NLP_Word2Vec_Category_classifier.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "mount_file_id": "1AQCLEZy-q__AuEn9-bxWQlJ9t-XnmD54",
      "authorship_tag": "ABX9TyNg1iQFYDe/JDAcKjt620H3",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/mvoassis/nlp-category-classifier/blob/main/NLP_Word2Vec_Category_classifier.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **NLP - Text category classifier**\n",
        "\n",
        "#### *Marcos V. O. Assis (mvoassis@gmail.com)*\n",
        "\n",
        "***\n",
        "> ## Objective: \n",
        "\n",
        "Generate an algorithm capable of classifying articles into categories based on their titles.\n",
        "\n",
        "> ## Language: \n",
        "\n",
        "For development and testing, words in Portuguese-BR were used.\n",
        "\n",
        "> ## Approach:\n",
        "\n",
        "1. Download Word2Vec pre-trained CBOW model from [NILC](http://nilc.icmc.usp.br/nilc/index.php/repositorio-de-word-embeddings-do-nilc) (cbow_s300)\n",
        "2. Load model using Gensin library.\n",
        "3. Vectorization of the article's titles using NLTK library and CBOW model.\n",
        "4. Training a Logistic Regression classification model using Scikit Learn library.\n",
        "\n",
        "> ## Categories considered:\n",
        "\n",
        "* Mundo (World)\n",
        "* Cotidiano (Daily life)\n",
        "* Mercado (Market)\n",
        "* Esporte (Sports)\n",
        "* Ilustrada (Illustrated)\n",
        "* Colunas (Column)"
      ],
      "metadata": {
        "id": "6Z6yf_QaFPxE"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Importing Training and Testing data\n",
        "\n",
        "Datasets containing articles title, text, date, category, subcategory and link. "
      ],
      "metadata": {
        "id": "Qo6PFXzGFs8o"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!unzip '/content/drive/MyDrive/Curso_Word2Vec/cbow_s300.zip'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qL8FjpzF5abu",
        "outputId": "30567ba3-98f8-4569-ed51-20728674d70c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  /content/drive/MyDrive/Curso_Word2Vec/cbow_s300.zip\n",
            "  inflating: cbow_s300.txt           \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "1E1YDMD1t5qS"
      },
      "outputs": [],
      "source": [
        "import pandas as pd"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train = pd.read_csv('https://caelum-online-public.s3.amazonaws.com/1638-word-embedding/treino.csv')\n",
        "test = pd.read_csv('https://caelum-online-public.s3.amazonaws.com/1638-word-embedding/teste.csv')"
      ],
      "metadata": {
        "id": "uzthwsX5uBVM"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        },
        "id": "3dkI1BfruJmC",
        "outputId": "c344fe41-9cd0-4e3d-e3af-0fc63b08b59a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                               title  \\\n",
              "0  Após polêmica, Marine Le Pen diz que abomina n...   \n",
              "1  Macron e Le Pen vão ao 2º turno na França, em ...   \n",
              "2  Apesar de larga vitória nas legislativas, Macr...   \n",
              "3  Governo antecipa balanço, e Alckmin anuncia qu...   \n",
              "4  Após queda em maio, a atividade econômica sobe...   \n",
              "\n",
              "                                                text        date   category  \\\n",
              "0  A candidata da direita nacionalista à Presidên...  2017-04-28      mundo   \n",
              "1  O centrista independente Emmanuel Macron e a d...  2017-04-23      mundo   \n",
              "2  As eleições legislativas deste domingo (19) na...  2017-06-19      mundo   \n",
              "3  O número de ocorrências de homicídios dolosos ...  2015-07-24  cotidiano   \n",
              "4  A economia cresceu 0,25% no segundo trimestre,...  2017-08-17    mercado   \n",
              "\n",
              "  subcategory                                               link  \n",
              "0         NaN  http://www1.folha.uol.com.br/mundo/2017/04/187...  \n",
              "1         NaN  http://www1.folha.uol.com.br/mundo/2017/04/187...  \n",
              "2         NaN  http://www1.folha.uol.com.br/mundo/2017/06/189...  \n",
              "3         NaN  http://www1.folha.uol.com.br/cotidiano/2015/07...  \n",
              "4         NaN  http://www1.folha.uol.com.br/mercado/2017/08/1...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-ead9b79c-634c-47c6-a13e-7f2f9f8ff54b\">\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>text</th>\n",
              "      <th>date</th>\n",
              "      <th>category</th>\n",
              "      <th>subcategory</th>\n",
              "      <th>link</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Após polêmica, Marine Le Pen diz que abomina n...</td>\n",
              "      <td>A candidata da direita nacionalista à Presidên...</td>\n",
              "      <td>2017-04-28</td>\n",
              "      <td>mundo</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://www1.folha.uol.com.br/mundo/2017/04/187...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Macron e Le Pen vão ao 2º turno na França, em ...</td>\n",
              "      <td>O centrista independente Emmanuel Macron e a d...</td>\n",
              "      <td>2017-04-23</td>\n",
              "      <td>mundo</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://www1.folha.uol.com.br/mundo/2017/04/187...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Apesar de larga vitória nas legislativas, Macr...</td>\n",
              "      <td>As eleições legislativas deste domingo (19) na...</td>\n",
              "      <td>2017-06-19</td>\n",
              "      <td>mundo</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://www1.folha.uol.com.br/mundo/2017/06/189...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Governo antecipa balanço, e Alckmin anuncia qu...</td>\n",
              "      <td>O número de ocorrências de homicídios dolosos ...</td>\n",
              "      <td>2015-07-24</td>\n",
              "      <td>cotidiano</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://www1.folha.uol.com.br/cotidiano/2015/07...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Após queda em maio, a atividade econômica sobe...</td>\n",
              "      <td>A economia cresceu 0,25% no segundo trimestre,...</td>\n",
              "      <td>2017-08-17</td>\n",
              "      <td>mercado</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://www1.folha.uol.com.br/mercado/2017/08/1...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ead9b79c-634c-47c6-a13e-7f2f9f8ff54b')\"\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-ead9b79c-634c-47c6-a13e-7f2f9f8ff54b 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-ead9b79c-634c-47c6-a13e-7f2f9f8ff54b');\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": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Iwd1zD7JuPcp",
        "outputId": "fe25fa08-e318-4dfe-a24d-21028acf7427"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 90000 entries, 0 to 89999\n",
            "Data columns (total 6 columns):\n",
            " #   Column       Non-Null Count  Dtype \n",
            "---  ------       --------------  ----- \n",
            " 0   title        90000 non-null  object\n",
            " 1   text         90000 non-null  object\n",
            " 2   date         90000 non-null  object\n",
            " 3   category     90000 non-null  object\n",
            " 4   subcategory  17175 non-null  object\n",
            " 5   link         90000 non-null  object\n",
            "dtypes: object(6)\n",
            "memory usage: 4.1+ MB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T41_9SauDjsb",
        "outputId": "7af81fe8-bb20-480e-fd83-69fcb186f921"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 20513 entries, 0 to 20512\n",
            "Data columns (total 6 columns):\n",
            " #   Column       Non-Null Count  Dtype \n",
            "---  ------       --------------  ----- \n",
            " 0   title        20513 non-null  object\n",
            " 1   text         20513 non-null  object\n",
            " 2   date         20513 non-null  object\n",
            " 3   category     20513 non-null  object\n",
            " 4   subcategory  6794 non-null   object\n",
            " 5   link         20513 non-null  object\n",
            "dtypes: object(6)\n",
            "memory usage: 961.7+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train.category.unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5aIjFfW3uQaa",
        "outputId": "af30ba77-cee9-4bf0-fbc5-5b1268941aaa"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['mundo', 'cotidiano', 'mercado', 'esporte', 'ilustrada', 'colunas'],\n",
              "      dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "\n",
        "text = [ 'tenha um bom dia',\n",
        "         'tenha um ótimo dia',\n",
        "         'tenha um péssimo dia']\n",
        "\n",
        "vectorizer = CountVectorizer()\n",
        "vectorizer.fit(text)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RFFfb8Ecw3AV",
        "outputId": "501e41ec-076e-4b54-94ee-d3d4df61c362"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CountVectorizer()"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(vectorizer.vocabulary_)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ejZQuYNKxi1N",
        "outputId": "4efab577-117d-4363-d4c3-c997e84775dd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'tenha': 3, 'um': 4, 'bom': 0, 'dia': 1, 'ótimo': 5, 'péssimo': 2}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(vectorizer.transform(['bom']).toarray())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MGX-uZh8xwGt",
        "outputId": "ffce7e08-0836-45e4-d0c1-bb9263de2ec7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[1 0 0 0 0 0]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(vectorizer.transform(['ótimo']).toarray())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "y5xvRxHwx-TL",
        "outputId": "685cd280-9666-460c-b1a8-9ed55130227f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[0 0 0 0 0 1]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Importing Word2Vec model \n",
        "\n",
        "Downloading and Importing Word2Vec pre-trained CBOW model from NILC from Google Drive."
      ],
      "metadata": {
        "id": "MnzjgtAdGER4"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "CBOW - Continuous Bag of Words -> Have the context (before and after), predict the word.\n",
        "\n",
        "Skip Gram -> Have the word, predict the context (before and after)"
      ],
      "metadata": {
        "id": "0eZM5Nv118_s"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "with open('cbow_s300.txt') as f:\n",
        "  for linha in range(10):\n",
        "    print(next(f))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ht5Iioo0yTfQ",
        "outputId": "b11a90cd-b6fb-4a79-8361-663e44970ccd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "929606 300\n",
            "\n",
            "</s> -0.001667 -0.000158 -0.000026 0.001300 -0.000796 0.001527 0.000046 0.000584 0.000449 -0.000100 0.000353 0.001251 0.001069 0.000506 0.000574 0.000838 -0.000930 -0.001220 0.000317 0.001315 -0.001120 0.001373 -0.000040 -0.001580 0.000421 -0.000667 -0.001556 -0.000746 0.001604 0.001157 -0.000027 0.000354 0.000358 -0.000527 -0.000573 -0.001512 -0.001557 -0.001637 0.001617 -0.001511 -0.001022 -0.001426 0.001086 -0.001033 0.000593 0.000724 0.000627 -0.000450 -0.001140 0.000333 0.000524 0.001541 0.000284 0.000617 -0.000807 -0.000088 -0.000364 0.001126 -0.001230 -0.001138 -0.001280 0.001330 0.001257 0.000576 0.000764 0.000684 0.001008 -0.000215 -0.000629 -0.001228 -0.001557 -0.000311 -0.000246 0.000045 0.001136 -0.000645 -0.000549 0.001099 0.000858 -0.000886 0.000553 0.000303 0.001433 0.000732 0.001321 -0.000894 -0.000700 -0.000661 -0.001484 -0.000950 -0.001556 -0.000809 0.000348 -0.000068 0.000724 -0.000569 -0.000161 -0.001628 -0.001437 -0.000259 -0.000296 -0.001571 0.000149 0.000847 0.000613 0.000802 0.001507 0.001015 0.000377 0.000255 -0.000458 -0.000777 -0.001561 0.001601 -0.001520 -0.001210 0.000106 0.000714 0.000392 0.001311 -0.001192 -0.000090 -0.001097 0.000424 -0.000954 -0.001272 -0.001178 0.000036 -0.000181 0.000331 -0.001453 -0.001488 -0.001033 -0.000377 0.000257 -0.001418 0.001109 0.000722 0.000936 -0.000113 0.001215 -0.000263 0.000652 0.001190 -0.000258 0.001391 0.001213 0.000783 -0.001202 0.000470 -0.000879 0.000688 -0.001163 -0.001105 0.001497 0.001304 -0.001322 -0.001501 0.001377 0.001439 0.000884 0.000484 0.001239 -0.001578 0.000981 -0.000318 -0.001180 -0.001375 -0.001491 0.001057 -0.001028 0.000893 0.001028 0.000772 0.001636 -0.000331 -0.000247 -0.001006 -0.000329 0.000837 0.000605 -0.000959 0.001410 0.000488 0.001167 -0.000293 -0.001188 -0.000001 0.001135 0.001141 0.001504 0.000198 -0.001060 0.001551 -0.000003 -0.001474 -0.000391 -0.000880 0.000433 -0.000976 -0.001417 0.000563 -0.001188 0.000593 0.001584 -0.001602 -0.000439 -0.001148 -0.001256 0.001185 -0.000738 0.001543 -0.000846 -0.001029 -0.000641 -0.001587 0.001439 -0.001251 0.000942 -0.001414 -0.001106 0.001087 -0.000027 0.000757 -0.000159 -0.001014 -0.000891 0.000024 -0.000238 0.000157 -0.001067 0.000902 -0.001050 -0.000428 -0.001606 -0.000988 0.001391 0.001165 -0.000113 -0.001000 -0.000055 -0.001369 0.000684 0.000715 0.001407 0.000613 0.001389 0.001315 -0.000130 -0.001044 0.000175 -0.000035 0.000959 -0.000345 0.001209 -0.001251 -0.001219 0.001231 -0.000996 -0.001388 0.001038 0.001336 -0.001066 -0.000881 -0.001066 -0.001466 -0.000274 0.000201 0.000401 0.000132 0.000588 0.000589 -0.000128 0.001073 0.001197 0.000109 0.000770 0.001221 0.000996 -0.001174 0.000135 -0.001134 -0.001385 -0.000311 -0.001631 -0.000564 0.001162 -0.000322 -0.000469 0.001312 -0.001402 0.000239 0.000184 0.001300 0.000021 -0.001065 0.000047 -0.000301 0.001336 0.000332\n",
            "\n",
            ", -0.061483 -0.094368 -0.008557 -0.034702 0.021108 -0.011873 -0.041133 -0.095925 0.034668 -0.085286 0.076174 0.003314 0.019222 -0.038695 -0.008963 0.053399 0.162935 0.050372 -0.020163 -0.027230 0.061531 0.060840 0.074610 -0.056173 0.007621 -0.055220 0.018008 0.026096 0.033154 -0.082612 -0.081761 0.164978 -0.034423 0.003094 -0.018217 -0.087445 -0.074446 -0.000142 0.004218 0.036585 0.016095 -0.129141 -0.119698 -0.053717 0.005053 -0.114520 -0.017219 0.023693 -0.024115 0.053125 0.024658 -0.037689 0.012078 0.112701 0.028037 0.047618 -0.024196 0.050112 -0.073095 -0.090859 -0.030613 -0.109599 0.037756 0.063827 0.022537 -0.029640 -0.016311 0.021875 0.064882 0.039002 -0.082970 -0.043166 0.013695 -0.043153 -0.082203 -0.020087 -0.100360 0.007033 -0.074208 -0.067789 -0.024897 -0.020358 0.041731 0.101332 0.020217 -0.032473 0.087827 -0.033611 -0.150526 -0.016615 0.021147 -0.025058 0.097833 0.065067 0.051287 -0.079191 0.089563 -0.008436 -0.038352 0.019787 -0.058452 -0.009696 -0.051077 -0.112245 0.024886 -0.015172 -0.129670 0.068672 0.068483 0.009049 0.007055 -0.032763 0.001527 0.054264 0.029924 -0.023482 0.047470 0.008044 0.014534 0.071155 0.016700 -0.027491 -0.155782 0.039370 0.116605 -0.001262 -0.026638 -0.067078 0.078015 -0.066153 -0.039303 0.009535 -0.055698 -0.022250 0.046948 0.053810 0.038096 0.032157 -0.075257 0.008125 -0.034598 -0.020667 -0.003153 0.032491 -0.031064 0.030744 -0.049023 -0.046149 0.000792 0.010385 -0.057119 -0.122554 0.003210 -0.054363 -0.100899 0.069873 -0.009758 0.055455 0.049243 0.008346 -0.016087 0.093572 0.024125 0.058736 0.037243 -0.007478 0.032175 -0.054205 0.008798 0.032326 0.028384 -0.032259 -0.041842 -0.058281 -0.025145 0.011097 0.023598 -0.033376 0.026204 0.032505 -0.009283 0.041076 0.055565 -0.081757 -0.010077 0.058251 -0.014379 -0.040951 0.006938 -0.004179 0.052006 -0.063725 0.035674 -0.066554 0.030910 -0.004032 0.077445 0.029495 -0.064931 0.040263 -0.055423 -0.021571 0.086767 -0.003583 0.073308 -0.043991 0.022503 -0.028692 -0.063626 -0.048238 0.013439 -0.043673 -0.101352 -0.004321 0.125507 0.088486 0.042756 -0.014497 -0.053445 0.021800 0.038406 -0.034023 -0.074428 -0.132825 0.082152 -0.068497 0.004738 0.047527 -0.073890 0.051089 -0.055886 -0.047786 0.040247 -0.053966 -0.015752 0.099451 0.008218 -0.010716 -0.031540 0.036168 0.054244 0.051809 0.035158 0.043006 -0.027902 0.000130 0.103397 -0.114831 -0.036648 -0.036143 0.024432 0.084740 0.001801 0.044475 -0.035746 -0.024109 0.051210 -0.025769 0.016073 -0.000351 -0.029183 -0.075292 0.042163 0.025010 -0.041439 -0.059192 0.026617 -0.040852 0.034697 0.014691 -0.057382 0.046141 0.070360 0.045274 0.065880 0.011023 -0.031292 -0.015784 -0.023421 -0.042788 0.019669 0.035010 0.036188 -0.058060 -0.093562 0.030321 -0.054753 0.097162 0.001134 0.018939 -0.150218 -0.009928 0.051118 0.105212 -0.055051 -0.047959 -0.136800 -0.003198 0.068969 -0.022456\n",
            "\n",
            "de -0.232068 0.066729 0.103946 -0.072608 0.126237 -0.004782 -0.025139 -0.141489 -0.069438 -0.071078 0.175772 -0.017257 0.094824 0.011020 0.029226 -0.010670 0.144973 0.105333 -0.088273 -0.070952 0.054747 -0.048955 -0.047809 -0.030763 -0.052293 -0.003596 0.078465 0.144430 0.129697 -0.078427 -0.025080 0.212887 0.119806 0.101703 -0.142488 -0.031272 -0.026594 -0.109429 -0.154688 -0.121492 -0.103781 -0.177948 -0.125544 -0.120136 0.031658 0.054160 0.060493 -0.115676 0.132511 -0.001668 -0.125569 -0.124120 0.029025 0.006200 -0.007915 0.058489 0.122710 0.004660 -0.082200 0.123740 0.052747 -0.065657 0.154747 0.155071 0.003547 -0.154675 -0.035110 0.119117 0.080579 0.044262 0.048753 0.054975 -0.064624 -0.046866 0.028242 0.144197 0.029916 -0.024569 0.195781 -0.028608 -0.008233 -0.032542 0.012042 0.011934 -0.068173 0.037028 0.018363 -0.038244 -0.091176 -0.019795 0.051640 -0.008025 -0.025825 -0.001686 -0.069845 -0.039987 0.066244 0.088229 -0.171358 -0.030783 0.082344 0.059267 0.000550 -0.216285 -0.201356 -0.131063 -0.033069 -0.057126 -0.127576 -0.099251 -0.044669 0.035363 -0.119503 -0.090019 -0.006226 -0.000546 0.020731 0.016140 -0.007438 0.032885 -0.161048 0.105681 -0.115705 -0.078013 0.102293 -0.107958 -0.025926 -0.067986 0.017502 -0.054865 -0.007930 -0.006999 -0.000949 -0.026097 0.106295 0.115368 -0.033742 -0.046025 0.009641 -0.065890 0.198822 0.078333 0.062547 0.044947 -0.098424 0.140951 0.109045 -0.004204 -0.174950 0.034234 0.040793 -0.210289 0.104861 0.083616 -0.042511 -0.031439 -0.113108 0.000401 -0.171665 0.097890 0.008065 0.080684 -0.009576 -0.042125 0.177337 -0.005585 -0.004165 0.069755 -0.100323 -0.023289 0.039458 -0.089292 0.069543 -0.009576 0.011562 -0.076654 0.009681 -0.018808 -0.047881 0.026709 -0.021200 -0.032544 0.100362 0.037157 0.005169 -0.001280 -0.064330 -0.049024 -0.002354 0.004443 0.049129 0.026813 0.034249 0.068827 -0.038583 -0.116914 -0.107378 0.046983 0.038218 -0.082186 -0.124208 0.066872 -0.081745 -0.016516 0.016110 0.046844 0.176223 0.083604 -0.086893 -0.114739 -0.159588 0.007700 0.004887 0.006024 -0.020026 0.045816 0.033604 0.054474 0.089348 -0.024353 0.104835 0.084334 0.052662 -0.041422 -0.027877 0.002816 0.150068 -0.052310 0.017154 -0.152326 0.067753 -0.082644 0.119430 -0.012345 0.082965 -0.005791 -0.082770 -0.030068 -0.037331 -0.075347 0.035060 -0.092023 -0.001051 0.012675 0.128757 -0.048579 -0.078527 -0.134126 -0.060009 0.096834 -0.045947 0.132404 -0.030576 -0.006618 -0.088179 -0.124597 -0.095311 -0.086943 0.007010 0.059925 -0.077005 -0.035458 -0.017957 0.081104 0.060141 0.152996 0.083737 -0.025909 0.005420 -0.006300 -0.075839 -0.012399 -0.001624 0.039090 -0.040755 -0.013051 -0.072733 -0.048062 -0.082573 -0.013851 -0.120222 0.011452 -0.083538 0.015996 -0.110800 0.012405 0.045823 0.026705 -0.040789 0.064309 0.007381 -0.037854 0.076050 0.104702 0.010307 -0.103478 0.085227 0.064233 -0.015908 -0.047998\n",
            "\n",
            ". 0.027867 0.077901 -0.054738 -0.095938 -0.010536 0.015269 -0.005752 -0.048440 -0.104021 0.054583 0.050108 0.054979 0.071112 -0.056665 -0.140868 0.203626 0.211330 0.137270 -0.038069 -0.108607 0.060027 0.052436 0.126758 -0.078212 0.008864 -0.028087 0.029884 0.114899 -0.058657 -0.166441 -0.188866 0.294844 -0.147703 -0.095883 -0.104021 -0.097608 -0.172317 0.035795 -0.019025 0.057503 -0.030564 -0.125182 -0.257359 -0.106566 -0.002708 -0.035769 -0.072053 -0.005640 0.016920 0.155101 -0.226088 -0.070904 -0.110280 -0.075773 -0.190123 0.197397 -0.108603 0.155954 0.096544 -0.183301 0.008154 -0.195997 0.035284 -0.025561 0.159142 -0.141005 0.140643 -0.160588 0.140709 0.059974 -0.012396 -0.074953 -0.038325 -0.165941 -0.226005 -0.043738 0.052905 0.085072 -0.165731 0.016894 -0.136569 0.088165 -0.032044 0.022789 0.051207 0.015907 0.093662 0.037576 -0.286139 -0.107587 0.165001 0.095293 0.181427 0.038072 0.078180 -0.116599 0.138464 0.138386 0.020201 -0.028087 0.038829 -0.040074 -0.044944 -0.179895 0.035044 0.090347 -0.036088 0.154828 0.090421 0.172175 -0.127294 0.060227 0.164694 0.141367 0.049326 -0.136940 0.060633 -0.089581 -0.069639 0.043680 0.141455 -0.060588 -0.118859 0.181506 -0.051302 0.165763 0.151366 -0.191232 -0.103888 -0.077344 -0.016534 0.073607 -0.008464 0.216368 0.038251 0.002552 -0.112613 -0.049596 0.018040 -0.032907 -0.131335 -0.049530 0.057485 0.129446 -0.017884 0.075667 -0.061577 -0.015305 0.007649 0.116636 -0.132568 -0.244250 -0.048748 0.003139 -0.123311 0.018462 0.126955 0.079116 -0.067653 0.019126 0.044182 0.037610 0.199809 -0.118487 0.011276 -0.061652 -0.034426 -0.120148 -0.171490 0.009182 0.182921 -0.165212 0.149898 0.062080 -0.129040 0.051808 -0.075070 -0.068688 0.018707 -0.008050 0.108429 -0.111089 0.021230 -0.125982 0.077740 0.140251 -0.051900 0.036107 0.019580 -0.084050 0.065747 0.015718 -0.016900 -0.026964 0.126565 -0.087491 0.069504 0.108362 -0.006355 0.118521 -0.002510 0.060457 0.082848 0.000294 -0.037938 -0.026097 0.084446 0.015568 -0.050540 -0.117080 0.100227 -0.046050 -0.059615 0.086632 0.097266 0.084578 -0.036891 -0.033246 0.021302 -0.149581 -0.006353 -0.035713 -0.177163 -0.016629 -0.028634 -0.147015 -0.017861 0.009931 -0.181849 0.065564 -0.007616 0.094861 0.089787 0.005743 -0.127185 0.093027 -0.107247 -0.019161 -0.201082 -0.008504 0.126500 0.034318 0.011876 0.012696 -0.090981 0.012814 0.038388 -0.274763 -0.123303 -0.073078 0.027702 0.035841 -0.080954 0.091305 0.029009 -0.006000 0.121087 0.016678 -0.070181 -0.080839 -0.016332 -0.081390 -0.059594 0.131390 0.039603 -0.222503 -0.039382 -0.071041 0.267399 -0.010011 -0.077326 -0.044107 0.052901 0.023623 -0.087243 0.011539 -0.006381 -0.074194 -0.041591 -0.061872 -0.098552 -0.094357 -0.062879 -0.150341 -0.087596 0.330078 0.028435 0.087784 0.034434 -0.008122 -0.258485 -0.109828 0.013279 0.043340 0.136803 -0.081573 -0.136733 -0.081884 0.306241 0.093106\n",
            "\n",
            "a -0.019764 -0.096043 -0.010960 0.102012 -0.101848 -0.010257 0.004692 -0.165735 -0.179723 0.067659 0.037108 0.100931 -0.211779 0.163603 -0.042376 -0.021683 -0.014900 0.101581 -0.053583 0.065435 0.126251 0.062320 -0.057445 -0.107130 -0.037044 -0.072958 -0.066655 -0.067962 -0.070879 -0.187941 0.065208 0.102302 -0.024876 0.151433 0.012228 -0.065570 -0.063793 0.015519 0.046230 0.174824 -0.100376 -0.081244 -0.039337 0.017851 0.125168 -0.096404 -0.133070 -0.023451 0.117965 0.005949 0.038928 0.060488 -0.014678 0.024891 -0.001455 0.048539 0.133535 0.084564 0.018362 -0.054294 -0.235691 -0.151528 0.043101 -0.097685 -0.034775 -0.127507 0.010346 -0.010458 0.047631 0.002045 -0.013826 -0.139644 0.128220 -0.020997 -0.057176 0.007458 -0.033133 -0.083235 -0.075585 -0.079414 -0.023531 0.076794 0.051968 0.026857 0.007574 -0.039602 0.027458 -0.146083 0.008662 0.113647 0.007963 0.023413 0.088046 -0.079530 0.027995 -0.017113 -0.056750 0.069447 0.094213 0.123728 0.010051 -0.054795 0.041003 -0.044294 -0.066911 -0.025221 0.059618 0.023494 -0.095354 0.029266 0.047979 0.079102 -0.061757 0.116363 0.011112 0.136784 -0.027020 -0.002793 -0.092629 -0.047217 0.111263 0.102819 0.090932 -0.085343 -0.026490 -0.033315 -0.028970 -0.118673 -0.067360 -0.207183 -0.076011 0.017469 -0.078908 0.031023 0.090820 -0.054443 0.062466 0.065952 -0.028265 0.051329 0.090671 0.106923 0.028011 0.014720 -0.036025 0.099207 -0.013510 -0.147551 -0.077657 -0.099269 0.001171 0.059405 -0.006946 0.123997 0.023131 0.102308 -0.040784 0.027455 -0.181898 0.095958 0.155467 -0.077250 -0.124941 0.058529 0.035851 -0.004447 0.035127 -0.024015 -0.029101 -0.107558 0.055650 -0.017027 0.008997 -0.016504 0.018768 -0.080671 0.029003 0.013937 0.002876 -0.064262 0.001246 0.073494 -0.013986 -0.100303 0.027376 0.030375 -0.177618 0.016413 -0.008711 -0.257308 -0.031904 0.068326 0.038748 0.006762 -0.045245 0.038158 0.093055 -0.069432 -0.140442 -0.030486 -0.097529 0.251017 -0.059000 -0.037363 -0.020030 0.116937 0.018191 -0.005464 0.016485 0.009930 0.007318 -0.131821 -0.005316 -0.063905 0.066936 -0.007235 0.066709 0.153262 -0.087015 -0.196866 0.042973 0.124447 0.034619 -0.021113 0.029881 -0.072679 0.068707 -0.033823 -0.140311 -0.158137 0.215081 -0.020170 -0.062018 -0.038955 0.158768 -0.155656 0.016471 -0.071324 0.069629 -0.070692 0.114725 0.114054 0.154133 -0.143244 -0.082015 -0.070578 0.094558 -0.073491 -0.084676 0.089027 -0.053428 0.034783 0.007488 0.008849 0.137206 -0.061942 -0.155206 -0.022297 -0.054165 0.206418 0.058566 0.033847 0.039555 0.042072 -0.003160 0.350700 -0.067416 -0.089351 0.147734 0.026404 -0.128322 -0.143885 -0.189820 0.162824 -0.038651 -0.045660 0.083507 0.118510 -0.054740 -0.053446 -0.106545 -0.087144 -0.159905 0.006153 0.057539 0.079486 -0.015659 -0.014220 -0.013549 -0.122177 -0.154875 0.319072 -0.251233 0.070468 -0.084041 -0.067580 -0.163386 -0.026553 -0.033245 -0.035199\n",
            "\n",
            "o 0.050016 0.094213 -0.234393 0.057870 -0.255710 -0.099076 0.162541 -0.073166 -0.072231 0.087253 0.154162 -0.021438 0.095246 0.086680 0.027481 0.055959 0.176635 -0.158901 -0.076106 -0.036695 0.115689 -0.039255 -0.116532 -0.143528 0.059030 0.049429 -0.009721 0.204578 -0.020946 -0.254149 -0.167956 0.024087 -0.094412 -0.066287 -0.085987 -0.081115 -0.035314 -0.000736 0.069831 -0.034319 -0.041741 0.007945 0.106777 0.119971 -0.051358 0.152951 -0.109669 0.102599 0.102598 0.044424 -0.088313 0.142374 0.042868 -0.042467 0.199348 -0.234265 -0.044845 -0.109492 0.005783 -0.045873 -0.049017 -0.037446 0.105329 -0.163747 -0.139140 0.153179 -0.117726 -0.113436 0.010813 0.097110 0.074618 0.041083 0.082912 -0.052041 0.009895 0.160828 0.058190 -0.215907 -0.024543 0.042466 -0.011712 0.064930 0.097498 -0.080108 0.019837 -0.059179 0.077248 -0.135312 0.132748 0.069083 0.075949 -0.232743 0.078279 -0.078944 0.116486 0.004099 0.057786 -0.125938 0.068976 0.176170 0.044454 0.085901 0.033849 0.057793 -0.083662 0.061258 -0.037814 0.050392 -0.067302 -0.102103 -0.011153 0.029286 0.032976 0.003400 0.061664 0.076337 -0.097293 0.026432 -0.059397 -0.160978 0.069399 0.307550 0.069981 -0.122410 0.003086 0.027841 0.134213 -0.018279 -0.025608 -0.156602 0.062460 -0.051757 0.030337 -0.042270 0.007516 0.055478 0.075345 0.112685 -0.047181 0.001009 0.031433 -0.134699 -0.034702 -0.082685 -0.182748 -0.097813 0.053774 -0.067876 -0.090774 -0.035145 0.080364 -0.242563 0.024905 -0.012293 0.110051 0.103592 0.138403 -0.035121 0.037955 -0.028042 0.005687 0.042462 0.045434 0.127590 -0.049950 0.079609 -0.152268 -0.102837 -0.083840 -0.132681 -0.073176 0.031672 0.104991 0.082840 0.142188 -0.047643 0.054661 -0.112325 0.006551 0.079335 0.011705 -0.042121 0.086597 0.088994 0.079949 -0.002677 0.106816 -0.055035 0.068153 -0.167192 0.009149 -0.028594 -0.039560 0.001607 -0.122615 0.106024 0.107250 0.050540 -0.043291 0.061397 0.001380 0.054964 0.040893 -0.045046 0.002312 0.032035 -0.026698 -0.042064 0.144622 0.076241 0.024255 -0.086833 0.098275 -0.083894 0.086363 -0.004560 -0.029056 0.126784 0.071785 -0.025692 0.000966 0.100620 0.030299 -0.163968 0.042070 -0.198341 0.099603 -0.113758 0.124685 -0.069966 -0.165131 -0.134782 -0.020419 -0.112747 0.123515 -0.045274 0.167949 -0.109681 -0.146224 0.075453 0.054352 0.023153 -0.062161 -0.059457 -0.124327 0.041855 0.205124 -0.083616 -0.062827 0.017066 0.025000 0.079311 0.018874 -0.055492 -0.129246 0.099183 -0.022749 -0.079510 -0.130005 0.012160 0.060842 0.018816 -0.180103 0.057191 0.018971 0.216519 -0.003726 0.155861 0.112271 0.091379 -0.073707 -0.059254 0.011068 -0.005690 -0.098677 -0.082454 0.105838 0.042144 0.100894 0.096960 -0.228625 0.064737 0.022971 -0.001549 -0.105277 0.149519 -0.055187 0.201671 0.098550 0.056229 -0.107189 -0.005223 -0.077126 0.020865 -0.098107 0.064205 -0.174567 0.196321 0.233813 0.038634\n",
            "\n",
            "e -0.168088 -0.001531 0.017038 -0.137315 -0.026399 -0.018215 0.026076 0.019779 -0.095212 -0.090352 -0.106060 0.061975 -0.000318 -0.002945 0.119734 0.149804 0.108117 0.064784 0.065574 -0.038343 0.159915 -0.006124 -0.038307 -0.094810 -0.019356 -0.045115 -0.121859 0.028930 0.045017 -0.138268 0.099283 0.128659 0.077656 0.032982 -0.086139 0.051552 -0.077685 -0.066857 -0.045233 -0.185625 0.013728 -0.067088 -0.061796 0.091931 0.007931 -0.033784 -0.026604 0.026487 0.002281 0.000939 -0.043053 -0.082718 -0.101476 0.014253 0.106507 0.059285 -0.034097 0.032848 -0.067852 0.103892 0.020742 -0.049702 0.085734 0.069140 -0.036293 -0.001048 -0.043403 -0.033758 -0.041652 -0.049051 -0.070893 -0.002921 0.096254 -0.046406 -0.016074 0.074894 0.098854 -0.015867 -0.024065 0.048818 -0.058226 -0.031364 0.055631 -0.001012 0.002966 0.017624 0.082472 -0.025849 -0.099736 -0.065253 0.002511 0.026550 -0.046750 0.005665 0.058496 -0.068569 0.110588 -0.003354 -0.133236 -0.140404 0.090284 -0.089065 0.010600 -0.143848 0.023044 0.035726 0.044885 -0.081761 0.020406 0.045142 -0.099682 0.013478 -0.102208 -0.048206 0.040736 0.012733 0.064284 0.052800 0.046179 -0.083605 -0.002064 0.003024 0.078588 -0.075731 -0.038192 -0.010662 -0.062210 -0.060302 0.005740 -0.119290 0.049789 -0.009878 0.033791 0.145393 -0.064118 0.005292 -0.046015 -0.012572 -0.002266 0.018877 -0.114718 0.043831 0.036031 -0.077071 0.079499 0.047130 0.136516 -0.029402 -0.095073 -0.198367 0.077792 -0.092858 -0.115212 0.064883 -0.083748 -0.115535 -0.017903 0.080732 0.088457 0.037072 0.044588 0.042071 0.064027 0.035189 0.071027 -0.028330 0.026847 -0.005846 -0.003655 -0.019447 0.057641 -0.057623 0.032575 -0.009248 0.103035 -0.087592 0.049022 -0.030181 0.055313 0.001550 0.136222 -0.081518 -0.003322 -0.196934 -0.125240 -0.039119 0.082201 -0.036932 -0.030666 0.101376 -0.008224 0.045433 0.055995 -0.108370 0.120019 0.024427 -0.055956 0.076503 -0.063588 -0.091926 -0.047620 0.071838 -0.050062 0.122101 0.011679 0.015549 0.060463 -0.174818 0.006131 -0.035602 -0.039755 0.044683 -0.002324 -0.096353 0.153145 -0.181759 0.059133 0.008242 0.009471 0.004472 -0.005789 0.035746 0.107733 0.019606 0.035323 0.013101 0.078772 0.031495 0.011914 0.035316 0.029965 0.010626 0.024444 -0.036373 0.041982 -0.059195 -0.062521 -0.032445 -0.050086 -0.002740 -0.033702 -0.052157 -0.100889 0.010753 -0.017195 0.076872 -0.114096 -0.134334 0.019812 -0.147168 -0.048026 0.055315 -0.021564 0.092953 -0.070098 -0.009133 -0.021408 -0.043442 -0.059603 -0.026196 -0.046402 -0.024446 0.036482 -0.021195 -0.014285 -0.003113 -0.053620 -0.057722 -0.005912 -0.228597 -0.022520 0.008453 -0.131441 0.074053 0.010990 -0.029988 -0.106588 -0.005512 0.036736 -0.002115 -0.078824 -0.067073 0.034071 -0.037642 0.019106 0.080103 0.068267 -0.052487 -0.070635 0.073949 -0.056421 0.017687 0.011800 -0.014045 0.006226 -0.014210 0.088889 -0.007613 -0.042151 0.091935\n",
            "\n",
            "que 0.105677 -0.012054 -0.134709 -0.013888 -0.080373 -0.084105 0.028820 -0.039461 -0.011410 0.033408 -0.031719 0.198935 0.047353 -0.090613 -0.098891 -0.010202 0.144315 0.151674 0.016502 -0.133896 0.031100 -0.007432 -0.059600 0.059022 -0.038658 0.167877 0.082268 0.217249 -0.072088 0.076910 0.079216 -0.037108 0.014821 -0.078732 -0.035230 0.024507 -0.126809 -0.017172 0.040998 -0.043692 -0.149663 -0.049520 0.024513 -0.063125 0.058306 -0.012108 -0.119330 0.034781 -0.102405 0.088640 -0.272843 -0.059915 -0.108119 -0.004685 0.184494 0.029357 -0.012836 0.087409 -0.093633 0.141491 -0.051050 -0.072004 -0.164353 -0.078488 -0.075351 -0.029836 -0.011346 -0.081588 0.032113 -0.120232 -0.028536 0.010188 0.169394 -0.077409 -0.026955 0.083380 0.010644 0.026978 0.131446 -0.055521 -0.142428 -0.164762 0.016372 -0.254786 -0.015511 -0.044090 -0.091434 0.042059 -0.182369 -0.057621 -0.113482 -0.031757 0.036860 0.035760 -0.084302 -0.244284 0.014317 -0.110019 0.076959 -0.003242 0.058425 -0.068168 -0.150191 -0.054872 0.067636 -0.130009 0.033544 -0.028672 -0.029417 -0.003639 0.008028 0.067533 0.015459 0.057868 -0.095344 0.023596 0.005216 -0.068982 0.110498 -0.045724 0.067068 -0.103614 -0.010749 0.079515 -0.160577 -0.037640 -0.090058 0.194754 -0.137685 0.048330 0.023402 -0.080694 -0.076719 0.081176 0.145217 -0.312484 -0.158713 0.003970 0.017241 0.020559 0.019566 -0.020686 0.027909 -0.088033 0.055311 0.114986 0.046477 -0.026289 -0.085503 0.072325 0.033167 -0.027302 0.043202 -0.125175 -0.046430 -0.089749 -0.052517 -0.026923 -0.157140 0.142859 -0.035472 0.053636 -0.015367 -0.041692 0.004743 -0.053162 -0.073490 -0.031711 0.009967 -0.034655 0.209237 -0.052801 0.121190 -0.048079 0.203458 0.037113 -0.095015 0.143972 0.046959 -0.014909 0.169301 -0.034984 0.006088 -0.180444 -0.098740 -0.082224 0.042493 -0.048264 0.034078 -0.155237 0.049812 0.108571 -0.103458 0.015055 -0.005921 -0.020380 -0.059664 0.063496 0.120677 0.028720 0.010933 -0.034094 -0.033195 0.065065 -0.150067 -0.127015 -0.115948 -0.168308 -0.079605 0.112220 -0.024935 0.008233 -0.050985 -0.047235 -0.041944 0.099358 -0.094031 0.008722 -0.044243 0.091927 0.029639 0.053425 -0.020329 -0.007264 0.029055 -0.027748 0.021123 0.023926 -0.068259 -0.034626 0.010976 -0.027882 0.074465 0.017960 -0.142224 0.010472 -0.005171 0.001003 0.031482 0.048073 0.122005 -0.012546 -0.127375 0.137744 -0.043389 -0.067031 -0.028339 -0.102349 -0.001278 -0.075125 -0.209901 0.022713 0.123317 -0.005108 0.206613 -0.119748 0.148568 0.189863 -0.005551 0.042991 0.024824 0.095960 0.049202 -0.115647 -0.075263 -0.101255 0.047671 0.006936 -0.147872 -0.048649 0.152138 0.090100 -0.054797 0.180955 -0.016752 -0.090723 -0.126502 0.137744 0.065590 -0.095213 -0.141592 -0.081258 0.061414 0.138765 -0.047507 0.155577 -0.004058 -0.054305 -0.064755 -0.155780 0.019290 0.092772 -0.113740 0.062508 -0.108115 -0.104516 0.233178 0.085033 -0.063993 0.107476\n",
            "\n",
            "do -0.180377 0.215497 -0.243985 -0.111583 -0.213779 -0.101028 -0.000539 -0.101256 -0.070914 0.038753 0.046168 -0.014838 0.180611 0.000883 0.060867 0.085324 0.226791 -0.021134 -0.137997 -0.202197 0.148571 -0.050565 -0.078147 -0.128638 -0.060884 0.050100 0.018876 0.214215 0.115665 -0.095654 -0.142162 0.120025 -0.097131 0.072397 -0.038019 0.105100 0.071983 -0.160280 -0.051489 -0.098811 -0.117858 -0.034939 -0.009614 0.107021 -0.047199 -0.005264 -0.000139 0.071706 0.161239 -0.161907 -0.337947 0.006058 0.022506 -0.037415 0.171261 -0.062566 -0.062829 -0.086872 0.049601 0.208633 -0.010851 0.096788 0.242130 -0.103836 -0.078473 -0.008937 -0.192728 0.135349 -0.090459 0.275743 0.134895 0.051969 -0.132400 0.040473 -0.095963 0.027610 0.029590 -0.188700 0.164033 0.032661 -0.109406 0.180492 0.147311 0.025806 -0.079043 -0.050046 0.065651 -0.066955 0.037583 -0.031008 0.012122 -0.225802 0.106807 0.067123 0.035571 -0.204494 -0.004813 -0.022719 -0.267100 0.241056 0.115854 0.094971 -0.032575 0.045279 0.030343 -0.103868 0.013372 0.008258 0.027818 -0.051606 0.060680 0.094947 -0.025672 -0.044170 -0.007713 -0.008499 -0.111320 0.112186 -0.091970 -0.085167 0.001894 0.156088 -0.067075 -0.027825 0.093561 0.070251 0.049796 -0.161722 -0.051384 -0.124694 -0.016951 -0.141747 0.002894 -0.133038 -0.015664 0.080839 -0.090980 0.084050 0.089413 0.021215 0.043560 -0.075679 -0.068300 -0.025250 -0.157061 0.022866 0.055717 -0.019705 0.036159 0.183264 0.073571 -0.227341 -0.020978 0.064482 0.075304 0.010691 -0.068159 0.110613 0.047416 -0.029155 -0.043249 -0.074944 0.011065 -0.047526 0.018594 0.114240 -0.265600 0.007173 -0.118795 -0.109128 0.000016 -0.025266 -0.101829 0.167882 0.163892 -0.112688 -0.195755 -0.162868 0.114179 0.149630 0.075342 0.024940 0.099575 0.154626 0.069327 0.046685 0.046306 -0.138379 0.070196 -0.071198 0.028745 0.052231 -0.034552 -0.065006 0.070413 -0.061854 0.102099 0.063467 -0.002194 0.041076 -0.095035 0.138216 0.123084 -0.022177 0.178009 0.171209 0.193143 -0.170747 0.013212 0.020312 -0.129422 0.079598 -0.068529 -0.147807 0.047131 -0.137531 0.091482 0.030125 0.145402 -0.080191 0.025542 0.178609 0.118927 -0.206678 0.094064 -0.042739 -0.084631 -0.076197 0.253984 -0.273463 -0.026424 0.022400 0.012012 0.061563 0.131845 0.000155 -0.068048 -0.138741 -0.137770 0.143289 -0.017924 -0.111125 -0.055598 -0.012842 0.040329 0.034049 0.148260 -0.121714 -0.091318 0.030290 0.127014 0.037068 -0.070445 -0.136299 -0.232076 -0.028718 0.111683 -0.116630 -0.077730 0.033543 -0.032343 0.091454 0.094135 0.163275 0.123356 0.155053 0.137251 0.054886 0.086155 0.132626 -0.112011 0.103465 -0.085405 -0.075092 0.038307 0.075783 0.028715 0.020250 0.078956 0.012173 -0.058316 0.084967 -0.099896 -0.090466 -0.126150 0.144591 0.025099 0.059846 0.102206 0.092871 -0.004201 -0.062104 0.143311 -0.046741 0.010705 -0.065122 -0.074470 0.072302 0.144289 0.067299\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from gensim.models import KeyedVectors\n",
        "\n",
        "model = KeyedVectors.load_word2vec_format('cbow_s300.txt')"
      ],
      "metadata": {
        "id": "NTd-T3xw8Tvp"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "len(model.get_vector('china'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZZJPc5TJ8b_4",
        "outputId": "bacde40d-ef32-42ee-cbf7-0c3317bc0015"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "300"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.most_similar('china')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TNUEZKEu9znm",
        "outputId": "f3e756b6-2343-4d9a-ed56-bd5ae77ffdb2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('rússia', 0.7320704460144043),\n",
              " ('índia', 0.7241617441177368),\n",
              " ('tailândia', 0.701935887336731),\n",
              " ('indonésia', 0.6860769987106323),\n",
              " ('turquia', 0.6741335988044739),\n",
              " ('malásia', 0.6665689945220947),\n",
              " ('mongólia', 0.6593616008758545),\n",
              " ('manchúria', 0.6581847667694092),\n",
              " ('urss', 0.6581669449806213),\n",
              " ('grã-bretanha', 0.6568098068237305)]"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.most_similar(positive=['brasil', 'argentina'])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3XFm8iSs94v_",
        "outputId": "89387bee-400e-4ab5-d41c-310359ed7438"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('chile', 0.6781662702560425),\n",
              " ('peru', 0.6348033547401428),\n",
              " ('venezuela', 0.6273865103721619),\n",
              " ('equador', 0.6037014722824097),\n",
              " ('bolívia', 0.6017140746116638),\n",
              " ('haiti', 0.5993807315826416),\n",
              " ('méxico', 0.5962306261062622),\n",
              " ('paraguai', 0.5957703590393066),\n",
              " ('uruguai', 0.5903672575950623),\n",
              " ('japão', 0.5893509984016418)]"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Predict plural:\n",
        "\n",
        "nuvens -> nuvem : estrelas -> estrela\n",
        "\n",
        "nuvens + estrela - nuvem -> estrelas"
      ],
      "metadata": {
        "id": "EUwM-9rtG07i"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.most_similar(positive=['nuvens', 'estrela'], negative=['nuvem'])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2X-CZuP0_2-g",
        "outputId": "f2f5f75a-bac6-4789-9e5e-ceffa45e9208"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('estrelas', 0.5497429966926575),\n",
              " ('plêiades', 0.379197895526886),\n",
              " ('colinas', 0.3746805191040039),\n",
              " ('trovoadas', 0.373703271150589),\n",
              " ('sombras', 0.3734194040298462),\n",
              " ('pombas', 0.3726757764816284),\n",
              " ('corredoras', 0.3640727698802948),\n",
              " ('cigarras', 0.36065393686294556),\n",
              " ('galáxias', 0.35754913091659546),\n",
              " ('luas', 0.3575345277786255)]"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Data Bias:\n",
        "\n",
        "médico + mulher - homem = médica? "
      ],
      "metadata": {
        "id": "oDymnTy6IEXi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.most_similar(positive=['médico', 'mulher'], negative=['homem'])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yESsMV4sHcJa",
        "outputId": "98a3cf06-a4a5-41db-9f79-ebe8da2e24ee"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('enfermeira', 0.6180862188339233),\n",
              " ('psicóloga', 0.47447532415390015),\n",
              " ('dama-de-companhia', 0.4738234877586365),\n",
              " ('esposa', 0.46559134125709534),\n",
              " ('parteira', 0.46362876892089844),\n",
              " ('mãe', 0.4581768214702606),\n",
              " ('governanta', 0.45722925662994385),\n",
              " ('madrasta', 0.4569782614707947),\n",
              " ('menina', 0.44435906410217285),\n",
              " ('filha', 0.44341108202934265)]"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "professor + mulher - homem = profassora?"
      ],
      "metadata": {
        "id": "LmsUeNBtIan1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.most_similar(positive=['professor', 'mulher'], negative=['homem'])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "__H5cEiHHuC9",
        "outputId": "096ef8ce-ea13-4b78-c291-599ecf49bdec"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('professora', 0.6192208528518677),\n",
              " ('aluna', 0.5449554324150085),\n",
              " ('esposa', 0.4978231191635132),\n",
              " ('ex-aluna', 0.4884248375892639),\n",
              " ('namorada', 0.4737858772277832),\n",
              " ('enfermeira', 0.4728144109249115),\n",
              " ('filha', 0.467373788356781),\n",
              " ('irmã', 0.45845913887023926),\n",
              " ('ex-namorada', 0.45824766159057617),\n",
              " ('ex-professora', 0.4510470926761627)]"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Text Vectorization "
      ],
      "metadata": {
        "id": "S7OfcGW17b_t"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train.title.iloc[12]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "_jrjXPDTIfi1",
        "outputId": "60aac895-5c5e-49c2-f1a1-5b7c6db3c772"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "\"Daniel Craig será stormtrooper em novo 'Star Wars', diz ator\""
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import nltk\n",
        "nltk.download('punkt')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Wc28YCTq9Qfw",
        "outputId": "f09fc092-ce75-4fce-8776-6c3cc613684f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Unzipping tokenizers/punkt.zip.\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import nltk\n",
        "import string\n",
        "\n",
        "def tokenizer(text):\n",
        "  text = text.lower()\n",
        "  alpha_numeric_list = []\n",
        "\n",
        "  for valid_token in nltk.word_tokenize(text):\n",
        "    if valid_token in string.punctuation: continue\n",
        "    alpha_numeric_list.append(valid_token)\n",
        "\n",
        "  return alpha_numeric_list\n",
        "\n",
        "tokenizer('texto exemplo guarda-roupa\\', 1234.')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yhaswhKM7jNf",
        "outputId": "0d7f19ee-9d36-48cf-c577-8d023600a142"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['texto', 'exemplo', 'guarda-roupa', '1234']"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "string.punctuation"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "VNX-bPRn8Kmg",
        "outputId": "4ce30381-a308-4375-ff8f-63b8b451596e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'!\"#$%&\\'()*+,-./:;<=>?@[\\\\]^_`{|}~'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "def sum_vector_combining(words_numbers):\n",
        "\n",
        "  result_vector = np.zeros(300)\n",
        "\n",
        "  for wn in words_numbers:\n",
        "    result_vector += model.get_vector(wn)\n",
        "\n",
        "\n",
        "  return result_vector"
      ],
      "metadata": {
        "id": "f9v9ViCQ9p5B"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "words_numbers = tokenizer('Texto exemplo')        # still generating errors when a word is not in the vocabulary\n",
        "words_vector = sum_vector_combining(words_numbers)# and when the vocab receive a number\n",
        "print(len(words_vector))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gvkVWnxX_gNg",
        "outputId": "69532263-2309-4570-92b9-60de157b293f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "300\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "NILC preprocessing:\n",
        "\n",
        "- rare word occurency -> normalized with  UNKNOWN\n",
        "- numbers -> normalized with 0 "
      ],
      "metadata": {
        "id": "RdYseT2OCCtu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "def sum_vector_combining(words_numbers):\n",
        "\n",
        "  result_vector = np.zeros(300)\n",
        "\n",
        "  for wn in words_numbers:\n",
        "    try:\n",
        "      result_vector += model.get_vector(wn)\n",
        "    except KeyError:\n",
        "      if wn.isnumeric():\n",
        "        wn = '0'* len(wn)\n",
        "      else:\n",
        "        wn = 'unknown'\n",
        "      result_vector += model.get_vector(wn)\n",
        "  return result_vector"
      ],
      "metadata": {
        "id": "OOucH0PNASwv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "words_numbers = tokenizer('Texto biruleibe')        # ok\n",
        "words_vector = sum_vector_combining(words_numbers)\n",
        "print(len(words_vector))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OGA-Ld0ADKCT",
        "outputId": "a0fb61a8-60fa-4ed4-8804-49693a2bad77"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "300\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def vectors_matrix(texts):\n",
        "  x = len(texts)\n",
        "  y = 300                   #word2vec representation\n",
        "  matrix = np.zeros((x,y))\n",
        "\n",
        "  for i in range(x):\n",
        "    words_numbers = tokenizer(texts.iloc[i])       \n",
        "    matrix[i] = sum_vector_combining(words_numbers)\n",
        "  return matrix"
      ],
      "metadata": {
        "id": "uwe9AFqNILh9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "vet_matrix_train = vectors_matrix(train.title)\n",
        "vet_matrix_test = vectors_matrix(test.title)\n",
        "print(vet_matrix_train.shape)\n",
        "print(vet_matrix_test.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wpRPO3A0JkKe",
        "outputId": "4ee2b900-9d97-4f78-dd40-86371b6157ac"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(90000, 300)\n",
            "(20513, 300)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Classifier model"
      ],
      "metadata": {
        "id": "SEzrgfZqKhdi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.linear_model import LogisticRegression\n",
        "\n",
        "LR = LogisticRegression(max_iter=200)\n",
        "LR.fit(vet_matrix_train, train.category)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fd2PpPzYKC-M",
        "outputId": "ca154445-371a-46e1-fba1-39af9bdc7bab"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(max_iter=200)"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "LR.score(vet_matrix_test, test.category)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Y2M9iWpjLp1n",
        "outputId": "4e78ade7-ef34-4fe8-f58f-bd5759064f26"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7957880368546775"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "label_pred = LR.predict(vet_matrix_test)"
      ],
      "metadata": {
        "id": "HPZfMoY3MSgJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "CR = classification_report(test.category, label_pred)\n",
        "\n",
        "print(CR)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ewu91LSNLxjC",
        "outputId": "7a077e2f-35a1-4b59-ea69-39d934e5689b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "     colunas       0.86      0.71      0.78      6103\n",
            "   cotidiano       0.61      0.79      0.69      1698\n",
            "     esporte       0.92      0.88      0.90      4663\n",
            "   ilustrada       0.13      0.88      0.23       131\n",
            "     mercado       0.84      0.79      0.81      5867\n",
            "       mundo       0.74      0.86      0.79      2051\n",
            "\n",
            "    accuracy                           0.80     20513\n",
            "   macro avg       0.68      0.82      0.70     20513\n",
            "weighted avg       0.83      0.80      0.81     20513\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Comparing with dummy classifier..."
      ],
      "metadata": {
        "id": "Y-RU3cCON8Wn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.dummy import DummyClassifier\n",
        "\n",
        "DC = DummyClassifier()\n",
        "DC.fit(vet_matrix_train, train.category)\n",
        "\n",
        "DC_pred = DC.predict(vet_matrix_test)\n",
        "\n",
        "CR2 = classification_report(test.category, DC_pred)\n",
        "\n",
        "print(CR2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kmt_V4drMnTn",
        "outputId": "31373429-a15f-4b1e-99d2-03eeb9167af7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "     colunas       0.30      1.00      0.46      6103\n",
            "   cotidiano       0.00      0.00      0.00      1698\n",
            "     esporte       0.00      0.00      0.00      4663\n",
            "   ilustrada       0.00      0.00      0.00       131\n",
            "     mercado       0.00      0.00      0.00      5867\n",
            "       mundo       0.00      0.00      0.00      2051\n",
            "\n",
            "    accuracy                           0.30     20513\n",
            "   macro avg       0.05      0.17      0.08     20513\n",
            "weighted avg       0.09      0.30      0.14     20513\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Comparing CBOW with Skip Gram"
      ],
      "metadata": {
        "id": "8Fo8i77USRHG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!unzip '/content/drive/MyDrive/Curso_Word2Vec/skip_s300.zip'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4CKwF0VySVEu",
        "outputId": "dd971c42-0578-4ca5-916f-c1dc54c19348"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  /content/drive/MyDrive/Curso_Word2Vec/skip_s300.zip\n",
            "  inflating: skip_s300.txt           \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_skipgram = KeyedVectors.load_word2vec_format(\"skip_s300.txt\")\n",
        "\n",
        "def sum_vector_combining_skipgram(words_numbers):\n",
        "    result_vector = np.zeros(300)\n",
        "    for wn in words_numbers:\n",
        "        try:\n",
        "            result_vector += model_skipgram.get_vector(wn)\n",
        "\n",
        "        except KeyError:\n",
        "            if wn.isnumeric():\n",
        "                wn = \"0\"*len(wn)\n",
        "                result_vector += model_skipgram.get_vector(wn)\n",
        "                \n",
        "            else:\n",
        "                result_vector += model_skipgram.get_vector(\"unknown\")\n",
        "                \n",
        "    return result_vector\n",
        "\n",
        "def vectors_matrix_skipgram(texts):\n",
        "  x = len(texts)\n",
        "  y = 300                   #word2vec representation\n",
        "  matrix = np.zeros((x,y))\n",
        "\n",
        "  for i in range(x):\n",
        "    words_numbers = tokenizer(texts.iloc[i])       \n",
        "    matrix[i] = sum_vector_combining_skipgram(words_numbers)\n",
        "  return matrix\n",
        "\n",
        "vet_matrix_train_skipgram = vectors_matrix_skipgram(train.title)\n",
        "vet_matrix_test_skipgram = vectors_matrix_skipgram(test.title)\n",
        "\n",
        "#Prediction\n",
        "LR_skipgram = LogisticRegression(max_iter = 1000)\n",
        "LR_skipgram.fit(vet_matrix_train_skipgram, train.category)\n",
        "label_pred_skipgram = LR_skipgram.predict(vet_matrix_test_skipgram)\n",
        "CR_skipgram = classification_report(test.category, label_pred_skipgram)\n",
        "print(CR_skipgram)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "X_BlqTExSXwM",
        "outputId": "a8879906-55a1-4a20-f65f-815f9f99afd9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "     colunas       0.86      0.72      0.78      6103\n",
            "   cotidiano       0.63      0.81      0.70      1698\n",
            "     esporte       0.93      0.89      0.91      4663\n",
            "   ilustrada       0.15      0.91      0.26       131\n",
            "     mercado       0.84      0.81      0.83      5867\n",
            "       mundo       0.76      0.86      0.80      2051\n",
            "\n",
            "    accuracy                           0.81     20513\n",
            "   macro avg       0.69      0.83      0.71     20513\n",
            "weighted avg       0.84      0.81      0.82     20513\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(CR)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KnvjurNTzr9P",
        "outputId": "913abefe-8447-43a2-c436-645085203122"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "     colunas       0.86      0.71      0.78      6103\n",
            "   cotidiano       0.61      0.79      0.69      1698\n",
            "     esporte       0.92      0.88      0.90      4663\n",
            "   ilustrada       0.13      0.88      0.23       131\n",
            "     mercado       0.84      0.79      0.81      5867\n",
            "       mundo       0.74      0.86      0.79      2051\n",
            "\n",
            "    accuracy                           0.80     20513\n",
            "   macro avg       0.68      0.82      0.70     20513\n",
            "weighted avg       0.83      0.80      0.81     20513\n",
            "\n"
          ]
        }
      ]
    }
  ]
}