots\n",
        "plt.figure(figsize=(15, 5))\n",
        "\n",
        "def print_label_distribution(df, title):\n",
        "    # Function that prints the label distribution of a given DataFrame\n",
        "    label_counts = df['Label'].value_counts()\n",
        "    print(title)\n",
        "    print(label_counts)\n",
        "\n",
        "# Plot label distributions for each dataset\n",
        "plot_label_distribution(train_df, 'Label Distribution in Training Set', 131)\n",
        "plot_label_distribution(val_df, 'Label Distribution in Validation Set', 132)\n",
        "plot_label_distribution(test_df, 'Label Distribution in Test Set', 133)\n",
        "\n",
        "# Adjust the layout and display the plots\n",
        "plt.tight_layout()\n",
        "plt.show()\n",
        "\n",
        "# Print label distributions for each dataset\n",
        "print_label_distribution(train_df, 'Label Distribution in Training Set')\n",
        "print_label_distribution(val_df, 'Label Distribution in Validation Set')\n",
        "print_label_distribution(test_df, 'Label Distribution in Test Set')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 587
        },
        "id": "Be7FoZ5_uaW2",
        "outputId": "34c2af70-73d6-45f2-8e44-4972567c4875"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x500 with 3 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABdEAAAHqCAYAAADrpwd3AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABueElEQVR4nO3deVwVdf///yeLLKKAGyBFhNrlvqWmZO4kKpmULZqVGumlF5jL59LyynCrvNLcs8wWsS4ptcVKDUXcSnFDcdeyNE0FTQXSFBTm90c/5usRBkWPLPq4327nVmferzPznsMcX8OTc+Y4GIZhCAAAAAAAAAAA5OFY3BMAAAAAAAAAAKCkIkQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBER6lx+PBhOTg46O2337bbOtesWSMHBwetWbPGbuvMNWbMGDk4ONh9vflp27at2rZta97P3a8vvviiSLbfp08f3XvvvUWyrSvlHhMxMTFFvu2b4eDgoDFjxtzQY++991716dPHrvMBgBtFb7ZGb44p8m1fr/yOg+vtrzExMXJwcNDhw4ftNp/S8JwBQGFwfmCN84OYIt82YC+E6Lilcn/R2Lp1a3FP5abk7kfuzc3NTf7+/goNDdWMGTP0559/2mU7x48f15gxY5ScnGyX9dlTSZ6bPVz9M7a6FccJR0lx7tw5jR49WvXq1ZOHh4cqVaqkRo0aafDgwTp+/Hih17d3716NGTPGrkEEgGujNxdOSe5/JXlu9nDy5Ek5Ozvr2Weftaz5888/5e7urscff7wIZ3ZjYmNjNW3atOKeRh7fffed2rRpIx8fH5UtW1bVqlXTU089pbi4uBta35tvvqnFixfbd5IAbjnODwqnJPfgkjw3eyjq3903bNigMWPGKC0t7bofQ2+9PTkX9wSA0mTcuHEKCgrSpUuXlJKSojVr1mjIkCGaMmWKvv32WzVo0MCsHTVqlF555ZVCrf/48eMaO3as7r33XjVq1Oi6H7dixYpCbedGFDS3Dz74QDk5Obd8DlcLDAzUhQsXVKZMmZteV+vWrfXpp5/aLHvxxRf1wAMPqH///uaycuXK3fS2Lly4IGfnG/vn98CBA3J0LPq/f166dEmtW7fW/v371bt3bw0aNEjnzp3Tnj17FBsbq8cee0z+/v6FWufevXs1duxYtW3b9o7+4wSAm0Nvvn17s4+Pjx5++GF98803+uuvv1S2bNk8NV999ZUuXrxYYNB+PYqiv8bGxmr37t0aMmSIzXJ7PmeF9fbbb2v48OFq06aNRo4cqbJly+rgwYNauXKlPv/8c3Xq1KnQ63zzzTf1xBNPKDw83P4TBoDrxPnB7Xt+UJS/u0t/h+hjx45Vnz595O3tfc16euvtixAdKITOnTuradOm5v2RI0dq1apVeuSRR/Too49q3759cnd3lyQ5OzvfcFB6vXJ/oXRxcbml27mW4vilT5L5zgJ7qFatmqpVq2azbMCAAapWrVqBv5hfvnxZOTk5hfoZ3MycXV1db/ixN2Px4sXavn275s+fr2eeecZm7OLFi8rKyiqWeQEAvTl/t0NvlqRevXopLi5O3377rXr06JFnPDY2Vl5eXgoLC7up7RRXf5Xs/5xdr8uXL2v8+PF6+OGH8w11Tp48WeRzAgB74fwgf7fD+cGN/u5eFOittzcu54Jil5WVpejoaDVp0kReXl7y8PBQq1attHr1asvHTJ06VYGBgXJ3d1ebNm20e/fuPDX79+/XE088oYoVK8rNzU1NmzbVt99+a/f5t2/fXq+99pp+++03/e9//zOX53ddtfj4eD300EPy9vZWuXLlVLNmTf3nP/+R9Pe10Jo1ayZJ6tu3r/kRpNxrhrVt21b16tVTUlKSWrdurbJly5qPvfq6armys7P1n//8R35+fvLw8NCjjz6qo0eP2tRYXQP0ynVea275XVft/Pnz+r//+z8FBATI1dVVNWvW1Ntvvy3DMGzqHBwcFBUVpcWLF6tevXpydXVV3bp1r+tjTvldV61Pnz4qV66cjh07pvDwcJUrV05VqlTRv//9b2VnZ19zndezvbffflvTpk1T9erV5erqqr179xbqOL76mui5x8rBgwfNv257eXmpb9+++uuvv2wee/XPK/ejbOvXr9ewYcNUpUoVeXh46LHHHtOpU6dsHpuTk6MxY8bI399fZcuWVbt27bR3797rug7sL7/8Iklq2bJlnjE3Nzd5enraLLvW6y8mJkZPPvmkJKldu3bmMXUrrnEIoPDozfTmktKbH3vsMXl4eCg2NjbP2MmTJ5WQkKAnnnhCrq6u+uGHH/Tkk0/qnnvukaurqwICAjR06FBduHDhmvPO7znfs2eP2rdvL3d3d9199916/fXX83333jfffKOwsDD5+/vL1dVV1atX1/jx4232rW3btlq6dKl+++23PB8zt7pO7KpVq9SqVSt5eHjI29tb3bp10759+2xqCnMOcbU//vhDGRkZ+fZ26e9PAlwpMzNTo0ePVo0aNcznd8SIEcrMzDRrHBwcdP78ec2bN8/cT77LBbh9cH7A+UFJOT+4HseOHdMLL7wgX19fc64ff/xxnrqZM2eqbt26Klu2rCpUqKCmTZua5x1jxozR8OHDJUlBQUHm82l1SVJ66+2Nd6Kj2GVkZOjDDz9Uz5491a9fP/3555/66KOPFBoaqs2bN+f5+NEnn3yiP//8U5GRkbp48aKmT5+u9u3ba9euXfL19ZX09y89LVu21F133aVXXnlFHh4eWrhwocLDw/Xll1/qscces+s+PPfcc/rPf/6jFStWqF+/fvnW7NmzR4888ogaNGigcePGydXVVQcPHtT69eslSbVr19a4ceMUHR2t/v37q1WrVpKkBx980FzH6dOn1blzZ/Xo0UPPPvusub9W3njjDTk4OOjll1/WyZMnNW3aNIWEhCg5Odn8q/v1uJ65XckwDD366KNavXq1IiIi1KhRIy1fvlzDhw/XsWPHNHXqVJv6H3/8UV999ZX+9a9/qXz58poxY4a6d++uI0eOqFKlStc9z1zZ2dkKDQ1V8+bN9fbbb2vlypWaPHmyqlevroEDBxZ6fVebO3euLl68qP79+8vV1VUVK1Ys9HGcn6eeekpBQUGaMGGCtm3bpg8//FA+Pj566623rvnYQYMGqUKFCho9erQOHz6sadOmKSoqSgsWLDBrRo4cqYkTJ6pr164KDQ3Vjh07FBoaqosXL15z/YGBgZL+fv2NGjWqwC/euZ7XX+vWrfXSSy9pxowZ+s9//qPatWtLkvlfAMWL3kxvLim92cPDQ926ddMXX3yhM2fOqGLFiubYggULlJ2drV69ekmSFi1apL/++ksDBw5UpUqVtHnzZs2cOVO///67Fi1aVKj5pqSkqF27drp8+bJ5vM6ZMyffn1FMTIzKlSunYcOGqVy5clq1apWio6OVkZGhSZMmSZJeffVVpaen6/fffzef64I+Zr5y5Up17txZ1apV05gxY3ThwgXNnDlTLVu21LZt2/IEIDdyDuHj4yN3d3d99913GjRokM1ze7WcnBw9+uij+vHHH9W/f3/Vrl1bu3bt0tSpU/XTTz+Z12n99NNP83ykvnr16pbrBVC6cH7A+UFJOT+4ltTUVLVo0cIM/qtUqaLvv/9eERERysjIMC+t9sEHH+ill17SE088ocGDB+vixYvauXOnNm3apGeeeUaPP/64fvrpJ3322WeaOnWqKleuLEmqUqVKvtult97mDOAWmjt3riHJ2LJli2XN5cuXjczMTJtlZ8+eNXx9fY0XXnjBXHbo0CFDkuHu7m78/vvv5vJNmzYZkoyhQ4eayzp06GDUr1/fuHjxorksJyfHePDBB4377rvPXLZ69WpDkrF69eqb3g8vLy+jcePG5v3Ro0cbV77Epk6dakgyTp06ZbmOLVu2GJKMuXPn5hlr06aNIcmYPXt2vmNt2rTJs1933XWXkZGRYS5fuHChIcmYPn26uSwwMNDo3bv3NddZ0Nx69+5tBAYGmvcXL15sSDJef/11m7onnnjCcHBwMA4ePGguk2S4uLjYLNuxY4chyZg5c2aebV0p95i4ck69e/c2JBnjxo2zqW3cuLHRpEmTAtd3NQ8PD5vnJnd7np6exsmTJ21qr/c4Noy/93n06NHm/dxj5eq6xx57zKhUqZLNsqt/XrnHZkhIiJGTk2MuHzp0qOHk5GSkpaUZhmEYKSkphrOzsxEeHm6zvjFjxhiS8j0GrvTXX38ZNWvWNCQZgYGBRp8+fYyPPvrISE1NzVN7va+/RYsWXdfrD4B90ZvpzVcqDb156dKlhiTj/ffft1neokUL46677jKys7MNw/i7V11twoQJhoODg/Hbb7+Zy64+Dgwj73M+ZMgQQ5KxadMmc9nJkycNLy8vQ5Jx6NAhc3l+2/3nP/9plC1b1uZ4DwsLs/mZ5MrvOWvUqJHh4+NjnD592ly2Y8cOw9HR0Xj++efz7Mv1nEPkJzo62pBkeHh4GJ07dzbeeOMNIykpKU/dp59+ajg6Oho//PCDzfLZs2cbkoz169eby64+fwJQOnB+wPnBlUrD+cGVru49ERERRtWqVY0//vjDpq5Hjx6Gl5eX2bu7detm1K1bt8B1T5o0KU/vLwi99fbF5VxQ7JycnMzrguXk5OjMmTO6fPmymjZtqm3btuWpDw8P11133WXef+CBB9S8eXMtW7ZMknTmzBmtWrVKTz31lP7880/98ccf+uOPP3T69GmFhobq559/1rFjx+y+H+XKlSvwm75zv4Dim2++ueEv8nB1dVXfvn2vu/75559X+fLlzftPPPGEqlataj5Xt8qyZcvk5OSkl156yWb5//3f/8kwDH3//fc2y0NCQmz+ktqgQQN5enrq119/veE5DBgwwOZ+q1atbmp9V+revXuevzwX9ji+3jmfPn1aGRkZ13xs//79bd4d3qpVK2VnZ+u3336TJCUkJOjy5cv617/+ZfO4QYMGXdfc3N3dtWnTJvOjbDExMYqIiFDVqlU1aNAg8+NmxfX6A2Bf9ObrR2++fjfamzt27KgqVarYXNLl0KFD2rhxo3r27Gl+IeiV79Q7f/68/vjjDz344IMyDEPbt28v1FyXLVumFi1a6IEHHjCXValSxXzX+5Wu3G7u8d2qVSv99ddf2r9/f6G2K0knTpxQcnKy+vTpY/MOtgYNGujhhx/O91i50XOIsWPHKjY2Vo0bN9by5cv16quvqkmTJrr//vttLh2zaNEi1a5dW7Vq1TJfv3/88Yfat28vSQVeygHA7YPzg+vH+cH1s/fv7oZh6Msvv1TXrl1lGIZN3woNDVV6erp5vHp7e+v333/Xli1bbnh7V6O33r4I0VEizJs3Tw0aNJCbm5sqVaqkKlWqaOnSpUpPT89Te9999+VZ9o9//MO8JtXBgwdlGIZee+01ValSxeY2evRoSbfmyxzOnTtn0/Su9vTTT6tly5Z68cUX5evrqx49emjhwoWFasp33XVXob6I5OrnysHBQTVq1LC8fpe9/Pbbb/L398/zfOReqiM32M11zz335FlHhQoVdPbs2RvavpubW56Q+2bWd7WgoKB8lxfmOM7P1c9DhQoVJOm65n2tx+Y+5zVq1LCpq1ixoll7LV5eXpo4caIOHz6sw4cP66OPPlLNmjX1zjvvaPz48ZKK7/UHwP7ozdeH3nx9bqY3Ozs76+mnn9YPP/xghim5gfqVofaRI0fM4Dn3uqpt2rSRpOvuxbl+++23fI/rmjVr5lm2Z88ePfbYY/Ly8pKnp6eqVKlifrFZYbebu22rbdWuXVt//PGHzp8/b7P8Zs4hevbsqR9++EFnz57VihUr9Mwzz2j79u3q2rWrecm3n3/+WXv27Mnz+v3HP/4hid4O3Ek4P7g+nB9cn1vxu/upU6eUlpamOXPm5Dmucv+wkXtcvfzyyypXrpweeOAB3XfffYqMjDQv23Mz6K23J66JjmL3v//9T3369FF4eLiGDx8uHx8fOTk5acKECeaXGRZGbmP797//rdDQ0Hxrrg4Sb9bvv/+u9PT0Atfr7u6udevWafXq1Vq6dKni4uK0YMECtW/fXitWrJCTk9M1t1OYa6FdL6trW2dnZ1/XnOzBajvGVV9kcrPrs5f8fg72OI5v5nmw93N4LYGBgXrhhRf02GOPqVq1apo/f77NF64V5esPgP3Rm+nNJa03P/vss3rnnXf02Wef6d///rc+++wz1alTx7z+bnZ2th5++GGdOXNGL7/8smrVqiUPDw8dO3ZMffr0ueF3El5LWlqa2rRpI09PT40bN07Vq1eXm5ubtm3bppdffvmWbfdq9vh5eXp66uGHH9bDDz+sMmXKaN68edq0aZPatGmjnJwc1a9fX1OmTMn3sQEBATc0bwClC+cHnB+UtPOD/OQeV88++6x69+6db02DBg0k/f3HggMHDmjJkiWKi4vTl19+qXfffVfR0dEaO3bsTc+F3np7IURHsfviiy9UrVo1ffXVVzZNIfcvz1f7+eef8yz76aefzC9YqlatmiSpTJkyCgkJsf+E8/Hpp59KkmXjz+Xo6KgOHTqoQ4cOmjJlit588029+uqrWr16tUJCQgr8ssYbcfVzZRiGDh48aDYM6e+/8qalpeV57G+//WY+l5J1w85PYGCgVq5cqT///NPmL9q5H2nO/ZLK20lhj+OilvucHzx40Oad9KdPn76pv/JXqFBB1atX1+7duyUV7vVn7+MdgP3Qm+nNJU3z5s1VvXp1xcbG6uGHH9aePXv0xhtvmOO7du3STz/9pHnz5un55583l8fHx9/Q9gIDA/M9rg8cOGBzf82aNTp9+rS++uortW7d2lx+6NChPI+93p9X7s/i6m1Jf/+8KleuLA8Pj+ta141q2rSp5s2bpxMnTkj6+wvMduzYoQ4dOlxzP+jvwO2L8wPOD0qDKlWqqHz58srOzr6u48rDw0NPP/20nn76aWVlZenxxx/XG2+8oZEjR8rNzc1uP2t6a+nH5VxQ7HL/8njlXy43bdqkxMTEfOsXL15sc120zZs3a9OmTercubOkv78NuW3btnr//ffNf5yudOrUKXtOX6tWrdL48eMVFBSU73Uyc505cybPstx3T+VeTzr3F6L8GuONyP029FxffPGFTpw4YT5X0t//cG/cuFFZWVnmsiVLlujo0aM26yrM3Lp06aLs7Gy98847NsunTp0qBwcHm+3fLgp7HBe1Dh06yNnZWe+9957N8qt/RlZ27NihP/74I8/y3377TXv37jU/cl6Y15+9j3cA9kNvpjeXRL169dL27ds1evRoOTg46JlnnjHH8jtmDcPQ9OnTb2hbXbp00caNG7V582Zz2alTpzR//nybuvy2m5WVpXfffTfPOj08PK7r8i5Vq1ZVo0aNNG/ePJuf7e7du7VixQp16dKlsLuTr7/++svyNZ17Ddzc/v7UU0/p2LFj+uCDD/LUXrhwwebyMh4eHvR24DbF+QHnB6WBk5OTunfvri+//NJ8s9eVrjyuTp8+bTPm4uKiOnXqyDAMXbp0SVLhnk966+2Nd6KjSHz88ceKi4vLs3zw4MF65JFH9NVXX+mxxx5TWFiYDh06pNmzZ6tOnTo6d+5cnsfUqFFDDz30kAYOHKjMzExNmzZNlSpV0ogRI8yaWbNm6aGHHlL9+vXVr18/VatWTampqUpMTNTvv/+uHTt23NB+fP/999q/f78uX76s1NRUrVq1SvHx8QoMDNS3334rNzc3y8eOGzdO69atU1hYmAIDA3Xy5Em9++67uvvuu/XQQw9J+rspent7a/bs2Spfvrw8PDzUvHlzy2twX0vFihX10EMPqW/fvkpNTdW0adNUo0YN9evXz6x58cUX9cUXX6hTp0566qmn9Msvv+h///ufzZeFFHZuXbt2Vbt27fTqq6/q8OHDatiwoVasWKFvvvlGQ4YMybPu20Fhj+Oi5uvrq8GDB2vy5Ml69NFH1alTJ+3YsUPff/+9KleufM2/bMfHx2v06NF69NFH1aJFC5UrV06//vqrPv74Y2VmZmrMmDFm7fW+/ho1aiQnJye99dZbSk9Pl6urq9q3by8fH59b+VQA+P/Rm+nNpa03P/vssxo3bpy++eYbtWzZ0nwnoyTVqlVL1atX17///W8dO3ZMnp6e+vLLL2/401YjRozQp59+qk6dOmnw4MHy8PDQnDlzFBgYqJ07d5p1Dz74oCpUqKDevXvrpZdekoODgz799NN8P9bepEkTLViwQMOGDVOzZs1Urlw5de3aNd/tT5o0SZ07d1ZwcLAiIiJ04cIFzZw5U15eXjY992b89ddfevDBB9WiRQt16tRJAQEBSktL0+LFi/XDDz8oPDxcjRs3liQ999xzWrhwoQYMGKDVq1erZcuWys7O1v79+7Vw4UItX75cTZs2Nfdz5cqVmjJlivz9/RUUFKTmzZvbZc4Abj3ODzg/KG3nB/n573//q9WrV6t58+bq16+f6tSpozNnzmjbtm1auXKl+YeSjh07ys/PTy1btpSvr6/27dund955R2FhYeY785s0aSJJevXVV9WjRw+VKVNGXbt2zfdTYfTW25wB3EJz5841JFnejh49auTk5BhvvvmmERgYaLi6uhqNGzc2lixZYvTu3dsIDAw013Xo0CFDkjFp0iRj8uTJRkBAgOHq6mq0atXK2LFjR55t//LLL8bzzz9v+Pn5GWXKlDHuuusu45FHHjG++OILs2b16tWGJGP16tWF2g8XFxfDz8/PePjhh43p06cbGRkZeR4zevRo48qXWEJCgtGtWzfD39/fcHFxMfz9/Y2ePXsaP/30k83jvvnmG6NOnTqGs7OzIcmYO3euYRiG0aZNG6Nu3br5zq9NmzZGmzZt8uzXZ599ZowcOdLw8fEx3N3djbCwMOO3337L8/jJkycbd911l+Hq6mq0bNnS2Lp1a551FjS3q39WhmEYf/75pzF06FDD39/fKFOmjHHfffcZkyZNMnJycmzqJBmRkZF55hQYGGj07t073/3NlXtM5M4jdy4eHh55aq/+eVwPDw8PmzlceQxe7XqPY8P4e59Hjx6dZ26nTp2yqcs97g4dOmQuu/p5ya3ZsmWLzWPzO7YvX75svPbaa4afn5/h7u5utG/f3ti3b59RqVIlY8CAAQU+F7/++qsRHR1ttGjRwvDx8TGcnZ2NKlWqGGFhYcaqVavy1F/P688wDOODDz4wqlWrZjg5OV3XaxHAzaM305tLc29u1qyZIcl4991384zt3bvXCAkJMcqVK2dUrlzZ6Nevn7Fjx44888lvu/nt286dO402bdoYbm5uxl133WWMHz/e+Oijj/L05vXr1xstWrQw3N3dDX9/f2PEiBHG8uXL8xzH586dM5555hnD29vbkGT+fPJ7zgzDMFauXGm0bNnScHd3Nzw9PY2uXbsae/futakpzDnE1S5dumR88MEHRnh4uPlaL1u2rNG4cWNj0qRJRmZmpk19VlaW8dZbbxl169Y1XF1djQoVKhhNmjQxxo4da6Snp5t1+/fvN1q3bm24u7sbkq55zAAoGTg/4PygNJ8fXP27u2EYRmpqqhEZGWkEBAQYZcqUMfz8/IwOHToYc+bMMWvef/99o3Xr1kalSpUMV1dXo3r16sbw4cNt+pphGMb48eONu+66y3B0dCywv9Jbb28OhnGLvnUOAFDipaWlqUKFCnr99df16quvFvd0AAAAAAAAShyuiQ4Ad4gLFy7kWTZt2jRJUtu2bYt2MgAAAAAAAKUE10QHgDvEggULFBMToy5duqhcuXL68ccf9dlnn6ljx45q2bJlcU8PAAAAAACgRCJEB4A7RIMGDeTs7KyJEycqIyPD/LLR119/vbinBgAAAAAAUGJxTXQAAAAAAAAAACxwTXQAAAAAAAAAACwQogMAAAAAAAAAYIFrottJTk6Ojh8/rvLly8vBwaG4pwMAuA0YhqE///xT/v7+cnTk7972Ru8GANgbvfvWoncDAOztens3IbqdHD9+XAEBAcU9DQDAbejo0aO6++67i3satx16NwDgVqF33xr0bgDArXKt3k2Ibifly5eX9PcT7unpWcyzAQDcDjIyMhQQEGD2GNgXvRsAYG/07luL3g0AsLfr7d2E6HaS+1EyT09PmjkAwK74uPKtQe8GANwq9O5bg94NALhVrtW7uUgbAAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwUa4j+3nvvqUGDBvL09JSnp6eCg4P1/fffm+MXL15UZGSkKlWqpHLlyql79+5KTU21WceRI0cUFhamsmXLysfHR8OHD9fly5dtatasWaP7779frq6uqlGjhmJiYvLMZdasWbr33nvl5uam5s2ba/PmzbdknwEAKM3WrVunrl27yt/fXw4ODlq8eLHNuGEYio6OVtWqVeXu7q6QkBD9/PPPNjVnzpxRr1695OnpKW9vb0VEROjcuXM2NTt37lSrVq3k5uamgIAATZw4Mc9cFi1apFq1asnNzU3169fXsmXL7L6/AACUdvRuAABuXrGG6Hfffbf++9//KikpSVu3blX79u3VrVs37dmzR5I0dOhQfffdd1q0aJHWrl2r48eP6/HHHzcfn52drbCwMGVlZWnDhg2aN2+eYmJiFB0dbdYcOnRIYWFhateunZKTkzVkyBC9+OKLWr58uVmzYMECDRs2TKNHj9a2bdvUsGFDhYaG6uTJk0X3ZAAAUAqcP39eDRs21KxZs/IdnzhxombMmKHZs2dr06ZN8vDwUGhoqC5evGjW9OrVS3v27FF8fLyWLFmidevWqX///uZ4RkaGOnbsqMDAQCUlJWnSpEkaM2aM5syZY9Zs2LBBPXv2VEREhLZv367w8HCFh4dr9+7dt27nAQAohejdAADYgVHCVKhQwfjwww+NtLQ0o0yZMsaiRYvMsX379hmSjMTERMMwDGPZsmWGo6OjkZKSYta89957hqenp5GZmWkYhmGMGDHCqFu3rs02nn76aSM0NNS8/8ADDxiRkZHm/ezsbMPf39+YMGHCdc87PT3dkGSkp6cXbocBALBQ0nuLJOPrr7827+fk5Bh+fn7GpEmTzGVpaWmGq6ur8dlnnxmGYRh79+41JBlbtmwxa77//nvDwcHBOHbsmGEYhvHuu+8aFSpUMHu5YRjGyy+/bNSsWdO8/9RTTxlhYWE282nevLnxz3/+87rnX9KfXwBA6VPSewu9GwAAW9fbW0rMNdGzs7P1+eef6/z58woODlZSUpIuXbqkkJAQs6ZWrVq65557lJiYKElKTExU/fr15evra9aEhoYqIyPDfDd7YmKizTpya3LXkZWVpaSkJJsaR0dHhYSEmDUAAODaDh06pJSUFJue6uXlpebNm9v0bm9vbzVt2tSsCQkJkaOjozZt2mTWtG7dWi4uLmZNaGioDhw4oLNnz5o1BfV3AABwbfRuAACuj3NxT2DXrl0KDg7WxYsXVa5cOX399deqU6eOkpOT5eLiIm9vb5t6X19fpaSkSJJSUlJsAvTc8dyxgmoyMjJ04cIFnT17VtnZ2fnW7N+/33LemZmZyszMNO9nZGQUbscBALjN5Pbe/HrqlX3Zx8fHZtzZ2VkVK1a0qQkKCsqzjtyxChUqWPb33HXkh94NAIAtejcAANen2EP0mjVrKjk5Wenp6friiy/Uu3dvrV27trindU0TJkzQ2LFj7bKuJsM/sct6UDSSJj1fZNs6Mq5+kW0L9nFP9K4i21bLmS2LbFu4eesHrS/uKdzx6N13Lno3CkLvhhV6d/Gjd9+56N0oCL0bVm5l7y72y7m4uLioRo0aatKkiSZMmKCGDRtq+vTp8vPzU1ZWltLS0mzqU1NT5efnJ0ny8/NTampqnvHcsYJqPD095e7ursqVK8vJySnfmtx15GfkyJFKT083b0ePHr2h/QcA4HaR2zcL6ql+fn55vrj78uXLOnPmjF36O70bAIDrR+8GAOD6FHuIfrWcnBxlZmaqSZMmKlOmjBISEsyxAwcO6MiRIwoODpYkBQcHa9euXTYNPT4+Xp6enqpTp45Zc+U6cmty1+Hi4qImTZrY1OTk5CghIcGsyY+rq6s8PT1tbgAA3MmCgoLk5+dn01MzMjK0adMmm96dlpampKQks2bVqlXKyclR8+bNzZp169bp0qVLZk18fLxq1qypChUqmDUF9ff80LsBALBF7wYA4PoUa4g+cuRIrVu3TocPH9auXbs0cuRIrVmzRr169ZKXl5ciIiI0bNgwrV69WklJSerbt6+Cg4PVokULSVLHjh1Vp04dPffcc9qxY4eWL1+uUaNGKTIyUq6urpKkAQMG6Ndff9WIESO0f/9+vfvuu1q4cKGGDh1qzmPYsGH64IMPNG/ePO3bt08DBw7U+fPn1bdv32J5XgAAKKnOnTun5ORkJScnS/r7C8mSk5N15MgROTg4aMiQIXr99df17bffateuXXr++efl7++v8PBwSVLt2rXVqVMn9evXT5s3b9b69esVFRWlHj16yN/fX5L0zDPPyMXFRREREdqzZ48WLFig6dOna9iwYeY8Bg8erLi4OE2ePFn79+/XmDFjtHXrVkVFRRX1UwIAQIlG7wYA4OYV6zXRT548qeeff14nTpyQl5eXGjRooOXLl+vhhx+WJE2dOlWOjo7q3r27MjMzFRoaqnfffdd8vJOTk5YsWaKBAwcqODhYHh4e6t27t8aNG2fWBAUFaenSpRo6dKimT5+uu+++Wx9++KFCQ0PNmqefflqnTp1SdHS0UlJS1KhRI8XFxeX50hMAAO50W7duVbt27cz7ub8c9+7dWzExMRoxYoTOnz+v/v37Ky0tTQ899JDi4uLk5uZmPmb+/PmKiopShw4dzD4/Y8YMc9zLy0srVqxQZGSkmjRposqVKys6Olr9+/c3ax588EHFxsZq1KhR+s9//qP77rtPixcvVr169YrgWQAAoPSgdwMAcPMcDMMwinsSt4OMjAx5eXkpPT290B8x4wtOShe+4AQF4QtOYOVGvuDkZnoLro3efeegd6Mg9G5YoXeXPPTuOwe9GwWhd8PKrezdJe6a6AAAAAAAAAAAlBSE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBQrCH6hAkT1KxZM5UvX14+Pj4KDw/XgQMHbGratm0rBwcHm9uAAQNsao4cOaKwsDCVLVtWPj4+Gj58uC5fvmxTs2bNGt1///1ydXVVjRo1FBMTk2c+s2bN0r333is3Nzc1b95cmzdvtvs+AwAAAAAAAABKj2IN0deuXavIyEht3LhR8fHxunTpkjp27Kjz58/b1PXr108nTpwwbxMnTjTHsrOzFRYWpqysLG3YsEHz5s1TTEyMoqOjzZpDhw4pLCxM7dq1U3JysoYMGaIXX3xRy5cvN2sWLFigYcOGafTo0dq2bZsaNmyo0NBQnTx58tY/EQAAAAAAAACAEsm5ODceFxdncz8mJkY+Pj5KSkpS69atzeVly5aVn59fvutYsWKF9u7dq5UrV8rX11eNGjXS+PHj9fLLL2vMmDFycXHR7NmzFRQUpMmTJ0uSateurR9//FFTp05VaGioJGnKlCnq16+f+vbtK0maPXu2li5dqo8//livvPLKrdh9AAAAAAAAAEAJV6KuiZ6eni5Jqlixos3y+fPnq3LlyqpXr55Gjhypv/76yxxLTExU/fr15evray4LDQ1VRkaG9uzZY9aEhITYrDM0NFSJiYmSpKysLCUlJdnUODo6KiQkxKwBAAAAAAAAANx5ivWd6FfKycnRkCFD1LJlS9WrV89c/swzzygwMFD+/v7auXOnXn75ZR04cEBfffWVJCklJcUmQJdk3k9JSSmwJiMjQxcuXNDZs2eVnZ2db83+/fvznW9mZqYyMzPN+xkZGTe45wAAAAAAAACAkqrEhOiRkZHavXu3fvzxR5vl/fv3N/+/fv36qlq1qjp06KBffvlF1atXL+ppmiZMmKCxY8cW2/YBAAAAAAAAALdeibicS1RUlJYsWaLVq1fr7rvvLrC2efPmkqSDBw9Kkvz8/JSammpTk3s/9zrqVjWenp5yd3dX5cqV5eTklG+N1bXYR44cqfT0dPN29OjR69xbAAAAAAAAAEBpUawhumEYioqK0tdff61Vq1YpKCjomo9JTk6WJFWtWlWSFBwcrF27dunkyZNmTXx8vDw9PVWnTh2zJiEhwWY98fHxCg4OliS5uLioSZMmNjU5OTlKSEgwa67m6uoqT09PmxsAAAAAAAAA4PZSrJdziYyMVGxsrL755huVL1/evIa5l5eX3N3d9csvvyg2NlZdunRRpUqVtHPnTg0dOlStW7dWgwYNJEkdO3ZUnTp19Nxzz2nixIlKSUnRqFGjFBkZKVdXV0nSgAED9M4772jEiBF64YUXtGrVKi1cuFBLly415zJs2DD17t1bTZs21QMPPKBp06bp/Pnz6tu3b9E/MQAAAAAAAACAEqFYQ/T33ntPktS2bVub5XPnzlWfPn3k4uKilStXmoF2QECAunfvrlGjRpm1Tk5OWrJkiQYOHKjg4GB5eHiod+/eGjdunFkTFBSkpUuXaujQoZo+fbruvvtuffjhhwoNDTVrnn76aZ06dUrR0dFKSUlRo0aNFBcXl+fLRgEAAAAAAAAAd45iDdENwyhwPCAgQGvXrr3megIDA7Vs2bICa9q2bavt27cXWBMVFaWoqKhrbg8AAAAAAAAAcGcoEV8sCgAAAAAAAABASUSIDgAAAAAAAACABUJ0AABgN9nZ2XrttdcUFBQkd3d3Va9eXePHj7e5hJthGIqOjlbVqlXl7u6ukJAQ/fzzzzbrOXPmjHr16iVPT095e3srIiJC586ds6nZuXOnWrVqJTc3NwUEBGjixIlFso8AANxO6N0AAFwbIToAALCbt956S++9957eeecd7du3T2+99ZYmTpyomTNnmjUTJ07UjBkzNHv2bG3atEkeHh4KDQ3VxYsXzZpevXppz549io+P15IlS7Ru3Tr179/fHM/IyFDHjh0VGBiopKQkTZo0SWPGjNGcOXOKdH8BACjt6N0AAFxbsX6xKAAAuL1s2LBB3bp1U1hYmCTp3nvv1WeffabNmzdL+vudbNOmTdOoUaPUrVs3SdInn3wiX19fLV68WD169NC+ffsUFxenLVu2qGnTppKkmTNnqkuXLnr77bfl7++v+fPnKysrSx9//LFcXFxUt25dJScna8qUKTa/sAMAgILRuwEAuDbeiQ4AAOzmwQcfVEJCgn766SdJ0o4dO/Tjjz+qc+fOkqRDhw4pJSVFISEh5mO8vLzUvHlzJSYmSpISExPl7e1t/hIuSSEhIXJ0dNSmTZvMmtatW8vFxcWsCQ0N1YEDB3T27Nlbvp8AANwu6N0AAFwb70QHAAB288orrygjI0O1atWSk5OTsrOz9cYbb6hXr16SpJSUFEmSr6+vzeN8fX3NsZSUFPn4+NiMOzs7q2LFijY1QUFBedaRO1ahQoU8c8vMzFRmZqZ5PyMj42Z2FQCA2wK9GwCAa+Od6AAAwG4WLlyo+fPnKzY2Vtu2bdO8efP09ttva968ecU9NU2YMEFeXl7mLSAgoLinBABAsaN3AwBwbYToAADAboYPH65XXnlFPXr0UP369fXcc89p6NChmjBhgiTJz89PkpSammrzuNTUVHPMz89PJ0+etBm/fPmyzpw5Y1OT3zqu3MbVRo4cqfT0dPN29OjRm9xbAABKP3o3AADXRogOAADs5q+//pKjo+3phZOTk3JyciRJQUFB8vPzU0JCgjmekZGhTZs2KTg4WJIUHBystLQ0JSUlmTWrVq1STk6OmjdvbtasW7dOly5dMmvi4+NVs2bNfD8OLkmurq7y9PS0uQEAcKejdwMAcG2E6AAAwG66du2qN954Q0uXLtXhw4f19ddfa8qUKXrsscckSQ4ODhoyZIhef/11ffvtt9q1a5eef/55+fv7Kzw8XJJUu3ZtderUSf369dPmzZu1fv16RUVFqUePHvL395ckPfPMM3JxcVFERIT27NmjBQsWaPr06Ro2bFhx7ToAAKUSvRsAgGvji0UBAIDdzJw5U6+99pr+9a9/6eTJk/L399c///lPRUdHmzUjRozQ+fPn1b9/f6Wlpemhhx5SXFyc3NzczJr58+crKipKHTp0kKOjo7p3764ZM2aY415eXlqxYoUiIyPVpEkTVa5cWdHR0erfv3+R7i8AAKUdvRsAgGsjRAcAAHZTvnx5TZs2TdOmTbOscXBw0Lhx4zRu3DjLmooVKyo2NrbAbTVo0EA//PDDjU4VAACI3g0AwPXgci4AAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAvFGqJPmDBBzZo1U/ny5eXj46Pw8HAdOHDApubixYuKjIxUpUqVVK5cOXXv3l2pqak2NUeOHFFYWJjKli0rHx8fDR8+XJcvX7apWbNmje6//365urqqRo0aiomJyTOfWbNm6d5775Wbm5uaN2+uzZs3232fAQAAAAAAAAClR7GG6GvXrlVkZKQ2btyo+Ph4Xbp0SR07dtT58+fNmqFDh+q7777TokWLtHbtWh0/flyPP/64OZ6dna2wsDBlZWVpw4YNmjdvnmJiYhQdHW3WHDp0SGFhYWrXrp2Sk5M1ZMgQvfjii1q+fLlZs2DBAg0bNkyjR4/Wtm3b1LBhQ4WGhurkyZNF82QAAAAAAAAAAEoc5+LceFxcnM39mJgY+fj4KCkpSa1bt1Z6ero++ugjxcbGqn379pKkuXPnqnbt2tq4caNatGihFStWaO/evVq5cqV8fX3VqFEjjR8/Xi+//LLGjBkjFxcXzZ49W0FBQZo8ebIkqXbt2vrxxx81depUhYaGSpKmTJmifv36qW/fvpKk2bNna+nSpfr444/1yiuvFOGzAgAAAAAAAAAoKUrUNdHT09MlSRUrVpQkJSUl6dKlSwoJCTFratWqpXvuuUeJiYmSpMTERNWvX1++vr5mTWhoqDIyMrRnzx6z5sp15NbkriMrK0tJSUk2NY6OjgoJCTFrAAAAAAAAAAB3nmJ9J/qVcnJyNGTIELVs2VL16tWTJKWkpMjFxUXe3t42tb6+vkpJSTFrrgzQc8dzxwqqycjI0IULF3T27FllZ2fnW7N///5855uZmanMzEzzfkZGRiH3GAAAAAAAAABQ0pWYd6JHRkZq9+7d+vzzz4t7KtdlwoQJ8vLyMm8BAQHFPSUAAAAAAAAAgJ2ViBA9KipKS5Ys0erVq3X33Xeby/38/JSVlaW0tDSb+tTUVPn5+Zk1qampecZzxwqq8fT0lLu7uypXriwnJ6d8a3LXcbWRI0cqPT3dvB09erTwOw4AAAAAAAAAKNGKNUQ3DENRUVH6+uuvtWrVKgUFBdmMN2nSRGXKlFFCQoK57MCBAzpy5IiCg4MlScHBwdq1a5dOnjxp1sTHx8vT01N16tQxa65cR25N7jpcXFzUpEkTm5qcnBwlJCSYNVdzdXWVp6enzQ0AAAAAAAAAcHsp1muiR0ZGKjY2Vt98843Kly9vXsPcy8tL7u7u8vLyUkREhIYNG6aKFSvK09NTgwYNUnBwsFq0aCFJ6tixo+rUqaPnnntOEydOVEpKikaNGqXIyEi5urpKkgYMGKB33nlHI0aM0AsvvKBVq1Zp4cKFWrp0qTmXYcOGqXfv3mratKkeeOABTZs2TefPn1ffvn2L/okBAAAAAAAAAJQIxRqiv/fee5Kktm3b2iyfO3eu+vTpI0maOnWqHB0d1b17d2VmZio0NFTvvvuuWevk5KQlS5Zo4MCBCg4OloeHh3r37q1x48aZNUFBQVq6dKmGDh2q6dOn6+6779aHH36o0NBQs+bpp5/WqVOnFB0drZSUFDVq1EhxcXF5vmwUAAAAAAAAAHDnKNYQ3TCMa9a4ublp1qxZmjVrlmVNYGCgli1bVuB62rZtq+3btxdYExUVpaioqGvOCQAAAAAAAABwZygRXywKAAAAAAAAAEBJRIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AACwq2PHjunZZ59VpUqV5O7urvr162vr1q3muGEYio6OVtWqVeXu7q6QkBD9/PPPNus4c+aMevXqJU9PT3l7eysiIkLnzp2zqdm5c6datWolNzc3BQQEaOLEiUWyfwAA3G7o3QAAFIwQHQAA2M3Zs2fVsmVLlSlTRt9//7327t2ryZMnq0KFCmbNxIkTNWPGDM2ePVubNm2Sh4eHQkNDdfHiRbOmV69e2rNnj+Lj47VkyRKtW7dO/fv3N8czMjLUsWNHBQYGKikpSZMmTdKYMWM0Z86cIt1fAABKO3o3AADX5lzcEwAAALePt956SwEBAZo7d665LCgoyPx/wzA0bdo0jRo1St26dZMkffLJJ/L19dXixYvVo0cP7du3T3FxcdqyZYuaNm0qSZo5c6a6dOmit99+W/7+/po/f76ysrL08ccfy8XFRXXr1lVycrKmTJli8ws7AAAoGL0bAIBr453oAADAbr799ls1bdpUTz75pHx8fNS4cWN98MEH5vihQ4eUkpKikJAQc5mXl5eaN2+uxMRESVJiYqK8vb3NX8IlKSQkRI6Ojtq0aZNZ07p1a7m4uJg1oaGhOnDggM6ePXurdxMAgNsGvRsAgGsjRAcAAHbz66+/6r333tN9992n5cuXa+DAgXrppZc0b948SVJKSookydfX1+Zxvr6+5lhKSop8fHxsxp2dnVWxYkWbmvzWceU2rpaZmamMjAybGwAAdzp6NwAA18blXAAAgN3k5OSoadOmevPNNyVJjRs31u7duzV79mz17t27WOc2YcIEjR07tljnAABASUPvBgDg2ngnOgAAsJuqVauqTp06Nstq166tI0eOSJL8/PwkSampqTY1qamp5pifn59OnjxpM3758mWdOXPGpia/dVy5jauNHDlS6enp5u3o0aM3sosAANxW6N0AAFwbIToAALCbli1b6sCBAzbLfvrpJwUGBkr6+4vK/Pz8lJCQYI5nZGRo06ZNCg4OliQFBwcrLS1NSUlJZs2qVauUk5Oj5s2bmzXr1q3TpUuXzJr4+HjVrFlTFSpUyHdurq6u8vT0tLkBAHCno3cDAHBthOgAAMBuhg4dqo0bN+rNN9/UwYMHFRsbqzlz5igyMlKS5ODgoCFDhuj111/Xt99+q127dun555+Xv7+/wsPDJf397rdOnTqpX79+2rx5s9avX6+oqCj16NFD/v7+kqRnnnlGLi4uioiI0J49e7RgwQJNnz5dw4YNK65dBwCgVKJ3AwBwbVwTHQAA2E2zZs309ddfa+TIkRo3bpyCgoI0bdo09erVy6wZMWKEzp8/r/79+ystLU0PPfSQ4uLi5ObmZtbMnz9fUVFR6tChgxwdHdW9e3fNmDHDHPfy8tKKFSsUGRmpJk2aqHLlyoqOjlb//v2LdH8BACjt6N0AAFwbIToAALCrRx55RI888ojluIODg8aNG6dx48ZZ1lSsWFGxsbEFbqdBgwb64YcfbnieAADgb/RuAAAKxuVcAAAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYOGGQvRq1arp9OnTeZanpaWpWrVqNz0pAABgX/RuAABKF3o3AAAlxw2F6IcPH1Z2dnae5ZmZmTp27NhNTwoAANgXvRsAgNKF3g0AQMnhXJjib7/91vz/5cuXy8vLy7yfnZ2thIQE3XvvvXabHAAAuDn0bgAAShd6NwAAJU+hQvTw8HBJkoODg3r37m0zVqZMGd17772aPHmy3SYHAABuDr0bAIDShd4NAEDJU6gQPScnR5IUFBSkLVu2qHLlyrdkUgAAwD7o3QAAlC70bgAASp5Chei5Dh06ZO95AACAW4jeDQBA6ULvBgCg5LihEF2SEhISlJCQoJMnT5p/Kc/18ccf3/TEAACAfdG7AQAoXejdAACUDDcUoo8dO1bjxo1T06ZNVbVqVTk4ONh7XgAAwI7o3QAAlC70bgAASo4bCtFnz56tmJgYPffcc/aeDwAAuAXo3QAAlC70bgAASg7HG3lQVlaWHnzwQXvPBQAA3CL0bgAAShd6NwAAJccNhegvvviiYmNj7T0XAABwi9C7AQAoXejdAACUHDd0OZeLFy9qzpw5WrlypRo0aKAyZcrYjE+ZMsUukwMAAPZB7wYAoHShdwMAUHLcUIi+c+dONWrUSJK0e/dumzG+7AQAgJKH3g0AQOlC7wYAoOS4oRB99erV9p4HAAC4hejdAACULvRuAABKjhu6JjoAAAAAAAAAAHeCG3onert27Qr8+NiqVatueEIAAMD+6N0AAJQu9G4AAEqOGwrRc6/LluvSpUtKTk7W7t271bt3b3vMCwAA2BG9GwCA0oXeDQBAyXFDIfrUqVPzXT5mzBidO3fupiYEAADsj94NAEDpQu8GAKDksOs10Z999ll9/PHH9lwlAAC4hejdAACULvRuAACKnl1D9MTERLm5udlzlQAA4BaidwMAULrQuwEAKHo3dDmXxx9/3Oa+YRg6ceKEtm7dqtdee80uEwMAAPZD7wYAoHShdwMAUHLcUIju5eVlc9/R0VE1a9bUuHHj1LFjR7tMDAAA2A+9GwCA0oXeDQBAyXFDIfrcuXPtPQ8AAHAL0bsBAChd6N0AAJQcNxSi50pKStK+ffskSXXr1lXjxo3tMikAAHBr0LsBAChd6N0AABS/G/pi0ZMnT6p9+/Zq1qyZXnrpJb300ktq0qSJOnTooFOnTl33etatW6euXbvK399fDg4OWrx4sc14nz595ODgYHPr1KmTTc2ZM2fUq1cveXp6ytvbWxERETp37pxNzc6dO9WqVSu5ubkpICBAEydOzDOXRYsWqVatWnJzc1P9+vW1bNmy639CAAAo4ezVuwEAQNGgdwMAUHLcUIg+aNAg/fnnn9qzZ4/OnDmjM2fOaPfu3crIyNBLL7103es5f/68GjZsqFmzZlnWdOrUSSdOnDBvn332mc14r169tGfPHsXHx2vJkiVat26d+vfvb45nZGSoY8eOCgwMVFJSkiZNmqQxY8Zozpw5Zs2GDRvUs2dPRUREaPv27QoPD1d4eLh2795diGcFAICSy169GwAAFA16NwAAJccNXc4lLi5OK1euVO3atc1lderU0axZswr1BSedO3dW586dC6xxdXWVn59fvmP79u1TXFyctmzZoqZNm0qSZs6cqS5duujtt9+Wv7+/5s+fr6ysLH388cdycXFR3bp1lZycrClTpphh+/Tp09WpUycNHz5ckjR+/HjFx8frnXfe0ezZs697fwAAKKns1bsBAEDRoHcDAFBy3NA70XNyclSmTJk8y8uUKaOcnJybntSV1qxZIx8fH9WsWVMDBw7U6dOnzbHExER5e3ubAbokhYSEyNHRUZs2bTJrWrduLRcXF7MmNDRUBw4c0NmzZ82akJAQm+2GhoYqMTHRcl6ZmZnKyMiwuQEAUFIVZe8GAAA3j94NAEDJcUMhevv27TV48GAdP37cXHbs2DENHTpUHTp0sNvkOnXqpE8++UQJCQl66623tHbtWnXu3FnZ2dmSpJSUFPn4+Ng8xtnZWRUrVlRKSopZ4+vra1OTe/9aNbnj+ZkwYYK8vLzMW0BAwM3tLAAAt1BR9W4AAGAf9G4AAEqOGwrR33nnHWVkZOjee+9V9erVVb16dQUFBSkjI0MzZ8602+R69OihRx99VPXr11d4eLiWLFmiLVu2aM2aNXbbxo0aOXKk0tPTzdvRo0eLe0oAAFgqqt4NAADsg94NAEDJcUPXRA8ICNC2bdu0cuVK7d+/X5JUu3btPJdEsbdq1aqpcuXKOnjwoDp06CA/Pz+dPHnSpuby5cs6c+aMeR11Pz8/paam2tTk3r9WjdW12KW/r9Xu6up60/sEAEBRKK7eDQAAbgy9GwCAkqNQ70RftWqV6tSpo4yMDDk4OOjhhx/WoEGDNGjQIDVr1kx169bVDz/8cKvmqt9//12nT59W1apVJUnBwcFKS0tTUlKSzRxzcnLUvHlzs2bdunW6dOmSWRMfH6+aNWuqQoUKZk1CQoLNtuLj4xUcHHzL9gUAgKJQ3L0bAAAUDr0bAICSp1Ah+rRp09SvXz95enrmGfPy8tI///lPTZky5brXd+7cOSUnJys5OVmSdOjQISUnJ+vIkSM6d+6chg8fro0bN+rw4cNKSEhQt27dVKNGDYWGhkr6+6/wnTp1Ur9+/bR582atX79eUVFR6tGjh/z9/SVJzzzzjFxcXBQREaE9e/ZowYIFmj59uoYNG2bOY/DgwYqLi9PkyZO1f/9+jRkzRlu3blVUVFRhnh4AAEoce/duAABwa9G7AQAoeQoVou/YsUOdOnWyHO/YsaPNu8KvZevWrWrcuLEaN24sSRo2bJgaN26s6OhoOTk5aefOnXr00Uf1j3/8QxEREWrSpIl++OEHm8uozJ8/X7Vq1VKHDh3UpUsXPfTQQ5ozZ4457uXlpRUrVujQoUNq0qSJ/u///k/R0dHq37+/WfPggw8qNjZWc+bMUcOGDfXFF19o8eLFqlevXmGeHgAAShx7924AAHBr0bsBACh5CnVN9NTUVJUpU8Z6Zc7OOnXq1HWvr23btjIMw3J8+fLl11xHxYoVFRsbW2BNgwYNrvlxtyeffFJPPvnkNbcHAEBpYu/eDQAAbi16NwAAJU+h3ol+1113affu3ZbjO3fuNK9XDgAAih+9GwCA0oXeDQBAyVOoEL1Lly567bXXdPHixTxjFy5c0OjRo/XII4/YbXIAAODm0LsBAChd6N0AAJQ8hbqcy6hRo/TVV1/pH//4h6KiolSzZk1J0v79+zVr1ixlZ2fr1VdfvSUTBQAAhUfvBgCgdKF3AwBQ8hQqRPf19dWGDRs0cOBAjRw50ryeuYODg0JDQzVr1iz5+vrekokCAIDCo3cDAFC60LsBACh5ChWiS1JgYKCWLVums2fP6uDBgzIMQ/fdd58qVKhwK+YHAABuEr0bAIDShd4NAEDJUugQPVeFChXUrFkze84FAADcQvRuAABKF3o3AAAlQ6G+WBQAAAAAAAAAgDsJIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAgFvmv//9rxwcHDRkyBBz2cWLFxUZGalKlSqpXLly6t69u1JTU20ed+TIEYWFhals2bLy8fHR8OHDdfnyZZuaNWvW6P7775erq6tq1KihmJiYItgjAABub/RuAADyIkQHAAC3xJYtW/T++++rQYMGNsuHDh2q7777TosWLdLatWt1/PhxPf744+Z4dna2wsLClJWVpQ0bNmjevHmKiYlRdHS0WXPo0CGFhYWpXbt2Sk5O1pAhQ/Tiiy9q+fLlRbZ/AADcbujdAADkjxAdAADY3blz59SrVy998MEHqlChgrk8PT1dH330kaZMmaL27durSZMmmjt3rjZs2KCNGzdKklasWKG9e/fqf//7nxo1aqTOnTtr/PjxmjVrlrKysiRJs2fPVlBQkCZPnqzatWsrKipKTzzxhKZOnVos+wsAQGlH7wYAwBohOgAAsLvIyEiFhYUpJCTEZnlSUpIuXbpks7xWrVq65557lJiYKElKTExU/fr15evra9aEhoYqIyNDe/bsMWuuXndoaKi5DgAAUDj0bgAArDkX9wQAAMDt5fPPP9e2bdu0ZcuWPGMpKSlycXGRt7e3zXJfX1+lpKSYNVf+Ep47njtWUE1GRoYuXLggd3f3PNvOzMxUZmameT8jI6PwOwcAwG2I3g0AQMF4JzoAALCbo0ePavDgwZo/f77c3NyKezo2JkyYIC8vL/MWEBBQ3FMCAKDY0bsBALg2QnQAAGA3SUlJOnnypO6//345OzvL2dlZa9eu1YwZM+Ts7CxfX19lZWUpLS3N5nGpqany8/OTJPn5+Sk1NTXPeO5YQTWenp75vpNNkkaOHKn09HTzdvToUXvsMgAApRq9GwCAayNEBwAAdtOhQwft2rVLycnJ5q1p06bq1auX+f9lypRRQkKC+ZgDBw7oyJEjCg4OliQFBwdr165dOnnypFkTHx8vT09P1alTx6y5ch25NbnryI+rq6s8PT1tbgAA3Ono3QAAXBvXRAcAAHZTvnx51atXz2aZh4eHKlWqZC6PiIjQsGHDVLFiRXl6emrQoEEKDg5WixYtJEkdO3ZUnTp19Nxzz2nixIlKSUnRqFGjFBkZKVdXV0nSgAED9M4772jEiBF64YUXtGrVKi1cuFBLly4t2h0GAKCUo3cDAHBthOgAAKBITZ06VY6OjurevbsyMzMVGhqqd9991xx3cnLSkiVLNHDgQAUHB8vDw0O9e/fWuHHjzJqgoCAtXbpUQ4cO1fTp03X33Xfrww8/VGhoaHHsEgAAtzV6NwDgTkeIDgAAbqk1a9bY3Hdzc9OsWbM0a9Ysy8cEBgZq2bJlBa63bdu22r59uz2mCAAArkDvBgDAFtdEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC8Uaoq9bt05du3aVv7+/HBwctHjxYptxwzAUHR2tqlWryt3dXSEhIfr5559tas6cOaNevXrJ09NT3t7eioiI0Llz52xqdu7cqVatWsnNzU0BAQGaOHFinrksWrRItWrVkpubm+rXr69ly5bZfX8BAAAAAAAAAKVLsYbo58+fV8OGDTVr1qx8xydOnKgZM2Zo9uzZ2rRpkzw8PBQaGqqLFy+aNb169dKePXsUHx+vJUuWaN26derfv785npGRoY4dOyowMFBJSUmaNGmSxowZozlz5pg1GzZsUM+ePRUREaHt27crPDxc4eHh2r17963beQAAAAAAAABAiedcnBvv3LmzOnfunO+YYRiaNm2aRo0apW7dukmSPvnkE/n6+mrx4sXq0aOH9u3bp7i4OG3ZskVNmzaVJM2cOVNdunTR22+/LX9/f82fP19ZWVn6+OOP5eLiorp16yo5OVlTpkwxw/bp06erU6dOGj58uCRp/Pjxio+P1zvvvKPZs2cXwTMBAAAAAAAAACiJSuw10Q8dOqSUlBSFhISYy7y8vNS8eXMlJiZKkhITE+Xt7W0G6JIUEhIiR0dHbdq0yaxp3bq1XFxczJrQ0FAdOHBAZ8+eNWuu3E5uTe528pOZmamMjAybGwAAAAAAAADg9lJiQ/SUlBRJkq+vr81yX19fcywlJUU+Pj42487OzqpYsaJNTX7ruHIbVjW54/mZMGGCvLy8zFtAQEBhdxEAAAAAAAAAUMKV2BC9pBs5cqTS09PN29GjR4t7SgAAAAAAAAAAOyuxIbqfn58kKTU11WZ5amqqOebn56eTJ0/ajF++fFlnzpyxqclvHVduw6omdzw/rq6u8vT0tLkBAAAAAAAAAG4vJTZEDwoKkp+fnxISEsxlGRkZ2rRpk4KDgyVJwcHBSktLU1JSklmzatUq5eTkqHnz5mbNunXrdOnSJbMmPj5eNWvWVIUKFcyaK7eTW5O7HQAAAAAAAADAnalYQ/Rz584pOTlZycnJkv7+MtHk5GQdOXJEDg4OGjJkiF5//XV9++232rVrl55//nn5+/srPDxcklS7dm116tRJ/fr10+bNm7V+/XpFRUWpR48e8vf3lyQ988wzcnFxUUREhPbs2aMFCxZo+vTpGjZsmDmPwYMHKy4uTpMnT9b+/fs1ZswYbd26VVFRUUX9lAAAAAAAAAAAShDn4tz41q1b1a5dO/N+brDdu3dvxcTEaMSIETp//rz69++vtLQ0PfTQQ4qLi5Obm5v5mPnz5ysqKkodOnSQo6OjunfvrhkzZpjjXl5eWrFihSIjI9WkSRNVrlxZ0dHR6t+/v1nz4IMPKjY2VqNGjdJ//vMf3XfffVq8eLHq1atXBM8CAAAAAAAAAKCkKtYQvW3btjIMw3LcwcFB48aN07hx4yxrKlasqNjY2AK306BBA/3www8F1jz55JN68sknC54wAAAAAAAAAOCOUmKviQ4AAAAAAAAAQHEjRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAwG4mTJigZs2aqXz58vLx8VF4eLgOHDhgU3Px4kVFRkaqUqVKKleunLp3767U1FSbmiNHjigsLExly5aVj4+Phg8frsuXL9vUrFmzRvfff79cXV1Vo0YNxcTE3OrdAwDgtkPvBgDg2gjRAQCA3axdu1aRkZHauHGj4uPjdenSJXXs2FHnz583a4YOHarvvvtOixYt0tq1a3X8+HE9/vjj5nh2drbCwsKUlZWlDRs2aN68eYqJiVF0dLRZc+jQIYWFhaldu3ZKTk7WkCFD9OKLL2r58uVFur8AAJR29G4AAK7NubgnAAAAbh9xcXE292NiYuTj46OkpCS1bt1a6enp+uijjxQbG6v27dtLkubOnavatWtr48aNatGihVasWKG9e/dq5cqV8vX1VaNGjTR+/Hi9/PLLGjNmjFxcXDR79mwFBQVp8uTJkqTatWvrxx9/1NSpUxUaGlrk+w0AQGlF7wYA4Np4JzoAALhl0tPTJUkVK1aUJCUlJenSpUsKCQkxa2rVqqV77rlHiYmJkqTExETVr19fvr6+Zk1oaKgyMjK0Z88es+bKdeTW5K4DAADcGHo3AAB58U50AABwS+Tk5GjIkCFq2bKl6tWrJ0lKSUmRi4uLvL29bWp9fX2VkpJi1lz5S3jueO5YQTUZGRm6cOGC3N3d88wnMzNTmZmZ5v2MjIyb20EAAG4z9G4AAPLHO9EBAMAtERkZqd27d+vzzz8v7qlI+vuL07y8vMxbQEBAcU8JAIAShd4NAED+CNEBAIDdRUVFacmSJVq9erXuvvtuc7mfn5+ysrKUlpZmU5+amio/Pz+zJjU1Nc947lhBNZ6envm+k02SRo4cqfT0dPN29OjRm9pHAABuJ/RuAACsEaIDAAC7MQxDUVFR+vrrr7Vq1SoFBQXZjDdp0kRlypRRQkKCuezAgQM6cuSIgoODJUnBwcHatWuXTp48adbEx8fL09NTderUMWuuXEduTe468uPq6ipPT0+bGwAAdzp6NwAA18Y10QEAgN1ERkYqNjZW33zzjcqXL29eB9XLy0vu7u7y8vJSRESEhg0bpooVK8rT01ODBg1ScHCwWrRoIUnq2LGj6tSpo+eee04TJ05USkqKRo0apcjISLm6ukqSBgwYoHfeeUcjRozQCy+8oFWrVmnhwoVaunRpse07AAClEb0bAIBr453oAADAbt577z2lp6erbdu2qlq1qnlbsGCBWTN16lQ98sgj6t69u1q3bi0/Pz999dVX5riTk5OWLFkiJycnBQcH69lnn9Xzzz+vcePGmTVBQUFaunSp4uPj1bBhQ02ePFkffvihQkNDi3R/AQAo7ejdAABcG+9EBwAAdmMYxjVr3NzcNGvWLM2aNcuyJjAwUMuWLStwPW3bttX27dsLPUcAAPD/0LsBALg23okOAAAAAAAAAIAFQnQAAAAAAAAAACyU6BB9zJgxcnBwsLnVqlXLHL948aIiIyNVqVIllStXTt27d1dqaqrNOo4cOaKwsDCVLVtWPj4+Gj58uC5fvmxTs2bNGt1///1ydXVVjRo1FBMTUxS7BwAAAAAAAAAo4Up0iC5JdevW1YkTJ8zbjz/+aI4NHTpU3333nRYtWqS1a9fq+PHjevzxx83x7OxshYWFKSsrSxs2bNC8efMUExOj6Ohos+bQoUMKCwtTu3btlJycrCFDhujFF1/U8uXLi3Q/AQAAAAAAAAAlT4n/YlFnZ2f5+fnlWZ6enq6PPvpIsbGxat++vSRp7ty5ql27tjZu3KgWLVpoxYoV2rt3r1auXClfX181atRI48eP18svv6wxY8bIxcVFs2fPVlBQkCZPnixJql27tn788UdNnTqVbwkHAAAAAAAAgDtciX8n+s8//yx/f39Vq1ZNvXr10pEjRyRJSUlJunTpkkJCQszaWrVq6Z577lFiYqIkKTExUfXr15evr69ZExoaqoyMDO3Zs8esuXIduTW56wAAAAAAAAAA3LlK9DvRmzdvrpiYGNWsWVMnTpzQ2LFj1apVK+3evVspKSlycXGRt7e3zWN8fX2VkpIiSUpJSbEJ0HPHc8cKqsnIyNCFCxfk7u6e79wyMzOVmZlp3s/IyLipfQUAAAAAAAAAlDwlOkTv3Lmz+f8NGjRQ8+bNFRgYqIULF1qG20VlwoQJGjt2bLHOAQAAAAAAAABwa5X4y7lcydvbW//4xz908OBB+fn5KSsrS2lpaTY1qamp5jXU/fz8lJqammc8d6ygGk9PzwKD+pEjRyo9Pd28HT169GZ3DwAAAAAAAABQwpSqEP3cuXP65ZdfVLVqVTVp0kRlypRRQkKCOX7gwAEdOXJEwcHBkqTg4GDt2rVLJ0+eNGvi4+Pl6empOnXqmDVXriO3JncdVlxdXeXp6WlzAwAAAAAAAADcXkp0iP7vf/9ba9eu1eHDh7VhwwY99thjcnJyUs+ePeXl5aWIiAgNGzZMq1evVlJSkvr27avg4GC1aNFCktSxY0fVqVNHzz33nHbs2KHly5dr1KhRioyMlKurqyRpwIAB+vXXXzVixAjt379f7777rhYuXKihQ4cW564DAAAAAAAAAEqAEn1N9N9//109e/bU6dOnVaVKFT300EPauHGjqlSpIkmaOnWqHB0d1b17d2VmZio0NFTvvvuu+XgnJyctWbJEAwcOVHBwsDw8PNS7d2+NGzfOrAkKCtLSpUs1dOhQTZ8+XXfffbc+/PBDhYaGFvn+AgAAAAAAAABKlhIdon/++ecFjru5uWnWrFmaNWuWZU1gYKCWLVtW4Hratm2r7du339AcAQAAAAAAAAC3rxJ9ORcAAAAAAAAAAIoTIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKIDAAAAAAAAAGCBEB0AAAAAAAAAAAuE6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACwQIgOAAAAAAAAAIAFQnQAAAAAAAAAACwQogMAAAAAAAAAYIEQHQAAAAAAAAAAC4ToAAAAAAAAAABYIEQHAAAAAAAAAMACIToAAAAAAAAAABYI0QEAAAAAAAAAsECIDgAAAAAAAACABUJ0AAAAAAAAAAAsEKJfZdasWbr33nvl5uam5s2ba/PmzcU9JQAAUAB6NwAApQu9GwBQ2hCiX2HBggUaNmyYRo8erW3btqlhw4YKDQ3VyZMni3tqAAAgH/RuAABKF3o3AKA0IkS/wpQpU9SvXz/17dtXderU0ezZs1W2bFl9/PHHxT01AACQD3o3AAClC70bAFAaEaL//7KyspSUlKSQkBBzmaOjo0JCQpSYmFiMMwMAAPmhdwMAULrQuwEApZVzcU+gpPjjjz+UnZ0tX19fm+W+vr7av39/nvrMzExlZmaa99PT0yVJGRkZhd52duaFQj8GxedGfsY36s+L2UW2LdhHUR4fly9cLrJt4ebdyLGR+xjDMOw9ndsCvRvXi96NgtC7YYXebX/0blwvejcKQu+GlVvZuwnRb9CECRM0duzYPMsDAgKKYTYoSl4zBxT3FFCSTfAq7hmghPJ6+caPjT///FNeXhxbN4vefeeid6NA9G5YoHcXP3r3nYvejQLRu2HhVvZuQvT/X+XKleXk5KTU1FSb5ampqfLz88tTP3LkSA0bNsy8n5OTozNnzqhSpUpycHC45fMt6TIyMhQQEKCjR4/K09OzuKeDEobjA1Y4NmwZhqE///xT/v7+xT2VEonebV+8/lAQjg9Y4diwRe8uGL3bvnj9oSAcH7DCsWHrens3Ifr/z8XFRU2aNFFCQoLCw8Ml/d2gExISFBUVlafe1dVVrq6uNsu8vb2LYKali6enJy9IWOL4gBWOjf+Hd7FZo3ffGrz+UBCOD1jh2Ph/6N3W6N23Bq8/FITjA1Y4Nv6f6+ndhOhXGDZsmHr37q2mTZvqgQce0LRp03T+/Hn17du3uKcGAADyQe8GAKB0oXcDAEojQvQrPP300zp16pSio6OVkpKiRo0aKS4uLs+XngAAgJKB3g0AQOlC7wYAlEaE6FeJiorK92NkKBxXV1eNHj06z0fvAInjA9Y4NnAj6N32wesPBeH4gBWODdwIerd98PpDQTg+YIVj48Y4GIZhFPckAAAAAAAAAAAoiRyLewIAAAAAAAAAAJRUhOgAAAAAAAAAAFggREeRWrNmjRwcHJSWllbcU0EpEhMTI29v7+KeBkogBwcHLV68uLinAdyReP2hIPRucN4PlDz0bvBvM24U53aE6ACAUuzEiRPq3LmzJOnw4cNycHBQcnJy8U4KuEPw+gMAoHShdwPAjXMu7gkAQEEuXbpU3FNACZSVlSUXFxf5+fkV91SAOw6vP1wLvRsAShZ6N4Cbwbnd33gnOm5K27ZtNWjQIA0ZMkQVKlSQr6+vPvjgA50/f159+/ZV+fLlVaNGDX3//ffFPVXYyRdffKH69evL3d1dlSpVUkhIiM6fP68+ffooPDxcY8eOVZUqVeTp6akBAwYoKyvLfGxcXJweeugheXt7q1KlSnrkkUf0yy+/mOO574ZYsGCB2rRpIzc3N82fPz/PHE6dOqWmTZvqscceU2ZmZpHsN6x/9pL04Ycfqnbt2nJzc1OtWrX07rvv2jz2999/V8+ePVWxYkV5eHioadOm2rRpkySZx86VhgwZorZt25r327Ztq6ioKA0ZMkSVK1dWaGioJNuPpAYFBUmSGjduLAcHB7Vt21br1q1TmTJllJKSkmf9rVq1stdTA9xyvP5wM+jdsAfO+4HCoXejKPBv852Jc7viQYiOmzZv3jxVrlxZmzdv1qBBgzRw4EA9+eSTevDBB7Vt2zZ17NhRzz33nP7666/inipu0okTJ9SzZ0+98MIL2rdvn9asWaPHH39chmFIkhISEszln332mb766iuNHTvWfPz58+c1bNgwbd26VQkJCXJ0dNRjjz2mnJwcm+288sorGjx4sPbt22ee8OU6evSoWrVqpXr16umLL76Qq6vrrd9xFPiznz9/vqKjo/XGG29o3759evPNN/Xaa69p3rx5kqRz586pTZs2OnbsmL799lvt2LFDI0aMyPNzv5Z58+bJxcVF69ev1+zZs/OMb968WZK0cuVKnThxQl999ZVat26tatWq6dNPPzXrLl26pPnz5+uFF164iWcEKDq8/nAz6N2wJ877getD70ZR4t/mOwvndsXIAG5CmzZtjIceesi8f/nyZcPDw8N47rnnzGUnTpwwJBmJiYnG6tWrDUnG2bNni2G2uFlJSUmGJOPw4cN5xnr37m1UrFjROH/+vLnsvffeM8qVK2dkZ2fnu75Tp04Zkoxdu3YZhmEYhw4dMiQZ06ZNs6mbO3eu4eXlZezfv98ICAgwXnrpJSMnJ8eOe4ZrKehnX716dSM2NtZm2fjx443g4GDDMAzj/fffN8qXL2+cPn0633X37t3b6Natm82ywYMHG23atDHvt2nTxmjcuHGex0oyvv76a8Mw/t/xs337dpuat956y6hdu7Z5/8svvzTKlStnnDt3zmp3gRKF1x9uBr0b9sJ5P3D96N0oKvzbfOfh3K748E503LQGDRqY/+/k5KRKlSqpfv365jJfX19J0smTJ4t8brCvhg0bqkOHDqpfv76efPJJffDBBzp79qzNeNmyZc37wcHBOnfunI4ePSpJ+vnnn9WzZ09Vq1ZNnp6euvfeeyVJR44csdlO06ZN82z7woULatWqlR5//HFNnz5dDg4Ot2APYcXqZ3/+/Hn98ssvioiIULly5czb66+/bn4kLDk5WY0bN1bFihVvag5NmjS5ocf16dNHBw8e1MaNGyX9/a3iTz31lDw8PG5qPkBR4fWHm0Hvhj1x3g9cH3o3ihL/Nt9ZOLcrPoTouGllypSxue/g4GCzLPdFVdiPn6HkcXJyUnx8vL7//nvVqVNHM2fOVM2aNXXo0KHrenzXrl115swZffDBB9q0aZN5Xb8rr88lKd8TNFdXV4WEhGjJkiU6duzYze8MCsXqZ797925J0gcffKDk5GTztnv3bvPE293dvcB1Ozo6mh89y5XfF5fc6Im7j4+Punbtqrlz5yo1NVXff/89H0dFqcLrDzeD3g174rwfuD70bhQl/m2+s3BuV3wI0QEUioODg1q2bKmxY8dq+/btcnFx0ddffy1J2rFjhy5cuGDWbty4UeXKlVNAQIBOnz6tAwcOaNSoUerQoYNq165t89fSa3F0dNSnn36qJk2aqF27djp+/Ljd9w0Fy+9nv379evn7++vXX39VjRo1bG65X1bUoEEDJScn68yZM/mut0qVKjpx4oTNsuTk5ELPz8XFRZKUnZ2dZ+zFF1/UggULNGfOHFWvXl0tW7Ys9PqB4sTrDzeD3g0ARY/eDeBW4dyueBCiA7humzZt0ptvvqmtW7fqyJEj+uqrr3Tq1CnVrl1b0t9/uYyIiNDevXu1bNkyjR49WlFRUXJ0dFSFChVUqVIlzZkzRwcPHtSqVas0bNiwQm3fyclJ8+fPV8OGDdW+ffs83xqPW6egn/3YsWM1YcIEzZgxQz/99JN27dqluXPnasqUKZKknj17ys/PT+Hh4Vq/fr1+/fVXffnll0pMTJQktW/fXlu3btUnn3yin3/+WaNHjzbfpVMYPj4+cnd3V1xcnFJTU5Wenm6OhYaGytPTU6+//rr69u1rnycFKCK8/nAz6N0AUPTo3QBuFc7tig8hOoDr5unpqXXr1qlLly76xz/+oVGjRmny5Mnq3LmzJKlDhw6677771Lp1az399NN69NFHNWbMGEl//8Xy888/V1JSkurVq6ehQ4dq0qRJhZ6Ds7OzPvvsM9WtW1ft27fnum5FpKCf/YsvvqgPP/xQc+fOVf369dWmTRvFxMSY76ZxcXHRihUr5OPjoy5duqh+/fr673//KycnJ0l/n6S/9tprGjFihJo1a6Y///xTzz//fKHn6OzsrBkzZuj999+Xv7+/unXrZo45OjqqT58+ys7OvqF1A8WJ1x9uBr0bAIoevRvArcK5XfFxMK6+mBYA3IA+ffooLS1NixcvLu6pAPmKiIjQqVOn9O233xb3VIA7Dq+/koneDQCwQu8GSh/O7W4t5+KeAAAAt1J6erp27dql2NhYfgkAihivPwAAShd6NwDkjxAdAHBb69atmzZv3qwBAwbo4YcfLu7pAHcUXn8AAJQu9G4AyB+XcwEAAAAAAAAAwAJfLAoAAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAdQZGJiYuTt7X3T63FwcNDixYtvej0AAKBg9G4AAEoXejdwaxCiAyiUPn36KDw8vLinAQAArhO9GwCA0oXeDZQ8hOgAAAAAAAAAAFggRAdgN1OmTFH9+vXl4eGhgIAA/etf/9K5c+fy1C1evFj33Xef3NzcFBoaqqNHj9qMf/PNN7r//vvl5uamatWqaezYsbp8+XJR7QYAAHcMejcAAKULvRsoHoToAOzG0dFRM2bM0J49ezRv3jytWrVKI0aMsKn566+/9MYbb+iTTz7R+vXrlZaWph49epjjP/zwg55//nkNHjxYe/fu1fvvv6+YmBi98cYbRb07AADc9ujdAACULvRuoHg4GIZhFPckAJQeffr0UVpa2nV9wcgXX3yhAQMG6I8//pD09xec9O3bVxs3blTz5s0lSfv371ft2rW1adMmPfDAAwoJCVGHDh00cuRIcz3/+9//NGLECB0/flzS319w8vXXX3ONOAAArgO9GwCA0oXeDZQ8zsU9AQC3j5UrV2rChAnav3+/MjIydPnyZV28eFF//fWXypYtK0lydnZWs2bNzMfUqlVL3t7e2rdvnx544AHt2LFD69evt/kLeHZ2dp71AACAm0fvBgCgdKF3A8WDEB2AXRw+fFiPPPKIBg4cqDfeeEMVK1bUjz/+qIiICGVlZV13Ez537pzGjh2rxx9/PM+Ym5ubvacNAMAdi94NAEDpQu8Gig8hOgC7SEpKUk5OjiZPnixHx7+/bmHhwoV56i5fvqytW7fqgQcekCQdOHBAaWlpql27tiTp/vvv14EDB1SjRo2imzwAAHcgejcAAKULvRsoPoToAAotPT1dycnJNssqV66sS5cuaebMmeratavWr1+v2bNn53lsmTJlNGjQIM2YMUPOzs6KiopSixYtzOYeHR2tRx55RPfcc4+eeOIJOTo6aseOHdq9e7def/31otg9AABuO/RuAABKF3o3ULI4FvcEAJQ+a9asUePGjW1un376qaZMmaK33npL9erV0/z58zVhwoQ8jy1btqxefvllPfPMM2rZsqXKlSunBQsWmOOhoaFasmSJVqxYoWbNmqlFixaaOnWqAgMDi3IXAQC4rdC7AQAoXejdQMniYBiGUdyTAAAAAAAAAACgJOKd6AAAAAAAAAAAWCBEBwAAAAAAAADAAiE6AAAAAAAAAAAWCNEBAAAAAAAAALBAiA4AAAAAAAAAgAVCdAAAAAAAAAAALBCiAwAAAAAAAABggRAdAAAAAAAAAAALhOgAAAAAAAAAAFggRAcAAAAAAAAAwAIhOgAAAAAAAAAAFgjRAQAAAAAAAACw8P8BqC+cjWHTVLwAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Label Distribution in Training Set\n",
            "ml          30000\n",
            "spark       30000\n",
            "security    29999\n",
            "Name: Label, dtype: int64\n",
            "Label Distribution in Validation Set\n",
            "security    10000\n",
            "spark       10000\n",
            "ml          10000\n",
            "Name: Label, dtype: int64\n",
            "Label Distribution in Test Set\n",
            "security    10000\n",
            "ml          10000\n",
            "spark       10000\n",
            "Name: Label, dtype: int64\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Build the model"
      ],
      "metadata": {
        "id": "14Df7jmFumlo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "from torch.utils.data import Dataset, DataLoader\n",
        "\n",
        "class StackOverflowDataset(Dataset):\n",
        "    \"\"\"\n",
        "    A custom Dataset class for StackOverflow data.\n",
        "    \"\"\"\n",
        "    def __init__(self, df, label_to_idx):\n",
        "        \"\"\"\n",
        "        Initializes a new StackOverflowDataset object.\n",
        "        \n",
        "        Args:\n",
        "        df (pandas.DataFrame): the DataFrame containing the data\n",
        "        label_to_idx (dict): a dictionary mapping labels to their corresponding index values\n",
        "        \"\"\"\n",
        "        self.df = df\n",
        "        self.label_to_idx = label_to_idx\n",
        "\n",
        "    def __len__(self):\n",
        "        \"\"\"\n",
        "        Returns the length of the dataset.\n",
        "        \n",
        "        Returns:\n",
        "        int: the number of samples in the dataset\n",
        "        \"\"\"\n",
        "        return len(self.df)\n",
        "\n",
        "    def __getitem__(self, idx):\n",
        "        \"\"\"\n",
        "        Retrieves a sample from the dataset at the specified index.\n",
        "        \n",
        "        Args:\n",
        "        idx (int): the index of the sample to retrieve\n",
        "        \n",
        "        Returns:\n",
        "        tuple: a tuple containing the vectorized title (torch.Tensor) and the label (torch.Tensor)\n",
        "        \"\"\"\n",
        "        text = torch.stack(list(self.df.iloc[idx][\"Vectorized_Title\"]))\n",
        "        label = torch.tensor(self.label_to_idx[self.df.iloc[idx][\"Label\"]], dtype=torch.long)\n",
        "        return text, label\n",
        "\n",
        "def collate_batch(batch):\n",
        "    \"\"\"\n",
        "    A custom collate function to pad the text sequences in a batch to the same length.\n",
        "    \n",
        "    Args:\n",
        "    batch (list): a list of samples (text, label) to be collated\n",
        "    \n",
        "    Returns:\n",
        "    tuple: a tuple containing the padded texts (torch.Tensor) and labels (torch.Tensor)\n",
        "    \"\"\"\n",
        "    texts, labels = zip(*batch)\n",
        "    text_lens = [len(text) for text in texts]\n",
        "    max_len = max(text_lens)\n",
        "    padded_texts = [torch.cat([text, torch.zeros(max_len - len(text), 100)]) for text in texts]\n",
        "    return torch.stack(padded_texts), torch.tensor(labels, dtype=torch.long)"
      ],
      "metadata": {
        "id": "Q5A8fLBuucWa"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import torch.nn as nn\n",
        "\n",
        "class LSTMClassifier(nn.Module):\n",
        "    \"\"\"\n",
        "    A custom LSTM-based classifier for text classification.\n",
        "    \"\"\"\n",
        "    def __init__(self, embed_dim, hidden_dim, num_classes, dropout=0.5):\n",
        "        \"\"\"\n",
        "        Initializes a new LSTMClassifier object.\n",
        "        \n",
        "        Args:\n",
        "        embed_dim (int): the dimension of the word embeddings\n",
        "        hidden_dim (int): the size of the hidden state in the LSTM layer\n",
        "        num_classes (int): the number of classes in the classification task\n",
        "        dropout (float, optional): the dropout probability (default: 0.5)\n",
        "        \"\"\"\n",
        "        super(LSTMClassifier, self).__init__()\n",
        "        self.lstm = nn.LSTM(embed_dim, hidden_dim, batch_first=True)\n",
        "        self.fc = nn.Linear(hidden_dim, num_classes)\n",
        "        self.dropout = nn.Dropout(dropout)\n",
        "    \n",
        "    def forward(self, text):\n",
        "        \"\"\"\n",
        "        Defines the forward pass for the LSTMClassifier.\n",
        "        \n",
        "        Args:\n",
        "        text (torch.Tensor): the input text tensor\n",
        "        \n",
        "        Returns:\n",
        "        torch.Tensor: the logits for the classification task\n",
        "        \"\"\"\n",
        "        lstm_out, _ = self.lstm(text)\n",
        "        pooled = self.dropout(torch.mean(lstm_out, dim=1))\n",
        "        logits = self.fc(pooled)\n",
        "        return logits\n",
        "\n",
        "class GRUClassifier(nn.Module):\n",
        "    \"\"\"\n",
        "    A custom GRU-based classifier for text classification.\n",
        "    \"\"\"\n",
        "    def __init__(self, embed_dim, hidden_dim, num_classes, dropout=0.5):\n",
        "        \"\"\"\n",
        "        Initializes a new GRUClassifier object.\n",
        "        \n",
        "        Args:\n",
        "        embed_dim (int): the dimension of the word embeddings\n",
        "        hidden_dim (int): the size of the hidden state in the GRU layer\n",
        "        num_classes (int): the number of classes in the classification task\n",
        "        dropout (float, optional): the dropout probability (default: 0.5)\n",
        "        \"\"\"\n",
        "        super(GRUClassifier, self).__init__()\n",
        "        self.gru = nn.GRU(embed_dim, hidden_dim, batch_first=True)\n",
        "        self.fc = nn.Linear(hidden_dim, num_classes)\n",
        "        self.dropout = nn.Dropout(dropout)\n",
        "    \n",
        "    def forward(self, text):\n",
        "        \"\"\"\n",
        "        Defines the forward pass for the GRUClassifier.\n",
        "        \n",
        "        Args:\n",
        "        text (torch.Tensor): the input text tensor\n",
        "        \n",
        "        Returns:\n",
        "        torch.Tensor: the logits for the classification task\n",
        "        \"\"\"\n",
        "        gru_out, _ = self.gru(text)\n",
        "        pooled = self.dropout(torch.mean(gru_out, dim=1))\n",
        "        logits = self.fc(pooled)\n",
        "        return logits"
      ],
      "metadata": {
        "id": "kzVpbYf1uxV5"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create a dictionary to map labels to their corresponding index values\n",
        "label_to_idx = {\"spark\": 0, \"ml\": 1, \"security\": 2}\n",
        "\n",
        "# Initialize the StackOverflowDataset objects for training, validation, and testing\n",
        "train_dataset = StackOverflowDataset(train_df, label_to_idx)\n",
        "val_dataset = StackOverflowDataset(val_df, label_to_idx)\n",
        "test_dataset = StackOverflowDataset(test_df, label_to_idx)\n",
        "\n",
        "# Set the batch size for data loading\n",
        "batch_size = 64\n",
        "\n",
        "# Initialize DataLoaders for training, validation, and testing\n",
        "train_dataloader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True, collate_fn=collate_batch)\n",
        "val_dataloader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False, collate_fn=collate_batch)\n",
        "test_dataloader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False, collate_fn=collate_batch)"
      ],
      "metadata": {
        "id": "RxqObmcSupK_"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Training"
      ],
      "metadata": {
        "id": "WLN9cVe22O1q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch.optim as optim\n",
        "\n",
        "def train_and_evaluate(model, num_epochs=10):\n",
        "    \"\"\"\n",
        "    Trains and evaluates a given PyTorch model on the train and validation datasets,\n",
        "    and tests the model on the test dataset.\n",
        "    \n",
        "    Args:\n",
        "    model: PyTorch model to be trained and evaluated\n",
        "    num_epochs: Number of epochs to train the model (default: 10)\n",
        "    \n",
        "    Returns:\n",
        "    true_labels: List of true labels for the test dataset\n",
        "    pred_labels: List of predicted labels for the test dataset\n",
        "    \"\"\"\n",
        "    device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "    model.to(device)\n",
        "    optimizer = optim.Adam(model.parameters(), lr=1e-3)\n",
        "    criterion = nn.CrossEntropyLoss()\n",
        "    \n",
        "    # Train and evaluate the model\n",
        "    for epoch in range(num_epochs):\n",
        "        # Training\n",
        "        model.train()\n",
        "        train_loss = 0\n",
        "        for texts, labels in train_dataloader:\n",
        "            texts, labels = texts.to(device), labels.to(device)\n",
        "            optimizer.zero_grad()\n",
        "            logits = model(texts)\n",
        "            loss = criterion(logits, labels)\n",
        "            loss.backward()\n",
        "            optimizer.step()\n",
        "            train_loss += loss.item()\n",
        "        \n",
        "        # Validation\n",
        "        model.eval()\n",
        "        val_loss = 0\n",
        "        with torch.no_grad():\n",
        "            for texts, labels in val_dataloader:\n",
        "                texts, labels = texts.to(device), labels.to(device)\n",
        "                logits = model(texts)\n",
        "                loss = criterion(logits, labels)\n",
        "                val_loss += loss.item()\n",
        "        \n",
        "        train_loss /= len(train_dataloader)\n",
        "        val_loss /= len(val_dataloader)\n",
        "        print(f\"Epoch {epoch + 1}/{num_epochs} | Train Loss: {train_loss:.4f} | Val Loss: {val_loss:.4f}\")\n",
        "\n",
        "    # Testing\n",
        "    model.eval()\n",
        "    true_labels = []\n",
        "    pred_labels = []\n",
        "    with torch.no_grad():\n",
        "        for texts, labels in test_dataloader:\n",
        "            texts, labels = texts.to(device), labels.to(device)\n",
        "            logits = model(texts)\n",
        "            predictions = torch.argmax(logits, dim=-1)\n",
        "            true_labels.extend(labels.tolist())\n",
        "            pred_labels.extend(predictions.tolist())\n",
        "    \n",
        "    return true_labels, pred_labels\n",
        "\n",
        "embed_dim = 100\n",
        "hidden_dim = 128\n",
        "num_classes = 3\n",
        "\n",
        "lstm_model = LSTMClassifier(embed_dim, hidden_dim, num_classes)\n",
        "print(\"Training and evaluating LSTM model:\")\n",
        "lstm_true_labels, lstm_pred_labels = train_and_evaluate(lstm_model)\n",
        "\n",
        "gru_model = GRUClassifier(embed_dim, hidden_dim, num_classes)\n",
        "print(\"\\nTraining and evaluating GRU model:\")\n",
        "gru_true_labels, gru_pred_labels = train_and_evaluate(gru_model)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "g4YJ22Jruzrl",
        "outputId": "907fe586-67dd-43ab-c6e3-c410d42908b6"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training and evaluating LSTM model:\n",
            "Epoch 1/10 | Train Loss: 0.3794 | Val Loss: 0.2864\n",
            "Epoch 2/10 | Train Loss: 0.2657 | Val Loss: 0.2542\n",
            "Epoch 3/10 | Train Loss: 0.2362 | Val Loss: 0.2426\n",
            "Epoch 4/10 | Train Loss: 0.2145 | Val Loss: 0.2349\n",
            "Epoch 5/10 | Train Loss: 0.2001 | Val Loss: 0.2355\n",
            "Epoch 6/10 | Train Loss: 0.1855 | Val Loss: 0.2321\n",
            "Epoch 7/10 | Train Loss: 0.1732 | Val Loss: 0.2329\n",
            "Epoch 8/10 | Train Loss: 0.1616 | Val Loss: 0.2507\n",
            "Epoch 9/10 | Train Loss: 0.1493 | Val Loss: 0.2517\n",
            "Epoch 10/10 | Train Loss: 0.1395 | Val Loss: 0.2609\n",
            "\n",
            "Training and evaluating GRU model:\n",
            "Epoch 1/10 | Train Loss: 0.3586 | Val Loss: 0.2856\n",
            "Epoch 2/10 | Train Loss: 0.2597 | Val Loss: 0.2620\n",
            "Epoch 3/10 | Train Loss: 0.2314 | Val Loss: 0.2410\n",
            "Epoch 4/10 | Train Loss: 0.2138 | Val Loss: 0.2355\n",
            "Epoch 5/10 | Train Loss: 0.1961 | Val Loss: 0.2308\n",
            "Epoch 6/10 | Train Loss: 0.1833 | Val Loss: 0.2306\n",
            "Epoch 7/10 | Train Loss: 0.1701 | Val Loss: 0.2359\n",
            "Epoch 8/10 | Train Loss: 0.1552 | Val Loss: 0.2360\n",
            "Epoch 9/10 | Train Loss: 0.1427 | Val Loss: 0.2512\n",
            "Epoch 10/10 | Train Loss: 0.1315 | Val Loss: 0.2799\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import roc_curve, auc\n",
        "from sklearn.preprocessing import label_binarize\n",
        "\n",
        "def compute_roc_auc(true_labels, pred_labels, num_classes):\n",
        "    \"\"\"\n",
        "    Computes ROC curves and AUC scores for multiclass classification.\n",
        "    \n",
        "    Args:\n",
        "    true_labels: List of true labels for the dataset\n",
        "    pred_labels: List of predicted labels for the dataset\n",
        "    num_classes: Number of classes in the classification problem\n",
        "    \n",
        "    Returns:\n",
        "    fpr: Dictionary containing false positive rates for each class\n",
        "    tpr: Dictionary containing true positive rates for each class\n",
        "    roc_auc: Dictionary containing ROC AUC scores for each class\n",
        "    \"\"\"\n",
        "    true_labels_bin = label_binarize(true_labels, classes=range(num_classes))\n",
        "    pred_labels_bin = label_binarize(pred_labels, classes=range(num_classes))\n",
        "    \n",
        "    fpr = dict()\n",
        "    tpr = dict()\n",
        "    roc_auc = dict()\n",
        "    for i in range(num_classes):\n",
        "        fpr[i], tpr[i], _ = roc_curve(true_labels_bin[:, i], pred_labels_bin[:, i])\n",
        "        roc_auc[i] = auc(fpr[i], tpr[i])\n",
        "    return fpr, tpr, roc_auc\n",
        "\n",
        "lstm_fpr, lstm_tpr, lstm_roc_auc = compute_roc_auc(lstm_true_labels, lstm_pred_labels, num_classes)\n",
        "gru_fpr, gru_tpr, gru_roc_auc = compute_roc_auc(gru_true_labels, gru_pred_labels, num_classes)"
      ],
      "metadata": {
        "id": "XJjGGGx5vIIP"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "print(\"LSTM Model Performance:\")\n",
        "print(classification_report(lstm_true_labels, lstm_pred_labels, target_names=list(label_to_idx.keys()), digits=4))\n",
        "\n",
        "\n",
        "print(\"GRU Model Performance:\")\n",
        "print(classification_report(gru_true_labels, gru_pred_labels, target_names=list(label_to_idx.keys()), digits=4))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cwKUGvrPxa68",
        "outputId": "74567754-c1e3-4542-e618-518c8c5d28d7"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LSTM Model Performance:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "       spark     0.8997    0.9087    0.9042     10000\n",
            "          ml     0.9204    0.8933    0.9066     10000\n",
            "    security     0.9264    0.9444    0.9353     10000\n",
            "\n",
            "    accuracy                         0.9155     30000\n",
            "   macro avg     0.9155    0.9155    0.9154     30000\n",
            "weighted avg     0.9155    0.9155    0.9154     30000\n",
            "\n",
            "GRU Model Performance:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "       spark     0.8938    0.9135    0.9036     10000\n",
            "          ml     0.9151    0.8930    0.9039     10000\n",
            "    security     0.9332    0.9352    0.9342     10000\n",
            "\n",
            "    accuracy                         0.9139     30000\n",
            "   macro avg     0.9140    0.9139    0.9139     30000\n",
            "weighted avg     0.9140    0.9139    0.9139     30000\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Based on the validation metrics, it appears that both models perform similarly. We can observe the ROC curve to determine which model to perform hyperparameter tuning on."
      ],
      "metadata": {
        "id": "FPU39ogTzrps"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "def plot_separate_roc_curves(lstm_fpr, lstm_tpr, lstm_roc_auc, gru_fpr, gru_tpr, gru_roc_auc):\n",
        "    \"\"\"\n",
        "    This function plots separate ROC curves for the LSTM and GRU classifiers for each class.\n",
        "    \n",
        "    Parameters:\n",
        "    lstm_fpr (dict): A dictionary containing false positive rates for each class in the LSTM classifier.\n",
        "    lstm_tpr (dict): A dictionary containing true positive rates for each class in the LSTM classifier.\n",
        "    lstm_roc_auc (dict): A dictionary containing the area under the ROC curve for each class in the LSTM classifier.\n",
        "    gru_fpr (dict): A dictionary containing false positive rates for each class in the GRU classifier.\n",
        "    gru_tpr (dict): A dictionary containing true positive rates for each class in the GRU classifier.\n",
        "    gru_roc_auc (dict): A dictionary containing the area under the ROC curve for each class in the GRU classifier.\n",
        "    \"\"\"\n",
        "    fig, axes = plt.subplots(1, 2, figsize=(12, 6))\n",
        "    \n",
        "    for ax, model_name, fpr, tpr, roc_auc in zip(axes, ['LSTM', 'GRU'], [lstm_fpr, gru_fpr], [lstm_tpr, gru_tpr], [lstm_roc_auc, gru_roc_auc]):\n",
        "        for i, label in enumerate(label_to_idx.keys()):\n",
        "            ax.plot(fpr[i], tpr[i], label=f\"{label} (AUC = {roc_auc[i]:.4f})\")\n",
        "        \n",
        "        ax.plot([0, 1], [0, 1], 'k--')\n",
        "        ax.set_xlim([0.0, 1.0])\n",
        "        ax.set_ylim([0.0, 1.05])\n",
        "        ax.set_xlabel(\"False Positive Rate\")\n",
        "        ax.set_ylabel(\"True Positive Rate\")\n",
        "        ax.set_title(f\"{model_name} ROC Curve\")\n",
        "        ax.legend(loc=\"lower right\")\n",
        "        \n",
        "        total_auc = np.mean(list(roc_auc.values()))\n",
        "        ax.text(0.6, 0.3, f\"Average AUC= {total_auc:.4f}\", fontsize=12, bbox=dict(facecolor='red', alpha=0.5))\n",
        "    \n",
        "    plt.tight_layout()\n",
        "    plt.suptitle(\"LSTM and GRU ROC Curves for Classifying Stack Overflow Posts\", fontsize=16, y=1.05)\n",
        "    plt.show()"
      ],
      "metadata": {
        "id": "t0xJMDtGxZJe"
      },
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_separate_roc_curves(lstm_fpr, lstm_tpr, lstm_roc_auc, gru_fpr, gru_tpr, gru_roc_auc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 530
        },
        "id": "FCkym7XC0VGC",
        "outputId": "4af1999a-8a75-4b50-da05-ff0243c66533"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Proceed with performing hyperparameter tuning on LSTM model since the average AUC between all three classes is higher."
      ],
      "metadata": {
        "id": "HZpTTC1Dz3hR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "import seaborn as sns\n",
        "\n",
        "def plot_confusion_matrices(true_labels, pred_labels_lstm, pred_labels_gru):\n",
        "    \"\"\"\n",
        "    This function plots the confusion matrices for the LSTM and GRU classifiers.\n",
        "    \n",
        "    Parameters:\n",
        "    true_labels (list): The list of true labels for the test dataset.\n",
        "    pred_labels_lstm (list): The list of predicted labels by the LSTM classifier.\n",
        "    pred_labels_gru (list): The list of predicted labels by the GRU classifier.\n",
        "    \"\"\"\n",
        "    labels = list(label_to_idx.keys())\n",
        "    lstm_cm = confusion_matrix(true_labels, pred_labels_lstm)\n",
        "    gru_cm = confusion_matrix(true_labels, pred_labels_gru)\n",
        "    \n",
        "    fig, axes = plt.subplots(1, 2, figsize=(12, 5))\n",
        "    \n",
        "    for ax, model_name, cm in zip(axes, ['LSTM', 'GRU'], [lstm_cm, gru_cm]):\n",
        "        sns.heatmap(cm, annot=True, fmt=\"d\", cmap=\"Blues\", ax=ax, xticklabels=labels, yticklabels=labels)\n",
        "        ax.set_xlabel(\"Predicted\")\n",
        "        ax.set_ylabel(\"True\")\n",
        "        ax.set_title(f\"{model_name} Confusion Matrix\")\n",
        "    \n",
        "    plt.tight_layout()\n",
        "    plt.suptitle(\"Confusion Matrices for LSTM and GRU Models for Stack Overflow Post Classification\", y=1.05)\n",
        "    plt.show()"
      ],
      "metadata": {
        "id": "-FSvbrTW0UK1"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_confusion_matrices(lstm_true_labels, lstm_pred_labels, gru_pred_labels)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "dEKAA43kzDKk",
        "outputId": "f614e274-a523-453f-f0d7-dfdff6d52d83"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Hyperparameter tuning\n",
        "Perform hyperparameter tuning on LSTM model"
      ],
      "metadata": {
        "id": "3zgDQDPZ1kPN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import ParameterGrid\n",
        "from sklearn.metrics import f1_score\n",
        "\n",
        "# Define hyperparameter search space\n",
        "param_grid = {\n",
        "    \"hidden_dim\": [64, 128, 256],\n",
        "    \"dropout\": [0.3, 0.5, 0.7]\n",
        "}\n",
        "param_combinations = list(ParameterGrid(param_grid))\n",
        "\n",
        "def hyperparameter_tuning(model_class, param_combinations):\n",
        "    \"\"\"\n",
        "    This function performs hyperparameter tuning on the given model class using a grid search.\n",
        "    \n",
        "    Parameters:\n",
        "    model_class (class): The class of the model to be tuned, e.g., LSTMClassifier or GRUClassifier.\n",
        "    param_combinations (list): A list of dictionaries, each representing a combination of hyperparameters.\n",
        "    \n",
        "    Returns:\n",
        "    best_model (object): The best model with the optimal hyperparameters.\n",
        "    best_params (dict): A dictionary containing the optimal hyperparameters.\n",
        "    best_true_labels (list): The list of true labels used for evaluation.\n",
        "    best_pred_labels (list): The list of predicted labels by the best model.\n",
        "    \"\"\"\n",
        "    best_model = None\n",
        "    best_params = None\n",
        "    best_f1 = -1\n",
        "    best_true_labels = None\n",
        "    best_pred_labels = None\n",
        "\n",
        "    for params in param_combinations:\n",
        "        model = model_class(embed_dim, params[\"hidden_dim\"], num_classes, dropout=params[\"dropout\"])\n",
        "        true_labels, pred_labels = train_and_evaluate(model)\n",
        "        f1 = f1_score(true_labels, pred_labels, average=\"weighted\")\n",
        "        if f1 > best_f1:\n",
        "            best_f1 = f1\n",
        "            best_model = model\n",
        "            best_params = params\n",
        "            best_true_labels = true_labels\n",
        "            best_pred_labels = pred_labels\n",
        "    return best_model, best_params, best_true_labels, best_pred_labels\n",
        "\n",
        "lstm_best_model, lstm_best_params, lstm_true_labels, lstm_best_pred_labels = hyperparameter_tuning(LSTMClassifier, param_combinations)\n",
        "print(\"Best LSTM parameters:\", lstm_best_params)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KAPyrELl032n",
        "outputId": "7a1ff08c-2636-439d-c624-9449cb665ea2"
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10 | Train Loss: 0.3866 | Val Loss: 0.2958\n",
            "Epoch 2/10 | Train Loss: 0.2710 | Val Loss: 0.2545\n",
            "Epoch 3/10 | Train Loss: 0.2389 | Val Loss: 0.2568\n",
            "Epoch 4/10 | Train Loss: 0.2221 | Val Loss: 0.2350\n",
            "Epoch 5/10 | Train Loss: 0.2079 | Val Loss: 0.2343\n",
            "Epoch 6/10 | Train Loss: 0.1970 | Val Loss: 0.2253\n",
            "Epoch 7/10 | Train Loss: 0.1873 | Val Loss: 0.2307\n",
            "Epoch 8/10 | Train Loss: 0.1783 | Val Loss: 0.2313\n",
            "Epoch 9/10 | Train Loss: 0.1704 | Val Loss: 0.2399\n",
            "Epoch 10/10 | Train Loss: 0.1630 | Val Loss: 0.2462\n",
            "Epoch 1/10 | Train Loss: 0.3736 | Val Loss: 0.2791\n",
            "Epoch 2/10 | Train Loss: 0.2612 | Val Loss: 0.2708\n",
            "Epoch 3/10 | Train Loss: 0.2310 | Val Loss: 0.2399\n",
            "Epoch 4/10 | Train Loss: 0.2119 | Val Loss: 0.2396\n",
            "Epoch 5/10 | Train Loss: 0.1950 | Val Loss: 0.2333\n",
            "Epoch 6/10 | Train Loss: 0.1817 | Val Loss: 0.2340\n",
            "Epoch 7/10 | Train Loss: 0.1682 | Val Loss: 0.2409\n",
            "Epoch 8/10 | Train Loss: 0.1566 | Val Loss: 0.2516\n",
            "Epoch 9/10 | Train Loss: 0.1440 | Val Loss: 0.2554\n",
            "Epoch 10/10 | Train Loss: 0.1332 | Val Loss: 0.2477\n",
            "Epoch 1/10 | Train Loss: 0.3752 | Val Loss: 0.2902\n",
            "Epoch 2/10 | Train Loss: 0.2597 | Val Loss: 0.2525\n",
            "Epoch 3/10 | Train Loss: 0.2261 | Val Loss: 0.2374\n",
            "Epoch 4/10 | Train Loss: 0.2049 | Val Loss: 0.2338\n",
            "Epoch 5/10 | Train Loss: 0.1848 | Val Loss: 0.2320\n",
            "Epoch 6/10 | Train Loss: 0.1664 | Val Loss: 0.2368\n",
            "Epoch 7/10 | Train Loss: 0.1497 | Val Loss: 0.2486\n",
            "Epoch 8/10 | Train Loss: 0.1339 | Val Loss: 0.2422\n",
            "Epoch 9/10 | Train Loss: 0.1189 | Val Loss: 0.2743\n",
            "Epoch 10/10 | Train Loss: 0.1048 | Val Loss: 0.2845\n",
            "Epoch 1/10 | Train Loss: 0.3932 | Val Loss: 0.2918\n",
            "Epoch 2/10 | Train Loss: 0.2760 | Val Loss: 0.2605\n",
            "Epoch 3/10 | Train Loss: 0.2470 | Val Loss: 0.2476\n",
            "Epoch 4/10 | Train Loss: 0.2293 | Val Loss: 0.2401\n",
            "Epoch 5/10 | Train Loss: 0.2151 | Val Loss: 0.2395\n",
            "Epoch 6/10 | Train Loss: 0.2033 | Val Loss: 0.2369\n",
            "Epoch 7/10 | Train Loss: 0.1946 | Val Loss: 0.2396\n",
            "Epoch 8/10 | Train Loss: 0.1862 | Val Loss: 0.2409\n",
            "Epoch 9/10 | Train Loss: 0.1789 | Val Loss: 0.2306\n",
            "Epoch 10/10 | Train Loss: 0.1720 | Val Loss: 0.2416\n",
            "Epoch 1/10 | Train Loss: 0.3780 | Val Loss: 0.2871\n",
            "Epoch 2/10 | Train Loss: 0.2658 | Val Loss: 0.2530\n",
            "Epoch 3/10 | Train Loss: 0.2357 | Val Loss: 0.2387\n",
            "Epoch 4/10 | Train Loss: 0.2161 | Val Loss: 0.2387\n",
            "Epoch 5/10 | Train Loss: 0.2016 | Val Loss: 0.2280\n",
            "Epoch 6/10 | Train Loss: 0.1855 | Val Loss: 0.2311\n",
            "Epoch 7/10 | Train Loss: 0.1735 | Val Loss: 0.2366\n",
            "Epoch 8/10 | Train Loss: 0.1617 | Val Loss: 0.2436\n",
            "Epoch 9/10 | Train Loss: 0.1512 | Val Loss: 0.2528\n",
            "Epoch 10/10 | Train Loss: 0.1400 | Val Loss: 0.2553\n",
            "Epoch 1/10 | Train Loss: 0.3693 | Val Loss: 0.2780\n",
            "Epoch 2/10 | Train Loss: 0.2593 | Val Loss: 0.2512\n",
            "Epoch 3/10 | Train Loss: 0.2271 | Val Loss: 0.2428\n",
            "Epoch 4/10 | Train Loss: 0.2053 | Val Loss: 0.2311\n",
            "Epoch 5/10 | Train Loss: 0.1865 | Val Loss: 0.2348\n",
            "Epoch 6/10 | Train Loss: 0.1695 | Val Loss: 0.2385\n",
            "Epoch 7/10 | Train Loss: 0.1524 | Val Loss: 0.2522\n",
            "Epoch 8/10 | Train Loss: 0.1366 | Val Loss: 0.2586\n",
            "Epoch 9/10 | Train Loss: 0.1218 | Val Loss: 0.2922\n",
            "Epoch 10/10 | Train Loss: 0.1100 | Val Loss: 0.2926\n",
            "Epoch 1/10 | Train Loss: 0.4104 | Val Loss: 0.3028\n",
            "Epoch 2/10 | Train Loss: 0.2894 | Val Loss: 0.2638\n",
            "Epoch 3/10 | Train Loss: 0.2567 | Val Loss: 0.2497\n",
            "Epoch 4/10 | Train Loss: 0.2391 | Val Loss: 0.2436\n",
            "Epoch 5/10 | Train Loss: 0.2257 | Val Loss: 0.2366\n",
            "Epoch 6/10 | Train Loss: 0.2143 | Val Loss: 0.2399\n",
            "Epoch 7/10 | Train Loss: 0.2058 | Val Loss: 0.2386\n",
            "Epoch 8/10 | Train Loss: 0.1982 | Val Loss: 0.2403\n",
            "Epoch 9/10 | Train Loss: 0.1896 | Val Loss: 0.2399\n",
            "Epoch 10/10 | Train Loss: 0.1818 | Val Loss: 0.2430\n",
            "Epoch 1/10 | Train Loss: 0.3876 | Val Loss: 0.3033\n",
            "Epoch 2/10 | Train Loss: 0.2731 | Val Loss: 0.2669\n",
            "Epoch 3/10 | Train Loss: 0.2411 | Val Loss: 0.2445\n",
            "Epoch 4/10 | Train Loss: 0.2208 | Val Loss: 0.2346\n",
            "Epoch 5/10 | Train Loss: 0.2061 | Val Loss: 0.2316\n",
            "Epoch 6/10 | Train Loss: 0.1926 | Val Loss: 0.2334\n",
            "Epoch 7/10 | Train Loss: 0.1813 | Val Loss: 0.2596\n",
            "Epoch 8/10 | Train Loss: 0.1706 | Val Loss: 0.2397\n",
            "Epoch 9/10 | Train Loss: 0.1576 | Val Loss: 0.2436\n",
            "Epoch 10/10 | Train Loss: 0.1494 | Val Loss: 0.2604\n",
            "Epoch 1/10 | Train Loss: 0.3747 | Val Loss: 0.2817\n",
            "Epoch 2/10 | Train Loss: 0.2653 | Val Loss: 0.2515\n",
            "Epoch 3/10 | Train Loss: 0.2322 | Val Loss: 0.2476\n",
            "Epoch 4/10 | Train Loss: 0.2097 | Val Loss: 0.2465\n",
            "Epoch 5/10 | Train Loss: 0.1930 | Val Loss: 0.2329\n",
            "Epoch 6/10 | Train Loss: 0.1749 | Val Loss: 0.2525\n",
            "Epoch 7/10 | Train Loss: 0.1609 | Val Loss: 0.2439\n",
            "Epoch 8/10 | Train Loss: 0.1459 | Val Loss: 0.2586\n",
            "Epoch 9/10 | Train Loss: 0.1328 | Val Loss: 0.2602\n",
            "Epoch 10/10 | Train Loss: 0.1182 | Val Loss: 0.3012\n",
            "Best LSTM parameters: {'dropout': 0.3, 'hidden_dim': 256}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "print('LSTM Tuned Model Performance:')\n",
        "print(classification_report(lstm_true_labels, lstm_best_pred_labels, target_names=label_to_idx.keys(), digits=4))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZIoUTZOm7Dil",
        "outputId": "87cb9657-3ff6-406b-baf9-3cb96772d33b"
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LSTM Tuned Model Performance:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "       spark     0.9058    0.9006    0.9032     10000\n",
            "          ml     0.9020    0.9102    0.9061     10000\n",
            "    security     0.9367    0.9335    0.9351     10000\n",
            "\n",
            "    accuracy                         0.9148     30000\n",
            "   macro avg     0.9148    0.9148    0.9148     30000\n",
            "weighted avg     0.9148    0.9148    0.9148     30000\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "from sklearn.metrics import confusion_matrix, roc_curve, auc\n",
        "from sklearn.preprocessing import label_binarize\n",
        "\n",
        "# Compute the confusion matrix\n",
        "best_lstm_cm = confusion_matrix(lstm_true_labels, lstm_best_pred_labels)\n",
        "\n",
        "# Plot the confusion matrix\n",
        "fig, ax = plt.subplots(figsize=(5, 5))\n",
        "sns.heatmap(best_lstm_cm, annot=True, fmt=\"d\", cmap=\"Blues\", ax=ax, xticklabels=label_to_idx.keys(), yticklabels=label_to_idx.keys())\n",
        "ax.set_xlabel(\"Predicted\")\n",
        "ax.set_ylabel(\"True\")\n",
        "ax.set_title(\"Best LSTM Model Confusion Matrix\")\n",
        "plt.show()\n",
        "\n",
        "# Compute ROC curve and ROC area for each class\n",
        "y_true_bin = label_binarize(lstm_true_labels, classes=[0, 1, 2])\n",
        "y_pred_bin = label_binarize(lstm_best_pred_labels, classes=[0, 1, 2])\n",
        "\n",
        "fpr = dict()\n",
        "tpr = dict()\n",
        "roc_auc = dict()\n",
        "\n",
        "for i in range(num_classes):\n",
        "    fpr[i], tpr[i], _ = roc_curve(y_true_bin[:, i], y_pred_bin[:, i])\n",
        "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
        "\n",
        "# Aggregate all false positive rates\n",
        "all_fpr = np.unique(np.concatenate([fpr[i] for i in range(num_classes)]))\n",
        "\n",
        "# Interpolate all ROC curves at this points\n",
        "mean_tpr = np.zeros_like(all_fpr)\n",
        "for i in range(num_classes):\n",
        "    mean_tpr += np.interp(all_fpr, fpr[i], tpr[i])\n",
        "\n",
        "# Average TPR and compute AUC\n",
        "mean_tpr /= num_classes\n",
        "fpr[\"macro\"] = all_fpr\n",
        "tpr[\"macro\"] = mean_tpr\n",
        "roc_auc[\"macro\"] = auc(fpr[\"macro\"], tpr[\"macro\"])\n",
        "\n",
        "# Calculate the average AUC\n",
        "average_auc = np.mean(list(roc_auc.values()))\n",
        "\n",
        "# Plot the ROC curve with individual classes\n",
        "plt.figure(figsize=(7, 7))\n",
        "colors = ['blue', 'green', 'orange']\n",
        "class_names = list(label_to_idx.keys())\n",
        "\n",
        "for i, color, class_name in zip(range(num_classes), colors, class_names):\n",
        "    plt.plot(fpr[i], tpr[i], color=color, lw=2, label=f\"Class '{class_name}' (AUC = {roc_auc[i]:.4f})\")\n",
        "\n",
        "plt.plot([0, 1], [0, 1], linestyle=\"--\", lw=2, color=\"r\", label=\"Chance\", alpha=0.8)\n",
        "plt.xlim([0.0, 1.0])\n",
        "plt.ylim([0.0, 1.05])\n",
        "plt.xlabel(\"False Positive Rate\")\n",
        "plt.ylabel(\"True Positive Rate\")\n",
        "plt.title(f\"ROC Curve for Best LSTM Model\")\n",
        "plt.legend(loc=\"lower right\")\n",
        "\n",
        "# Display the average AUC on the graph with a red box\n",
        "plt.text(0.55, 0.325, f\"Average AUC: {average_auc:.4f}\", fontsize=12, bbox=dict(facecolor=\"red\", alpha=0.5))\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "xCMogebHJ2Ay",
        "outputId": "f11cca46-945a-4f57-9034-e1cba43f9d92"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 500x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 700x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Results\n",
        "\n",
        "In selecting RNN models, LSTM and GRU were considered beacuse they are both popular types of RNNs that excel at text classification tasks. I decided to compare the performance between the two models through a series comparison of ROC curves, confusion matrices, and classification reports.\n",
        "\n",
        "The slightly higher average AUC in the LSTM ROC curve tells us that this model slightly outperforms GRU model when it comes to comparison between all three classes.\n",
        "\n",
        "Confusion matrices and Classification report also slightly favour LSTM over GRU.\n",
        "\n",
        "\n",
        "### Receiver Operating Characteristic (ROC) curves\n",
        "\n",
        "| LSTM Model | GRU Model | Tuned LSTM Model (ndim=256, dr=0.3) |\n",
        "|------------|-----------|-----------------------------------------------|\n",
        "| ![LSTM ROC](assets/lstm_roc.png) | ![GRU ROC](assets/gru_roc.png) | ![Tuned LSTM ROC](assets/tuned_lstm_roc.png) |\n",
        "\n",
        "### Confusion matrices\n",
        "\n",
        "| LSTM Model | GRU Model | Tuned LSTM Model (ndim=256, dr=0.3) |\n",
        "|------------|-----------|-----------------------------------------------|\n",
        "| ![LSTM CM](assets/lstm_cm.png) | ![GRU CM](assets/gru_cm.png) | ![Tuned LSTM CM](assets/tuned_lstm_cm.png) |\n",
        "\n",
        "\n",
        "### Classification report\n",
        "\n",
        "```\n",
        "LSTM Model Performance:\n",
        "              precision    recall  f1-score   support\n",
        "\n",
        "       spark     0.9111    0.8986    0.9048     10000\n",
        "          ml     0.9085    0.9051    0.9068     10000\n",
        "    security     0.9239    0.9400    0.9319     10000\n",
        "\n",
        "    accuracy                         0.9146     30000\n",
        "   macro avg     0.9145    0.9146    0.9145     30000\n",
        "weighted avg     0.9145    0.9146    0.9145     30000\n",
        "```\n",
        "\n",
        "```\n",
        "GRU Model Performance:\n",
        "              precision    recall  f1-score   support\n",
        "\n",
        "       spark     0.8998    0.9075    0.9036     10000\n",
        "          ml     0.9018    0.9014    0.9016     10000\n",
        "    security     0.9392    0.9315    0.9353     10000\n",
        "\n",
        "    accuracy                         0.9135     30000\n",
        "   macro avg     0.9136    0.9135    0.9135     30000\n",
        "weighted avg     0.9136    0.9135    0.9135     30000\n",
        "\n",
        "```\n",
        "\n",
        "```\n",
        "LSTM Tuned Model Performance:\n",
        " precision    recall  f1-score   support\n",
        "\n",
        "       spark     0.8868    0.9228    0.9044     10000\n",
        "          ml     0.9127    0.9021    0.9074     10000\n",
        "    security     0.9530    0.9254    0.9390     10000\n",
        "\n",
        "    accuracy                         0.9168     30000\n",
        "   macro avg     0.9175    0.9168    0.9169     30000\n",
        "weighted avg     0.9175    0.9168    0.9169     30000\n",
        "```"
      ],
      "metadata": {
        "id": "E5-NIMNZ2e6_"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Interpretation\n",
        "*i. As described in the proposal, was the question answered/topic investigated and how?*\n",
        "\n",
        "The project aimed to deepen the understanding of natural language processing and recurrent neural networks by implementing a topic classification model for categorizing Stack Overflow questions into Spark, ML, or Security. This goal was successfully achieved through a systematic approach that involved the use of the PyTorch framework and RNN-based models, specifically LSTMs or GRUs, for handling sequential text data. High levels of accuracy, precision, recall, and F1-score across all categories demonstrated the effectiveness of the developed model. The positive outcomes highlight the proficiency of RNN-based models in processing sequential text data and capturing the inherent structure and dependencies in natural language.\n",
        "\n",
        "A practical application of this model is to assist users in determining the appropriate category for posting their questions on Stack Overflow. By effectively categorizing questions, the model can enhance the user experience, guiding users to the correct category and facilitating quicker, more accurate responses from the community.\n",
        "\n",
        "## Reflection\n",
        "\n",
        "*i. Why did you select this problem to solve?*\n",
        "\n",
        "I selected this problem to solve as it allowed me to learn and reinforce my understanding of NLP, and to apply it in different domains. I had previously used NLP in the context of Spark, but never with PyTorch. Additionally, I was interested in learning more about neural networks and the PyTorch framework, so this project provided a perfect opportunity for me to delve into these topics.\n",
        "\n",
        "\n",
        "\n",
        "*ii. Were there any deviations from your proposal? Explain why (or why not).*\n",
        "\n",
        "There were no significant deviations from the original project proposal. I followed a project guide based on Amazon reviews, which provided a solid foundation for building my project. This allowed me to focus on understanding and applying NLP and PyTorch concepts effectively throughout the project.\n",
        "\n",
        "*iii. What did you find difficult about this project? What did you find easy? What did you learn?*\n",
        "\n",
        "The most challenging aspect of this project was preprocessing the data, as it contained a lot of noise and required careful NLP text preprocessing. Converting between different data types such as strings, vectors, and tensors while ensuring the types are valid was particularly tricky.\n",
        "\n",
        "On the other hand, I found designing the model and building the training loop relatively easy, since I could consult various sources for guidance, and the process of using PyTorch for machine learning appeared to be consistent across different cases. This allowed me to focus on the specific aspects of NLP and RNN-based models without getting bogged down in the details of the machine learning process itself.\n",
        "\n",
        "Throughout this project, I learned a great deal about natural language processing, RNNs (specifically LSTMs and GRUs), and the PyTorch framework. The hands-on experience helped me gain a deeper understanding of the challenges and intricacies associated with NLP and text data. I also learned about the importance of data preprocessing and the various techniques that can be employed to clean and prepare text data for model training. Furthermore, I gained valuable experience in implementing and training neural network models using PyTorch and evaluating their performance.\n",
        "\n",
        "Overall, this project has significantly increased my familiarity with PyTorch and deep learning, serving as an excellent stepping stone for further exploration of these topics. I thoroughly enjoyed applying the concepts learned in class to create my own machine learning models and look forward to building on this experience in future projects.\n"
      ],
      "metadata": {
        "id": "RwhUN0KF6p07"
      }
    }
  ]
}