Word2vec_RF_and_LR.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Word2vec_RF_and_LR.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "H7B8Xm3CVqTj"
      },
      "source": [
        "import tarfile\n",
        "import codecs\n",
        "import numpy as np \n",
        "import pandas as pd\n",
        "from io import BytesIO\n",
        "import json\n",
        "import re\n",
        "import altair as alt\n",
        "import itertools\n",
        "from mlxtend.preprocessing import TransactionEncoder\n",
        "from mlxtend.frequent_patterns import apriori\n",
        "from mlxtend.frequent_patterns import association_rules\n",
        "import matplotlib.pyplot as plt\n",
        "import string\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "from sklearn.feature_extraction import text\n",
        "from collections import Counter\n",
        "from wordcloud import WordCloud\n",
        "import gensim\n",
        "import pickle\n",
        "\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer, TfidfTransformer\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from sklearn.pipeline import make_union\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.feature_selection import SelectFromModel"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xw_3JHHGVwLS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5a9d1cb3-a0e9-4279-ab1c-182f70782d32"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/data586_Project')"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/data586_Project\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YT28PlYIV1Fc"
      },
      "source": [
        "**Describe Dataset**\n",
        "\n",
        "Response variable: Label - Either the log is normal or anomaly"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yj68pzj7Vy2M",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "56b6e7ee-0582-4b0d-fdf4-1166a3d13b31"
      },
      "source": [
        "df = pd.read_csv('/content/data586_Project/MyDrive/DATA586_Project/processed_log.csv')\n",
        "df.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Date</th>\n",
              "      <th>Time</th>\n",
              "      <th>Pid</th>\n",
              "      <th>Level</th>\n",
              "      <th>Component</th>\n",
              "      <th>Content</th>\n",
              "      <th>BlockId</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>81109</td>\n",
              "      <td>203518</td>\n",
              "      <td>143</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.DataNode$DataXceiver</td>\n",
              "      <td>Receiving block blk_-1608999687919862906 src: ...</td>\n",
              "      <td>1608999687919862906</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>81109</td>\n",
              "      <td>203518</td>\n",
              "      <td>35</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.FSNamesystem</td>\n",
              "      <td>BLOCK* NameSystem.allocateBlock: /mnt/hadoop/m...</td>\n",
              "      <td>1608999687919862906</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>81109</td>\n",
              "      <td>203519</td>\n",
              "      <td>143</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.DataNode$DataXceiver</td>\n",
              "      <td>Receiving block blk_-1608999687919862906 src: ...</td>\n",
              "      <td>1608999687919862906</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>81109</td>\n",
              "      <td>203519</td>\n",
              "      <td>145</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.DataNode$DataXceiver</td>\n",
              "      <td>Receiving block blk_-1608999687919862906 src: ...</td>\n",
              "      <td>1608999687919862906</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>81109</td>\n",
              "      <td>203519</td>\n",
              "      <td>145</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.DataNode$PacketResponder</td>\n",
              "      <td>PacketResponder 1 for block blk_-1608999687919...</td>\n",
              "      <td>1608999687919862906</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    Date  ...              BlockId\n",
              "0  81109  ...  1608999687919862906\n",
              "1  81109  ...  1608999687919862906\n",
              "2  81109  ...  1608999687919862906\n",
              "3  81109  ...  1608999687919862906\n",
              "4  81109  ...  1608999687919862906\n",
              "\n",
              "[5 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FcfDSdKR73u8",
        "outputId": "7cdef0fb-99c1-40e0-98c0-13a220ef3c68"
      },
      "source": [
        "df.BlockId.nunique()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "575061"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aclExQ4g8CcE"
      },
      "source": [
        "event_description=[]\n",
        "for line in df.Content:\n",
        "  # extract only character add description of the event \n",
        "  match=re.findall('[a-zA-Z]+',str(line))\n",
        "  event_description.append(match[0:6])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ItHvNdsP9uwJ"
      },
      "source": [
        "df['Event_Description'] = event_description"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Eync0-S4_SJR"
      },
      "source": [
        "df.to_csv('event_desc.csv', index=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4M1rbktx_aAY"
      },
      "source": [
        "df = pd.read_csv('/content/data586_Project/MyDrive/DATA586_Project/event_desc.csv')"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-QqYwOnM-VFc"
      },
      "source": [
        "df['Event_id'] = df.groupby('Event_Description').ngroup()"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "vgyHRYiR_peV",
        "outputId": "2c157b21-08d7-4373-b802-60f8b0dfcc4e"
      },
      "source": [
        "df_event = df.iloc[:, 6:8].copy()\n",
        "df_event.head()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>BlockId</th>\n",
              "      <th>Event_Description</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1608999687919862906</td>\n",
              "      <td>['Receiving', 'block', 'blk', 'src', 'dest']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1608999687919862906</td>\n",
              "      <td>['BLOCK', 'NameSystem', 'allocateBlock', 'mnt'...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1608999687919862906</td>\n",
              "      <td>['Receiving', 'block', 'blk', 'src', 'dest']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1608999687919862906</td>\n",
              "      <td>['Receiving', 'block', 'blk', 'src', 'dest']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1608999687919862906</td>\n",
              "      <td>['PacketResponder', 'for', 'block', 'blk', 'te...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "               BlockId                                  Event_Description\n",
              "0  1608999687919862906       ['Receiving', 'block', 'blk', 'src', 'dest']\n",
              "1  1608999687919862906  ['BLOCK', 'NameSystem', 'allocateBlock', 'mnt'...\n",
              "2  1608999687919862906       ['Receiving', 'block', 'blk', 'src', 'dest']\n",
              "3  1608999687919862906       ['Receiving', 'block', 'blk', 'src', 'dest']\n",
              "4  1608999687919862906  ['PacketResponder', 'for', 'block', 'blk', 'te..."
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "R7OmuG7EsyOO",
        "outputId": "b6466ff5-5ad6-4b35-a994-be3a8bb73209"
      },
      "source": [
        "df_event = df.groupby('BlockId')['Event_Description'].apply(sum).reset_index(name='Events')\n",
        "df_event.head()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>BlockId</th>\n",
              "      <th>Events</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>24807389905763</td>\n",
              "      <td>['BLOCK', 'NameSystem', 'allocateBlock', 'user...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>27644297472372</td>\n",
              "      <td>['Receiving', 'block', 'blk', 'src', 'dest']['...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>33380551139732</td>\n",
              "      <td>['BLOCK', 'NameSystem', 'allocateBlock', 'user...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>35641716531335</td>\n",
              "      <td>['BLOCK', 'NameSystem', 'allocateBlock', 'user...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>40083153975957</td>\n",
              "      <td>['Receiving', 'block', 'blk', 'src', 'dest']['...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          BlockId                                             Events\n",
              "0  24807389905763  ['BLOCK', 'NameSystem', 'allocateBlock', 'user...\n",
              "1  27644297472372  ['Receiving', 'block', 'blk', 'src', 'dest']['...\n",
              "2  33380551139732  ['BLOCK', 'NameSystem', 'allocateBlock', 'user...\n",
              "3  35641716531335  ['BLOCK', 'NameSystem', 'allocateBlock', 'user...\n",
              "4  40083153975957  ['Receiving', 'block', 'blk', 'src', 'dest']['..."
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0oiRos0txKXq"
      },
      "source": [
        "df_event['Events'] = df_event['Events'].apply(lambda x: x.replace('][',','))\n",
        "df_event['Events'] = df_event['Events'].apply(lambda x: x.replace(\"'\",\"\"))\n",
        "df_event['Events'] = df_event['Events'].apply(lambda x: x.replace(\" \",\"\"))"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2CK-xxp6_gf-"
      },
      "source": [
        "df_event['Events'] = df_event['Events'].apply(lambda x: x[1:-1].split(','))\n",
        "# df_event['Events'] = df_event['Events'].apply(lambda x: x.replace(\"\\n\",\"\").replace('\\r', ''))"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6CAWs0pk6Bch"
      },
      "source": [
        "events = []\n",
        "for event in df_event[\"Events\"]:\n",
        "  events.append(event)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DhHvd9XACDXn"
      },
      "source": [
        "df_event.to_csv('BlockId_events.csv', index=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "9RNJkAkK0Zj9",
        "outputId": "7d0447f6-00e6-4148-c7eb-ae9e253fc5d0"
      },
      "source": [
        "df = pd.read_csv('/content/data586_Project/MyDrive/DATA586_Project/BlockId_events.csv')\n",
        "df.head()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>BlockId</th>\n",
              "      <th>Events</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>24807389905763</td>\n",
              "      <td>['BLOCK', 'NameSystem', 'allocateBlock', 'user...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>27644297472372</td>\n",
              "      <td>['Receiving', 'block', 'blk', 'src', 'dest', '...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>33380551139732</td>\n",
              "      <td>['BLOCK', 'NameSystem', 'allocateBlock', 'user...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>35641716531335</td>\n",
              "      <td>['BLOCK', 'NameSystem', 'allocateBlock', 'user...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>40083153975957</td>\n",
              "      <td>['Receiving', 'block', 'blk', 'src', 'dest', '...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          BlockId                                             Events\n",
              "0  24807389905763  ['BLOCK', 'NameSystem', 'allocateBlock', 'user...\n",
              "1  27644297472372  ['Receiving', 'block', 'blk', 'src', 'dest', '...\n",
              "2  33380551139732  ['BLOCK', 'NameSystem', 'allocateBlock', 'user...\n",
              "3  35641716531335  ['BLOCK', 'NameSystem', 'allocateBlock', 'user...\n",
              "4  40083153975957  ['Receiving', 'block', 'blk', 'src', 'dest', '..."
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Uh2c35u4WA4B"
      },
      "source": [
        "label = pd.read_csv('/content/data586_Project/MyDrive/DATA586_Project/anomaly_label.csv')"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2SjmdHguWErR"
      },
      "source": [
        "label['BlockId2'] = label.BlockId.str.extract('(\\d+)') \n",
        "label.drop('BlockId', axis=1, inplace=True)"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vZnIbd8PWGaA"
      },
      "source": [
        "label.BlockId2 = pd.to_numeric(label.BlockId2, errors='coerce').fillna(0).astype(np.int64)"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "w7el8Iz6WJAN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "39f28448-4945-489b-f87f-d82ecee6aa6b"
      },
      "source": [
        "log = pd.merge(df_event, label, left_on='BlockId', right_on='BlockId2', how='left').drop('BlockId2', axis=1)\n",
        "log.head()"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>BlockId</th>\n",
              "      <th>Events</th>\n",
              "      <th>Label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>24807389905763</td>\n",
              "      <td>[BLOCK, NameSystem, allocateBlock, user, root,...</td>\n",
              "      <td>Normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>27644297472372</td>\n",
              "      <td>[Receiving, block, blk, src, dest, Receiving, ...</td>\n",
              "      <td>Normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>33380551139732</td>\n",
              "      <td>[BLOCK, NameSystem, allocateBlock, user, root,...</td>\n",
              "      <td>Normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>35641716531335</td>\n",
              "      <td>[BLOCK, NameSystem, allocateBlock, user, root,...</td>\n",
              "      <td>Normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>40083153975957</td>\n",
              "      <td>[Receiving, block, blk, src, dest, BLOCK, Name...</td>\n",
              "      <td>Normal</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          BlockId                                             Events   Label\n",
              "0  24807389905763  [BLOCK, NameSystem, allocateBlock, user, root,...  Normal\n",
              "1  27644297472372  [Receiving, block, blk, src, dest, Receiving, ...  Normal\n",
              "2  33380551139732  [BLOCK, NameSystem, allocateBlock, user, root,...  Normal\n",
              "3  35641716531335  [BLOCK, NameSystem, allocateBlock, user, root,...  Normal\n",
              "4  40083153975957  [Receiving, block, blk, src, dest, BLOCK, Name...  Normal"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "y4EX5eoi00qJ",
        "outputId": "755ded49-5897-49b6-819b-072736e0b020"
      },
      "source": [
        "len(log)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "575061"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "woOw4Dg-FRog",
        "outputId": "38e8b20f-0be6-40d1-9875-1b529b9a7686"
      },
      "source": [
        "print(log.Events[1])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['Receiving', 'block', 'blk', 'src', 'dest', 'Receiving', 'block', 'blk', 'src', 'dest', 'Receiving', 'block', 'blk', 'src', 'dest', 'BLOCK', 'NameSystem', 'allocateBlock', 'user', 'root', 'sortrand', 'PacketResponder', 'for', 'block', 'blk', 'terminating', 'Received', 'block', 'blk', 'of', 'size', 'from', 'PacketResponder', 'for', 'block', 'blk', 'terminating', 'Received', 'block', 'blk', 'of', 'size', 'from', 'PacketResponder', 'for', 'block', 'blk', 'terminating', 'Received', 'block', 'blk', 'of', 'size', 'from', 'BLOCK', 'NameSystem', 'addStoredBlock', 'blockMap', 'updated', 'is', 'BLOCK', 'NameSystem', 'addStoredBlock', 'blockMap', 'updated', 'is', 'BLOCK', 'NameSystem', 'addStoredBlock', 'blockMap', 'updated', 'is', 'BLOCK', 'NameSystem', 'delete', 'blk', 'is', 'added', 'BLOCK', 'NameSystem', 'delete', 'blk', 'is', 'added', 'BLOCK', 'NameSystem', 'delete', 'blk', 'is', 'added', 'Deleting', 'block', 'blk', 'file', 'mnt', 'hadoop', 'Deleting', 'block', 'blk', 'file', 'mnt', 'hadoop', 'Deleting', 'block', 'blk', 'file', 'mnt', 'hadoop']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AUdTn6AfFKg8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 142
        },
        "outputId": "78b1a715-da53-437f-b473-dc002bd01504"
      },
      "source": [
        "log.groupby('Label').count()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>BlockId</th>\n",
              "      <th>Events</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Label</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Anomaly</th>\n",
              "      <td>16838</td>\n",
              "      <td>16838</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Normal</th>\n",
              "      <td>558223</td>\n",
              "      <td>558223</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         BlockId  Events\n",
              "Label                   \n",
              "Anomaly    16838   16838\n",
              "Normal    558223  558223"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VCnl9o5m1EEc"
      },
      "source": [
        "# **Word2vec Training - Vector Averaging**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "R3h7YhGi1Dgv",
        "outputId": "ddeeaf0d-21b0-4fad-b8fe-6ddb918066b6"
      },
      "source": [
        "import logging\n",
        "logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s',\\\n",
        "    level=logging.INFO)\n",
        "\n",
        "num_features = 300    # Word vector dimensionality                      \n",
        "min_word_count = 40   # Minimum word count                        \n",
        "num_workers = 4       # Number of threads to run in parallel\n",
        "context = 10          # Context window size                                                                                    \n",
        "downsampling = 1e-3   # Downsample setting for frequent words\n",
        "\n",
        "# Initialize and train the model\n",
        "from gensim.models import word2vec\n",
        "print (\"Training model...\")\n",
        "model = word2vec.Word2Vec(events, workers=num_workers, \\\n",
        "            size=num_features, min_count = min_word_count, \\\n",
        "            window = context, sample = downsampling)\n",
        "\n",
        "model.init_sims(replace=True)\n",
        "\n",
        "word2vec_1 = \"word2vec_origin\"\n",
        "model.save(word2vec_1)"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2021-05-06 05:43:51,309 : INFO : collecting all words and their counts\n",
            "2021-05-06 05:43:51,311 : INFO : PROGRESS: at sentence #0, processed 0 words, keeping 0 word types\n",
            "2021-05-06 05:43:51,484 : INFO : PROGRESS: at sentence #10000, processed 1088731 words, keeping 69 word types\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Training model...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "2021-05-06 05:43:51,663 : INFO : PROGRESS: at sentence #20000, processed 2176989 words, keeping 73 word types\n",
            "2021-05-06 05:43:51,847 : INFO : PROGRESS: at sentence #30000, processed 3262924 words, keeping 74 word types\n",
            "2021-05-06 05:43:52,008 : INFO : PROGRESS: at sentence #40000, processed 4346954 words, keeping 75 word types\n",
            "2021-05-06 05:43:52,173 : INFO : PROGRESS: at sentence #50000, processed 5428037 words, keeping 75 word types\n",
            "2021-05-06 05:43:52,347 : INFO : PROGRESS: at sentence #60000, processed 6515740 words, keeping 76 word types\n",
            "2021-05-06 05:43:52,512 : INFO : PROGRESS: at sentence #70000, processed 7601043 words, keeping 78 word types\n",
            "2021-05-06 05:43:52,682 : INFO : PROGRESS: at sentence #80000, processed 8690870 words, keeping 78 word types\n",
            "2021-05-06 05:43:52,851 : INFO : PROGRESS: at sentence #90000, processed 9773669 words, keeping 78 word types\n",
            "2021-05-06 05:43:53,013 : INFO : PROGRESS: at sentence #100000, processed 10862826 words, keeping 78 word types\n",
            "2021-05-06 05:43:53,198 : INFO : PROGRESS: at sentence #110000, processed 11947478 words, keeping 78 word types\n",
            "2021-05-06 05:43:53,366 : INFO : PROGRESS: at sentence #120000, processed 13028842 words, keeping 78 word types\n",
            "2021-05-06 05:43:53,536 : INFO : PROGRESS: at sentence #130000, processed 14110920 words, keeping 82 word types\n",
            "2021-05-06 05:43:53,711 : INFO : PROGRESS: at sentence #140000, processed 15197565 words, keeping 82 word types\n",
            "2021-05-06 05:43:53,874 : INFO : PROGRESS: at sentence #150000, processed 16281461 words, keeping 82 word types\n",
            "2021-05-06 05:43:54,040 : INFO : PROGRESS: at sentence #160000, processed 17374081 words, keeping 82 word types\n",
            "2021-05-06 05:43:54,196 : INFO : PROGRESS: at sentence #170000, processed 18458105 words, keeping 82 word types\n",
            "2021-05-06 05:43:54,355 : INFO : PROGRESS: at sentence #180000, processed 19553100 words, keeping 82 word types\n",
            "2021-05-06 05:43:54,519 : INFO : PROGRESS: at sentence #190000, processed 20641835 words, keeping 82 word types\n",
            "2021-05-06 05:43:54,680 : INFO : PROGRESS: at sentence #200000, processed 21729170 words, keeping 83 word types\n",
            "2021-05-06 05:43:54,853 : INFO : PROGRESS: at sentence #210000, processed 22811502 words, keeping 83 word types\n",
            "2021-05-06 05:43:55,010 : INFO : PROGRESS: at sentence #220000, processed 23899003 words, keeping 85 word types\n",
            "2021-05-06 05:43:55,167 : INFO : PROGRESS: at sentence #230000, processed 24987765 words, keeping 85 word types\n",
            "2021-05-06 05:43:55,327 : INFO : PROGRESS: at sentence #240000, processed 26070494 words, keeping 85 word types\n",
            "2021-05-06 05:43:55,484 : INFO : PROGRESS: at sentence #250000, processed 27157655 words, keeping 85 word types\n",
            "2021-05-06 05:43:55,648 : INFO : PROGRESS: at sentence #260000, processed 28240531 words, keeping 85 word types\n",
            "2021-05-06 05:43:55,820 : INFO : PROGRESS: at sentence #270000, processed 29330379 words, keeping 85 word types\n",
            "2021-05-06 05:43:55,986 : INFO : PROGRESS: at sentence #280000, processed 30414561 words, keeping 85 word types\n",
            "2021-05-06 05:43:56,141 : INFO : PROGRESS: at sentence #290000, processed 31496783 words, keeping 85 word types\n",
            "2021-05-06 05:43:56,303 : INFO : PROGRESS: at sentence #300000, processed 32581095 words, keeping 85 word types\n",
            "2021-05-06 05:43:56,467 : INFO : PROGRESS: at sentence #310000, processed 33667909 words, keeping 85 word types\n",
            "2021-05-06 05:43:56,641 : INFO : PROGRESS: at sentence #320000, processed 34753865 words, keeping 85 word types\n",
            "2021-05-06 05:43:56,802 : INFO : PROGRESS: at sentence #330000, processed 35841083 words, keeping 85 word types\n",
            "2021-05-06 05:43:56,961 : INFO : PROGRESS: at sentence #340000, processed 36932665 words, keeping 85 word types\n",
            "2021-05-06 05:43:57,118 : INFO : PROGRESS: at sentence #350000, processed 38015716 words, keeping 85 word types\n",
            "2021-05-06 05:43:57,280 : INFO : PROGRESS: at sentence #360000, processed 39103336 words, keeping 85 word types\n",
            "2021-05-06 05:43:57,440 : INFO : PROGRESS: at sentence #370000, processed 40191984 words, keeping 85 word types\n",
            "2021-05-06 05:43:57,596 : INFO : PROGRESS: at sentence #380000, processed 41275910 words, keeping 85 word types\n",
            "2021-05-06 05:43:57,749 : INFO : PROGRESS: at sentence #390000, processed 42361682 words, keeping 85 word types\n",
            "2021-05-06 05:43:57,909 : INFO : PROGRESS: at sentence #400000, processed 43446449 words, keeping 87 word types\n",
            "2021-05-06 05:43:58,066 : INFO : PROGRESS: at sentence #410000, processed 44533083 words, keeping 87 word types\n",
            "2021-05-06 05:43:58,236 : INFO : PROGRESS: at sentence #420000, processed 45620059 words, keeping 87 word types\n",
            "2021-05-06 05:43:58,391 : INFO : PROGRESS: at sentence #430000, processed 46702807 words, keeping 87 word types\n",
            "2021-05-06 05:43:58,544 : INFO : PROGRESS: at sentence #440000, processed 47796744 words, keeping 88 word types\n",
            "2021-05-06 05:43:58,705 : INFO : PROGRESS: at sentence #450000, processed 48886545 words, keeping 88 word types\n",
            "2021-05-06 05:43:58,859 : INFO : PROGRESS: at sentence #460000, processed 49978221 words, keeping 88 word types\n",
            "2021-05-06 05:43:59,025 : INFO : PROGRESS: at sentence #470000, processed 51060803 words, keeping 88 word types\n",
            "2021-05-06 05:43:59,181 : INFO : PROGRESS: at sentence #480000, processed 52150655 words, keeping 88 word types\n",
            "2021-05-06 05:43:59,337 : INFO : PROGRESS: at sentence #490000, processed 53239237 words, keeping 88 word types\n",
            "2021-05-06 05:43:59,496 : INFO : PROGRESS: at sentence #500000, processed 54330058 words, keeping 88 word types\n",
            "2021-05-06 05:43:59,667 : INFO : PROGRESS: at sentence #510000, processed 55418099 words, keeping 88 word types\n",
            "2021-05-06 05:43:59,821 : INFO : PROGRESS: at sentence #520000, processed 56507676 words, keeping 88 word types\n",
            "2021-05-06 05:43:59,982 : INFO : PROGRESS: at sentence #530000, processed 57601177 words, keeping 88 word types\n",
            "2021-05-06 05:44:00,137 : INFO : PROGRESS: at sentence #540000, processed 58693443 words, keeping 88 word types\n",
            "2021-05-06 05:44:00,293 : INFO : PROGRESS: at sentence #550000, processed 59780844 words, keeping 88 word types\n",
            "2021-05-06 05:44:00,449 : INFO : PROGRESS: at sentence #560000, processed 60869192 words, keeping 88 word types\n",
            "2021-05-06 05:44:00,610 : INFO : PROGRESS: at sentence #570000, processed 61959371 words, keeping 88 word types\n",
            "2021-05-06 05:44:00,691 : INFO : collected 88 word types from a corpus of 62512354 raw words and 575061 sentences\n",
            "2021-05-06 05:44:00,692 : INFO : Loading a fresh vocabulary\n",
            "2021-05-06 05:44:00,696 : INFO : effective_min_count=40 retains 67 unique words (76% of original 88, drops 21)\n",
            "2021-05-06 05:44:00,697 : INFO : effective_min_count=40 leaves 62512073 word corpus (99% of original 62512354, drops 281)\n",
            "2021-05-06 05:44:00,702 : INFO : deleting the raw counts dictionary of 88 items\n",
            "2021-05-06 05:44:00,703 : INFO : sample=0.001 downsamples 35 most-common words\n",
            "2021-05-06 05:44:00,705 : INFO : downsampling leaves estimated 13238329 word corpus (21.2% of prior 62512073)\n",
            "2021-05-06 05:44:00,707 : INFO : estimated required memory for 67 words and 300 dimensions: 194300 bytes\n",
            "2021-05-06 05:44:00,709 : INFO : resetting layer weights\n",
            "2021-05-06 05:44:00,732 : INFO : training model with 4 workers on 67 vocabulary and 300 features, using sg=0 hs=0 sample=0.001 negative=5 window=10\n",
            "2021-05-06 05:44:01,740 : INFO : EPOCH 1 - PROGRESS: at 3.34% examples, 440724 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:02,746 : INFO : EPOCH 1 - PROGRESS: at 6.74% examples, 444388 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:03,749 : INFO : EPOCH 1 - PROGRESS: at 10.25% examples, 449857 words/s, in_qsize 7, out_qsize 2\n",
            "2021-05-06 05:44:04,755 : INFO : EPOCH 1 - PROGRESS: at 13.75% examples, 452192 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:44:05,756 : INFO : EPOCH 1 - PROGRESS: at 17.11% examples, 450351 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:06,764 : INFO : EPOCH 1 - PROGRESS: at 20.55% examples, 450328 words/s, in_qsize 8, out_qsize 1\n",
            "2021-05-06 05:44:07,771 : INFO : EPOCH 1 - PROGRESS: at 23.99% examples, 450112 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:08,772 : INFO : EPOCH 1 - PROGRESS: at 27.22% examples, 447107 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:09,775 : INFO : EPOCH 1 - PROGRESS: at 30.44% examples, 444834 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:44:10,777 : INFO : EPOCH 1 - PROGRESS: at 33.63% examples, 442804 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:11,784 : INFO : EPOCH 1 - PROGRESS: at 36.90% examples, 441656 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:12,786 : INFO : EPOCH 1 - PROGRESS: at 40.00% examples, 438905 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:13,787 : INFO : EPOCH 1 - PROGRESS: at 43.32% examples, 438831 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:14,803 : INFO : EPOCH 1 - PROGRESS: at 46.35% examples, 435507 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:44:15,804 : INFO : EPOCH 1 - PROGRESS: at 49.54% examples, 434453 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:16,808 : INFO : EPOCH 1 - PROGRESS: at 52.63% examples, 432620 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:17,811 : INFO : EPOCH 1 - PROGRESS: at 55.80% examples, 431755 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:44:18,821 : INFO : EPOCH 1 - PROGRESS: at 59.05% examples, 431514 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:19,829 : INFO : EPOCH 1 - PROGRESS: at 62.16% examples, 430219 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:20,833 : INFO : EPOCH 1 - PROGRESS: at 65.36% examples, 429827 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:21,841 : INFO : EPOCH 1 - PROGRESS: at 68.56% examples, 429359 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:22,851 : INFO : EPOCH 1 - PROGRESS: at 71.70% examples, 428468 words/s, in_qsize 8, out_qsize 1\n",
            "2021-05-06 05:44:23,856 : INFO : EPOCH 1 - PROGRESS: at 74.72% examples, 427133 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:24,865 : INFO : EPOCH 1 - PROGRESS: at 77.76% examples, 426014 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:25,866 : INFO : EPOCH 1 - PROGRESS: at 80.88% examples, 425510 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:26,871 : INFO : EPOCH 1 - PROGRESS: at 84.16% examples, 425845 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:27,873 : INFO : EPOCH 1 - PROGRESS: at 87.40% examples, 425978 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:28,874 : INFO : EPOCH 1 - PROGRESS: at 90.61% examples, 425998 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:29,878 : INFO : EPOCH 1 - PROGRESS: at 93.85% examples, 426213 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:30,880 : INFO : EPOCH 1 - PROGRESS: at 97.08% examples, 426225 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:31,782 : INFO : worker thread finished; awaiting finish of 3 more threads\n",
            "2021-05-06 05:44:31,786 : INFO : worker thread finished; awaiting finish of 2 more threads\n",
            "2021-05-06 05:44:31,799 : INFO : worker thread finished; awaiting finish of 1 more threads\n",
            "2021-05-06 05:44:31,801 : INFO : worker thread finished; awaiting finish of 0 more threads\n",
            "2021-05-06 05:44:31,802 : INFO : EPOCH - 1 : training on 62512354 raw words (13236192 effective words) took 31.1s, 426074 effective words/s\n",
            "2021-05-06 05:44:32,811 : INFO : EPOCH 2 - PROGRESS: at 3.07% examples, 406705 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:44:33,813 : INFO : EPOCH 2 - PROGRESS: at 6.14% examples, 406220 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:34,816 : INFO : EPOCH 2 - PROGRESS: at 9.32% examples, 409585 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:35,822 : INFO : EPOCH 2 - PROGRESS: at 12.51% examples, 411902 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:36,831 : INFO : EPOCH 2 - PROGRESS: at 15.75% examples, 414175 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:37,835 : INFO : EPOCH 2 - PROGRESS: at 18.86% examples, 413564 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:38,839 : INFO : EPOCH 2 - PROGRESS: at 21.98% examples, 412670 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:39,847 : INFO : EPOCH 2 - PROGRESS: at 24.96% examples, 409869 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:40,851 : INFO : EPOCH 2 - PROGRESS: at 28.11% examples, 410617 words/s, in_qsize 8, out_qsize 0\n",
            "2021-05-06 05:44:41,857 : INFO : EPOCH 2 - PROGRESS: at 31.26% examples, 410967 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:42,860 : INFO : EPOCH 2 - PROGRESS: at 34.40% examples, 411469 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:43,861 : INFO : EPOCH 2 - PROGRESS: at 37.57% examples, 412111 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:44,865 : INFO : EPOCH 2 - PROGRESS: at 40.72% examples, 412367 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:45,871 : INFO : EPOCH 2 - PROGRESS: at 43.93% examples, 412970 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:44:46,873 : INFO : EPOCH 2 - PROGRESS: at 47.03% examples, 412730 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:47,873 : INFO : EPOCH 2 - PROGRESS: at 50.24% examples, 413359 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:48,878 : INFO : EPOCH 2 - PROGRESS: at 53.36% examples, 413191 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:49,881 : INFO : EPOCH 2 - PROGRESS: at 56.55% examples, 413566 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:50,882 : INFO : EPOCH 2 - PROGRESS: at 59.82% examples, 414592 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:51,887 : INFO : EPOCH 2 - PROGRESS: at 62.95% examples, 414563 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:52,889 : INFO : EPOCH 2 - PROGRESS: at 66.18% examples, 415148 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:53,898 : INFO : EPOCH 2 - PROGRESS: at 69.42% examples, 415538 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:54,907 : INFO : EPOCH 2 - PROGRESS: at 72.65% examples, 415915 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:55,911 : INFO : EPOCH 2 - PROGRESS: at 75.84% examples, 416183 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:56,919 : INFO : EPOCH 2 - PROGRESS: at 79.05% examples, 416410 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:57,925 : INFO : EPOCH 2 - PROGRESS: at 82.26% examples, 416724 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:58,927 : INFO : EPOCH 2 - PROGRESS: at 85.48% examples, 417120 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:44:59,929 : INFO : EPOCH 2 - PROGRESS: at 88.58% examples, 416874 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:00,937 : INFO : EPOCH 2 - PROGRESS: at 91.78% examples, 417124 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:45:01,938 : INFO : EPOCH 2 - PROGRESS: at 94.95% examples, 417293 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:02,954 : INFO : EPOCH 2 - PROGRESS: at 98.16% examples, 417364 words/s, in_qsize 8, out_qsize 1\n",
            "2021-05-06 05:45:03,496 : INFO : worker thread finished; awaiting finish of 3 more threads\n",
            "2021-05-06 05:45:03,499 : INFO : worker thread finished; awaiting finish of 2 more threads\n",
            "2021-05-06 05:45:03,502 : INFO : worker thread finished; awaiting finish of 1 more threads\n",
            "2021-05-06 05:45:03,505 : INFO : worker thread finished; awaiting finish of 0 more threads\n",
            "2021-05-06 05:45:03,507 : INFO : EPOCH - 2 : training on 62512354 raw words (13243770 effective words) took 31.7s, 417832 effective words/s\n",
            "2021-05-06 05:45:04,525 : INFO : EPOCH 3 - PROGRESS: at 3.24% examples, 425889 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:45:05,526 : INFO : EPOCH 3 - PROGRESS: at 6.60% examples, 434395 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:06,529 : INFO : EPOCH 3 - PROGRESS: at 9.93% examples, 435444 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:07,547 : INFO : EPOCH 3 - PROGRESS: at 13.24% examples, 434005 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:08,551 : INFO : EPOCH 3 - PROGRESS: at 16.59% examples, 435328 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:09,554 : INFO : EPOCH 3 - PROGRESS: at 19.92% examples, 435899 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:10,557 : INFO : EPOCH 3 - PROGRESS: at 23.17% examples, 434642 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:11,559 : INFO : EPOCH 3 - PROGRESS: at 26.51% examples, 435371 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:12,562 : INFO : EPOCH 3 - PROGRESS: at 29.84% examples, 435667 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:13,567 : INFO : EPOCH 3 - PROGRESS: at 33.15% examples, 436035 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:14,568 : INFO : EPOCH 3 - PROGRESS: at 36.32% examples, 434398 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:15,579 : INFO : EPOCH 3 - PROGRESS: at 39.75% examples, 435721 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:16,579 : INFO : EPOCH 3 - PROGRESS: at 43.15% examples, 436701 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:17,590 : INFO : EPOCH 3 - PROGRESS: at 46.55% examples, 437307 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:18,591 : INFO : EPOCH 3 - PROGRESS: at 49.87% examples, 437382 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:19,594 : INFO : EPOCH 3 - PROGRESS: at 53.22% examples, 437553 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:20,596 : INFO : EPOCH 3 - PROGRESS: at 56.55% examples, 437575 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:21,596 : INFO : EPOCH 3 - PROGRESS: at 59.84% examples, 437412 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:22,604 : INFO : EPOCH 3 - PROGRESS: at 63.17% examples, 437496 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:23,608 : INFO : EPOCH 3 - PROGRESS: at 66.45% examples, 437213 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:24,616 : INFO : EPOCH 3 - PROGRESS: at 69.79% examples, 437233 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:25,622 : INFO : EPOCH 3 - PROGRESS: at 73.03% examples, 436840 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:26,626 : INFO : EPOCH 3 - PROGRESS: at 76.34% examples, 436795 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:45:27,639 : INFO : EPOCH 3 - PROGRESS: at 79.65% examples, 436735 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:28,645 : INFO : EPOCH 3 - PROGRESS: at 82.95% examples, 436615 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:29,648 : INFO : EPOCH 3 - PROGRESS: at 86.23% examples, 436604 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:30,659 : INFO : EPOCH 3 - PROGRESS: at 89.50% examples, 436339 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:31,662 : INFO : EPOCH 3 - PROGRESS: at 92.78% examples, 436365 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:32,673 : INFO : EPOCH 3 - PROGRESS: at 96.10% examples, 436355 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:33,681 : INFO : EPOCH 3 - PROGRESS: at 99.36% examples, 436198 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:33,858 : INFO : worker thread finished; awaiting finish of 3 more threads\n",
            "2021-05-06 05:45:33,861 : INFO : worker thread finished; awaiting finish of 2 more threads\n",
            "2021-05-06 05:45:33,864 : INFO : worker thread finished; awaiting finish of 1 more threads\n",
            "2021-05-06 05:45:33,870 : INFO : worker thread finished; awaiting finish of 0 more threads\n",
            "2021-05-06 05:45:33,871 : INFO : EPOCH - 3 : training on 62512354 raw words (13242916 effective words) took 30.4s, 436271 effective words/s\n",
            "2021-05-06 05:45:34,879 : INFO : EPOCH 4 - PROGRESS: at 3.26% examples, 431766 words/s, in_qsize 8, out_qsize 1\n",
            "2021-05-06 05:45:35,885 : INFO : EPOCH 4 - PROGRESS: at 6.66% examples, 439524 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:36,890 : INFO : EPOCH 4 - PROGRESS: at 10.01% examples, 439284 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:37,891 : INFO : EPOCH 4 - PROGRESS: at 13.32% examples, 438483 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:45:38,897 : INFO : EPOCH 4 - PROGRESS: at 16.65% examples, 438253 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:39,900 : INFO : EPOCH 4 - PROGRESS: at 19.98% examples, 438058 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:40,912 : INFO : EPOCH 4 - PROGRESS: at 23.33% examples, 437830 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:41,914 : INFO : EPOCH 4 - PROGRESS: at 26.70% examples, 438662 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:45:42,914 : INFO : EPOCH 4 - PROGRESS: at 30.06% examples, 439204 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:43,922 : INFO : EPOCH 4 - PROGRESS: at 33.36% examples, 438846 words/s, in_qsize 8, out_qsize 0\n",
            "2021-05-06 05:45:44,931 : INFO : EPOCH 4 - PROGRESS: at 36.67% examples, 438331 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:45:45,947 : INFO : EPOCH 4 - PROGRESS: at 39.97% examples, 437588 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:46,948 : INFO : EPOCH 4 - PROGRESS: at 43.24% examples, 437068 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:47,952 : INFO : EPOCH 4 - PROGRESS: at 46.54% examples, 436856 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:48,953 : INFO : EPOCH 4 - PROGRESS: at 49.87% examples, 436993 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:49,961 : INFO : EPOCH 4 - PROGRESS: at 53.22% examples, 437049 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:45:50,967 : INFO : EPOCH 4 - PROGRESS: at 56.45% examples, 436363 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:51,982 : INFO : EPOCH 4 - PROGRESS: at 59.90% examples, 437088 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:52,985 : INFO : EPOCH 4 - PROGRESS: at 63.14% examples, 436617 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:53,993 : INFO : EPOCH 4 - PROGRESS: at 66.50% examples, 436846 words/s, in_qsize 8, out_qsize 0\n",
            "2021-05-06 05:45:55,003 : INFO : EPOCH 4 - PROGRESS: at 69.77% examples, 436358 words/s, in_qsize 8, out_qsize 0\n",
            "2021-05-06 05:45:56,014 : INFO : EPOCH 4 - PROGRESS: at 73.14% examples, 436610 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:57,020 : INFO : EPOCH 4 - PROGRESS: at 76.43% examples, 436455 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:58,021 : INFO : EPOCH 4 - PROGRESS: at 79.68% examples, 436241 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:45:59,021 : INFO : EPOCH 4 - PROGRESS: at 83.07% examples, 436705 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:00,021 : INFO : EPOCH 4 - PROGRESS: at 86.31% examples, 436481 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:01,023 : INFO : EPOCH 4 - PROGRESS: at 89.55% examples, 436256 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:02,024 : INFO : EPOCH 4 - PROGRESS: at 92.76% examples, 435995 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:03,031 : INFO : EPOCH 4 - PROGRESS: at 96.13% examples, 436277 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:04,036 : INFO : EPOCH 4 - PROGRESS: at 99.44% examples, 436379 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:04,201 : INFO : worker thread finished; awaiting finish of 3 more threads\n",
            "2021-05-06 05:46:04,205 : INFO : worker thread finished; awaiting finish of 2 more threads\n",
            "2021-05-06 05:46:04,213 : INFO : worker thread finished; awaiting finish of 1 more threads\n",
            "2021-05-06 05:46:04,216 : INFO : worker thread finished; awaiting finish of 0 more threads\n",
            "2021-05-06 05:46:04,218 : INFO : EPOCH - 4 : training on 62512354 raw words (13235539 effective words) took 30.3s, 436234 effective words/s\n",
            "2021-05-06 05:46:05,227 : INFO : EPOCH 5 - PROGRESS: at 3.32% examples, 439137 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:06,230 : INFO : EPOCH 5 - PROGRESS: at 6.63% examples, 437686 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:07,236 : INFO : EPOCH 5 - PROGRESS: at 10.03% examples, 439640 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:08,237 : INFO : EPOCH 5 - PROGRESS: at 13.45% examples, 442702 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:09,239 : INFO : EPOCH 5 - PROGRESS: at 16.83% examples, 443130 words/s, in_qsize 6, out_qsize 0\n",
            "2021-05-06 05:46:10,244 : INFO : EPOCH 5 - PROGRESS: at 20.12% examples, 441602 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:11,245 : INFO : EPOCH 5 - PROGRESS: at 23.46% examples, 441184 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:12,261 : INFO : EPOCH 5 - PROGRESS: at 26.88% examples, 441761 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:13,263 : INFO : EPOCH 5 - PROGRESS: at 30.22% examples, 441607 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:14,269 : INFO : EPOCH 5 - PROGRESS: at 33.56% examples, 441568 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:15,287 : INFO : EPOCH 5 - PROGRESS: at 36.83% examples, 440044 words/s, in_qsize 7, out_qsize 2\n",
            "2021-05-06 05:46:16,290 : INFO : EPOCH 5 - PROGRESS: at 40.28% examples, 441186 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:17,292 : INFO : EPOCH 5 - PROGRESS: at 43.73% examples, 442340 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:18,296 : INFO : EPOCH 5 - PROGRESS: at 47.03% examples, 441740 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:19,308 : INFO : EPOCH 5 - PROGRESS: at 50.43% examples, 441707 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:20,312 : INFO : EPOCH 5 - PROGRESS: at 53.78% examples, 441575 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:21,316 : INFO : EPOCH 5 - PROGRESS: at 57.01% examples, 440626 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:22,328 : INFO : EPOCH 5 - PROGRESS: at 60.28% examples, 439912 words/s, in_qsize 5, out_qsize 2\n",
            "2021-05-06 05:46:23,334 : INFO : EPOCH 5 - PROGRESS: at 63.59% examples, 439645 words/s, in_qsize 6, out_qsize 1\n",
            "2021-05-06 05:46:24,343 : INFO : EPOCH 5 - PROGRESS: at 66.82% examples, 438851 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:25,343 : INFO : EPOCH 5 - PROGRESS: at 70.17% examples, 438973 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:26,344 : INFO : EPOCH 5 - PROGRESS: at 73.51% examples, 439134 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:27,348 : INFO : EPOCH 5 - PROGRESS: at 76.81% examples, 438990 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:28,354 : INFO : EPOCH 5 - PROGRESS: at 80.16% examples, 439115 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:29,360 : INFO : EPOCH 5 - PROGRESS: at 83.55% examples, 439448 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:30,367 : INFO : EPOCH 5 - PROGRESS: at 86.94% examples, 439796 words/s, in_qsize 8, out_qsize 0\n",
            "2021-05-06 05:46:31,367 : INFO : EPOCH 5 - PROGRESS: at 90.23% examples, 439709 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:32,368 : INFO : EPOCH 5 - PROGRESS: at 93.46% examples, 439434 words/s, in_qsize 8, out_qsize 3\n",
            "2021-05-06 05:46:33,371 : INFO : EPOCH 5 - PROGRESS: at 96.81% examples, 439576 words/s, in_qsize 7, out_qsize 0\n",
            "2021-05-06 05:46:34,339 : INFO : worker thread finished; awaiting finish of 3 more threads\n",
            "2021-05-06 05:46:34,346 : INFO : worker thread finished; awaiting finish of 2 more threads\n",
            "2021-05-06 05:46:34,348 : INFO : worker thread finished; awaiting finish of 1 more threads\n",
            "2021-05-06 05:46:34,353 : INFO : worker thread finished; awaiting finish of 0 more threads\n",
            "2021-05-06 05:46:34,354 : INFO : EPOCH - 5 : training on 62512354 raw words (13237415 effective words) took 30.1s, 439370 effective words/s\n",
            "2021-05-06 05:46:34,355 : INFO : training on a 312561770 raw words (66195832 effective words) took 153.6s, 430899 effective words/s\n",
            "2021-05-06 05:46:34,360 : INFO : precomputing L2-norms of word weight vectors\n",
            "2021-05-06 05:46:34,365 : INFO : saving Word2Vec object under word2vec_origin, separately None\n",
            "2021-05-06 05:46:34,367 : INFO : not storing attribute vectors_norm\n",
            "2021-05-06 05:46:34,369 : INFO : not storing attribute cum_table\n",
            "2021-05-06 05:46:34,381 : INFO : saved word2vec_origin\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "566a4oXFGJw3",
        "outputId": "f5e0bef8-fba2-4766-844f-d29d87ca3327"
      },
      "source": [
        "from gensim.models import Word2Vec\n",
        "model = Word2Vec.load(\"word2vec_origin\")\n",
        "model.wv.syn0.shape"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2021-05-06 05:47:24,441 : INFO : loading Word2Vec object from word2vec_origin\n",
            "2021-05-06 05:47:24,447 : INFO : loading wv recursively from word2vec_origin.wv.* with mmap=None\n",
            "2021-05-06 05:47:24,449 : INFO : setting ignored attribute vectors_norm to None\n",
            "2021-05-06 05:47:24,450 : INFO : loading vocabulary recursively from word2vec_origin.vocabulary.* with mmap=None\n",
            "2021-05-06 05:47:24,451 : INFO : loading trainables recursively from word2vec_origin.trainables.* with mmap=None\n",
            "2021-05-06 05:47:24,453 : INFO : setting ignored attribute cum_table to None\n",
            "2021-05-06 05:47:24,455 : INFO : loaded word2vec_origin\n",
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:3: DeprecationWarning: Call to deprecated `syn0` (Attribute will be removed in 4.0.0, use self.wv.vectors instead).\n",
            "  This is separate from the ipykernel package so we can avoid doing imports until\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(67, 300)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "grB82qaKGbxh"
      },
      "source": [
        "def makeFeatureVec(words, model, num_features):\n",
        "    # Function to average all of the word vectors in a given block id\n",
        "\n",
        "    featureVec = np.zeros((num_features,), dtype=\"float32\")\n",
        "    nwords = 0\n",
        "\n",
        "    index2word_set = set(model.wv.index2word)\n",
        "\n",
        "    for word in words:\n",
        "        if word in index2word_set: \n",
        "            nwords = nwords + 1.\n",
        "            featureVec = np.add(featureVec,model[word])\n",
        "\n",
        "    # Divide the result by the number of words to get the average\n",
        "    featureVec = np.divide(featureVec,nwords)\n",
        "    return featureVec\n",
        "\n",
        "\n",
        "def getAvgFeatureVecs(ids, model, num_features):\n",
        "    # Given a set of ids calculate the average feature vector for each one\n",
        "\n",
        "    counter = 0\n",
        "    idFeatureVecs = np.zeros((len(ids), num_features), dtype=\"float32\")\n",
        "\n",
        "    for element in ids:\n",
        "\n",
        "       if counter%10000 == 0:\n",
        "           print(\"Block id %d of %d\" % (counter, len(ids)))\n",
        "\n",
        "       # Call the function that makes average feature vectors\n",
        "       idFeatureVecs[counter] = makeFeatureVec(element, model, num_features)\n",
        "\n",
        "       counter = counter + 1\n",
        "    return idFeatureVecs"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CBPcM7D-JFgS"
      },
      "source": [
        "# **Parse HDFS.2K Sample Set**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TU1xD5AJhykM"
      },
      "source": [
        "with open('/content/data586_Project/MyDrive/DATA586_Project/HDFS_2k.log') as f:\n",
        "    data = f.readlines()"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T0f3EGI3i_Dg"
      },
      "source": [
        "test_log = [t.strip() for t in data]\n",
        "Date = []\n",
        "Time = []\n",
        "Pid = []\n",
        "Level = []\n",
        "Component = []\n",
        "Content = []"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SSFMV5THjIEZ"
      },
      "source": [
        "for line in test_log:\n",
        "  s1=line.split(' ')[0].strip()\n",
        "  Date.append(s1)\n",
        "  s2=line.split(' ')[1].strip()\n",
        "  Time.append(s2)\n",
        "  s3=line.split(' ')[2].strip()\n",
        "  Pid.append(s3)\n",
        "  s4=line.split(' ')[3].strip()\n",
        "  Level.append(s4)\n",
        "  s5=line.split(' ')[4].strip(': ')\n",
        "  Component.append(s5)\n",
        "  s6=line.split(' ')[5:]\n",
        "  Content.append(' '.join(s6))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OKvVzlOajMjo"
      },
      "source": [
        "test = pd.DataFrame([Date, Time, Pid, Level, Component, Content])\n",
        "test = test.T\n",
        "test.columns=['Date','Time', 'Pid', 'Level', 'Component', 'Content']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xDDPZjStjb1p"
      },
      "source": [
        "test['BlockId'] = test.Content.str.extract(r'(?:blk_-)(\\d+)')\n",
        "test['BlockId2'] = test.Content.str.extract(r'(?:blk_)(\\d+)')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "g-wKy4V2jolK",
        "outputId": "70a5a3cb-a25b-4bc6-f7e3-e5275fb88724"
      },
      "source": [
        "test['BlockId'] = test['BlockId'].fillna(test['BlockId2'])\n",
        "test.drop('BlockId2', axis=1, inplace=True)\n",
        "test.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Date</th>\n",
              "      <th>Time</th>\n",
              "      <th>Pid</th>\n",
              "      <th>Level</th>\n",
              "      <th>Component</th>\n",
              "      <th>Content</th>\n",
              "      <th>BlockId</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>081109</td>\n",
              "      <td>203615</td>\n",
              "      <td>148</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.DataNode$PacketResponder</td>\n",
              "      <td>PacketResponder 1 for block blk_38865049064139...</td>\n",
              "      <td>38865049064139660</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>081109</td>\n",
              "      <td>203807</td>\n",
              "      <td>222</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.DataNode$PacketResponder</td>\n",
              "      <td>PacketResponder 0 for block blk_-6952295868487...</td>\n",
              "      <td>6952295868487656571</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>081109</td>\n",
              "      <td>204005</td>\n",
              "      <td>35</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.FSNamesystem</td>\n",
              "      <td>BLOCK* NameSystem.addStoredBlock: blockMap upd...</td>\n",
              "      <td>7128370237687728475</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>081109</td>\n",
              "      <td>204015</td>\n",
              "      <td>308</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.DataNode$PacketResponder</td>\n",
              "      <td>PacketResponder 2 for block blk_82291938032499...</td>\n",
              "      <td>8229193803249955061</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>081109</td>\n",
              "      <td>204106</td>\n",
              "      <td>329</td>\n",
              "      <td>INFO</td>\n",
              "      <td>dfs.DataNode$PacketResponder</td>\n",
              "      <td>PacketResponder 2 for block blk_-6670958622368...</td>\n",
              "      <td>6670958622368987959</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     Date  ...              BlockId\n",
              "0  081109  ...    38865049064139660\n",
              "1  081109  ...  6952295868487656571\n",
              "2  081109  ...  7128370237687728475\n",
              "3  081109  ...  8229193803249955061\n",
              "4  081109  ...  6670958622368987959\n",
              "\n",
              "[5 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VeanokGym1_Q"
      },
      "source": [
        "test_event=[]\n",
        "for line in test.Content:\n",
        "  # extract only character add description of the event \n",
        "  match=re.findall('[a-zA-Z]+',str(line))\n",
        "  test_event.append(match[0:6])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AFjWufJZnDuo"
      },
      "source": [
        "test['Event_Description'] = test_event"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "eZRHGmKnnY_R",
        "outputId": "3af62af0-abb9-4d54-c246-09826d8bffeb"
      },
      "source": [
        "test_event = test.iloc[:, 6:8].copy()\n",
        "test_event.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>BlockId</th>\n",
              "      <th>Event_Description</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>38865049064139660</td>\n",
              "      <td>[PacketResponder, for, block, blk, terminating]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>6952295868487656571</td>\n",
              "      <td>[PacketResponder, for, block, blk, terminating]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>7128370237687728475</td>\n",
              "      <td>[BLOCK, NameSystem, addStoredBlock, blockMap, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>8229193803249955061</td>\n",
              "      <td>[PacketResponder, for, block, blk, terminating]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>6670958622368987959</td>\n",
              "      <td>[PacketResponder, for, block, blk, terminating]</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "               BlockId                                  Event_Description\n",
              "0    38865049064139660    [PacketResponder, for, block, blk, terminating]\n",
              "1  6952295868487656571    [PacketResponder, for, block, blk, terminating]\n",
              "2  7128370237687728475  [BLOCK, NameSystem, addStoredBlock, blockMap, ...\n",
              "3  8229193803249955061    [PacketResponder, for, block, blk, terminating]\n",
              "4  6670958622368987959    [PacketResponder, for, block, blk, terminating]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "PPzNx_2cnnBr",
        "outputId": "7774ade7-1eb9-4353-fe5b-0bc71ef9584d"
      },
      "source": [
        "test_event = test_event.groupby('BlockId')['Event_Description'].apply(sum).reset_index(name='Events')\n",
        "test_event.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>BlockId</th>\n",
              "      <th>Events</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1002945451269828722</td>\n",
              "      <td>[PacketResponder, for, block, blk, terminating]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1018961650823645078</td>\n",
              "      <td>[Deleting, block, blk, file, mnt, hadoop]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1021828391502582165</td>\n",
              "      <td>[BLOCK, NameSystem, addStoredBlock, blockMap, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>102441832197223997</td>\n",
              "      <td>[Receiving, block, blk, src, dest]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1030832046197982436</td>\n",
              "      <td>[BLOCK, NameSystem, delete, blk, is, added]</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "               BlockId                                             Events\n",
              "0  1002945451269828722    [PacketResponder, for, block, blk, terminating]\n",
              "1  1018961650823645078          [Deleting, block, blk, file, mnt, hadoop]\n",
              "2  1021828391502582165  [BLOCK, NameSystem, addStoredBlock, blockMap, ...\n",
              "3   102441832197223997                 [Receiving, block, blk, src, dest]\n",
              "4  1030832046197982436        [BLOCK, NameSystem, delete, blk, is, added]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Dzdxu3fcooWY",
        "outputId": "c86a4363-ac32-4d51-a966-f7ffc553f333"
      },
      "source": [
        "test_event.Events[0]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['PacketResponder', 'for', 'block', 'blk', 'terminating']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eRJRJeWaoLF7"
      },
      "source": [
        "test = []\n",
        "for event in test_event[\"Events\"]:\n",
        "  test.append(event)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EuE_Zvv9m3NY"
      },
      "source": [
        "# **Prediction**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JfRxWYhiITUG",
        "outputId": "2450f5ed-b845-400d-a3be-05d90a94ff63"
      },
      "source": [
        "trainDataVecs = getAvgFeatureVecs(events, model, num_features )"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Block id 0 of 575061\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:12: DeprecationWarning: Call to deprecated `__getitem__` (Method will be removed in 4.0.0, use self.wv.__getitem__() instead).\n",
            "  if sys.path[0] == '':\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Block id 10000 of 575061\n",
            "Block id 20000 of 575061\n",
            "Block id 30000 of 575061\n",
            "Block id 40000 of 575061\n",
            "Block id 50000 of 575061\n",
            "Block id 60000 of 575061\n",
            "Block id 70000 of 575061\n",
            "Block id 80000 of 575061\n",
            "Block id 90000 of 575061\n",
            "Block id 100000 of 575061\n",
            "Block id 110000 of 575061\n",
            "Block id 120000 of 575061\n",
            "Block id 130000 of 575061\n",
            "Block id 140000 of 575061\n",
            "Block id 150000 of 575061\n",
            "Block id 160000 of 575061\n",
            "Block id 170000 of 575061\n",
            "Block id 180000 of 575061\n",
            "Block id 190000 of 575061\n",
            "Block id 200000 of 575061\n",
            "Block id 210000 of 575061\n",
            "Block id 220000 of 575061\n",
            "Block id 230000 of 575061\n",
            "Block id 240000 of 575061\n",
            "Block id 250000 of 575061\n",
            "Block id 260000 of 575061\n",
            "Block id 270000 of 575061\n",
            "Block id 280000 of 575061\n",
            "Block id 290000 of 575061\n",
            "Block id 300000 of 575061\n",
            "Block id 310000 of 575061\n",
            "Block id 320000 of 575061\n",
            "Block id 330000 of 575061\n",
            "Block id 340000 of 575061\n",
            "Block id 350000 of 575061\n",
            "Block id 360000 of 575061\n",
            "Block id 370000 of 575061\n",
            "Block id 380000 of 575061\n",
            "Block id 390000 of 575061\n",
            "Block id 400000 of 575061\n",
            "Block id 410000 of 575061\n",
            "Block id 420000 of 575061\n",
            "Block id 430000 of 575061\n",
            "Block id 440000 of 575061\n",
            "Block id 450000 of 575061\n",
            "Block id 460000 of 575061\n",
            "Block id 470000 of 575061\n",
            "Block id 480000 of 575061\n",
            "Block id 490000 of 575061\n",
            "Block id 500000 of 575061\n",
            "Block id 510000 of 575061\n",
            "Block id 520000 of 575061\n",
            "Block id 530000 of 575061\n",
            "Block id 540000 of 575061\n",
            "Block id 550000 of 575061\n",
            "Block id 560000 of 575061\n",
            "Block id 570000 of 575061\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "x5qfE8NToWqW"
      },
      "source": [
        "from sklearn.decomposition import PCA\n",
        "\n",
        "from gensim.test.utils import datapath, get_tmpfile\n",
        "from gensim.models import KeyedVectors\n",
        "from gensim.scripts.glove2word2vec import glove2word2vec\n",
        "\n",
        "def display_pca_scatterplot(model, words=None, sample=0):\n",
        "    if words == None:\n",
        "        if sample > 0:\n",
        "            words = np.random.choice(list(model.wv.vocab.keys()), sample)\n",
        "        else:\n",
        "            words = [word for word in model.wv.vocab ]\n",
        "        \n",
        "    word_vectors = np.array([model.wv[w] for w in words])\n",
        "\n",
        "    twodim = PCA().fit_transform(word_vectors)[:,:2]\n",
        "    \n",
        "    plt.figure(figsize=(10,7))\n",
        "    plt.scatter(twodim[:,0], twodim[:,1], edgecolors='k', c='r')\n",
        "    for word, (x,y) in zip(words, twodim):\n",
        "        plt.text(x+0.05, y+0.05, word)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 431
        },
        "id": "pAiitC7_oXhA",
        "outputId": "509af2da-84d9-44e2-8ef2-d317acba265a"
      },
      "source": [
        "display_pca_scatterplot(model, sample=300)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnwAAAGeCAYAAAAOidxCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3xUVf74/9eZmZRJITEUKQIBlGAoYaUsKCiKUpSl/yzwWbGBBlBcXddl/ayAfi0ry7oCgor7WXGBRcWCuBQpyoqCEJAAQRAQQYiRZiBlUmbm/ftjMteZFAgkmPZ+Ph73kbnnnnvuuSOGN6caEUEppZRSStVetqqugFJKKaWUurg04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKXXTGmD3GmDlVXQ+llKqrHFVdAaVUnXAF+vtGKaWqTLX9BdygQQOJj4+v6moopUpx9OhRMjIygtKuvPJKTp48ybFjx6y0hg0bkpeX5z9tY4yRRo0a0bx581+uskop9QvYunXrCRFpWNX1KEu1Dfji4+NJSUmp6moopUrx1FNPMWXKFO644w4WLVqEzWbj9OnTHDt2jEcffZS//vWv3HLLLSxfvhwRwWazER8fz7ffflvVVVdKqYvCGHOoqutwNjqGTyl1wRYtWgSA0+nk1KlTAMyYMQNjDMuXL6/KqimllApQbVv4lFI1hzGGwsJCAHR/bqWUqn60hU8pVSkcDt+/H1u3bg1AYWEho0ePBsBms5GTk1NldVNKqbpOAz6lVKWZMmUKBw8exBhDaGgo7733HgB9+vTh2LFjGGOYPHlyFddSKaXqHlNdu1+6du0qOmlDKaWUUjWBMWariHSt6nqURVv4lFJKKaVquUoJ+IwxA4wxe40x+40xfyzlegtjzCfGmK+MMTuMMTdXxnOVUhVnjKnqKiillLrIKhzwGWPswMvAQCARuMMYk1gs2/8Cb4vIr4DbAd1iSalqoroO61BKKVV5KqOFrzuwX0S+FZECYDEwpFgeAeoVfY4B0ivhuUqpSuBv4cvPz8cYYx3R0dFVXDOllFKVpTICvmbA9wHnR4rSAk0F/scYcwRYDjxYWkHGmHHGmBRjTMrx48croWpKqfK69NJLATh27BjTpk0jOzubP/3pT1VcK6WUUpXhl5q0cQfwhohcBtwM/MsYU+LZIvKaiHQVka4NG1bb7eiUqpVOnz5NeHg4DRs25MknnwTg1VdfreJaKaWUqgyVEfAdBQJ3Qr+sKC3QvcDbACKyEQgHGlTCs5VSSiml1DlURsC3BbjCGNPKGBOKb1LGh8XyHAb6AhhjrsQX8GmfrVLVSExMDHl5eRw/fpwXXngBgAkTJlRxrZRSSlWGSll4uWiZlb8DduD/ROQZY8xTQIqIfFg0a3ceEIVvAscfROTjs5WpCy8rVblaNmtGZno6WUA0ENu0KYeOHsUYg4iQn59PeHi4lT8qKoqsrKwqq69SStUk1X3hZUdlFCIiy/FNxghMezLg827gmsp4llLq/LVs1oyC9HQ+AHoBG4BR6em0bNbMWpYlLCxMl2hRSqlaSnfaUKoOyExPZxFwPRBS9HNRUXpxuhyLUkrVPhrwKVUHZOFr2QvUqyi9RF7txlVKqVpHAz6l6oBofN24gTYUpRdnjCEjI4O4uDgiIiIIDw+31uPr0aMHt956q5W3T58+DBo0qMz8SimlqgcN+JSqA2KbNmUU8AlQWPRzVFF6qfljY9mxYwe5ubmkpqYyffp0vF4vkyZN4uOPf55vtXHjRv785z+Xmf98PPHEEwwYMOCC3q9Pnz7Y7XacTidhYWG0b98et9sNwOWXX86jjz563mXed999dOrU6YLqo5RS1Y0GfErVAYeOHiW0aVOGAmHAUCC0aJZuabxeLzfffDNOp5POnTvjdrvZtWsXd9xxB6dPn2bbtm28/fbbhIWF0a1bNwoKCkrNfz6eeeYZVq5cecHvOHDgQFwuFzk5OXz//ffMnDnzgstSSqnaRgM+peqIQ0ePcloErwinRcoM9gAeeughMjMzOXnyJC6XC7vdzpkzZwDo3bs3v/vd77jtttvweDxERESQkJDArl278Hq9dO/e3co/btw4nE4nTqeT1q1bA/D111/TrFkzIiMjiYyMZO7cucDPLWqHDx/G4XBYLXTHjh3D4XCQm5vLunXraNCgAREREcTExLB8+fISdc/Ozsbj8dCsWfEdHmH69OlWt3Pbtm2td5o/fz7R0dE4nU6ioqJILzaZZerUqURHR7N3794L+OaVUqrqacCnlCrh1KlTxMbGEhERwd/+9jc8Ho917bPPPmPLli0AeDwewsLCyMjIwOFwkJWVxdatW/F4PPz3v//ljTfeYPv27bhcLlavXg34WuKeeOIJcnJyWLNmDQ8//HDQs1u0aEHDhg2tFrpnnnmGNm3aEBERwciRI/n3v/9Nbm4uf/vb3xg9erR134oVK3A6ncTFxdGwYUNuu+22oHIzMzOZPHky77//Pnl5eXg8HsaMGUN2djb33Xcfs2fPxuVysWfPHuLi4qz7Hn/8cV588UV27NhBQkJC5X7RSin1C9GATylVwnPPPcfBgwcJDw9n7ty5hIaGBl33t77l5+fTtGlTjDEUFhYSFhZGTk4ONpuNVatW0a1bNytIatOmDQCHDx/m0Ucfxel0csMNN+B2u8nIyAgqf/jw4bz++usAvPPOO9x9991kZGTw008/MXjwYJxOJxMnTiQ3N9e6x9+lm52dTX5+Pg899FBQmR9//DGRkZH0798fgAceeIAvv/ySVatW4XQ6GTNmDACXXXaZtQD13r17efnll0lLS6NVq1aV8t0qpVRV0IBPqRqoXkQEMcZgM4YYY6gXEVEp5e7btw+73U5CQgJZWVnk5eWxb98+8vPz6dXr54Vd1q1bh91uJy4ujsjISO644w4SExN54IEHEBE8Hg8NGzYs8zk//PADLpcLl8uFx+OhcePGQdf//Oc/880333DgwAGOHTvGI488gtvtxmazWfe5XC7y8/NLlB0REUGfPn1Yt25dhb+P2NhY3G53pZSllFJVSQM+pWqYehERRLpcfADkAx8AkS5XhYO+bdu20b59e4YOHVruexwOByNGjGDp0qV4vV4KCwtJSUkhLS2N3/72t2zZsoV9+/YBcODAAcDXZRvYFfvWW2+VKLdx48Y0btyYm266iXbt2hEaGspll11GREQEjzzyCOCbWPL222+XuNfr9bJp06YSLXL9+vUjJyeHtWvXAvDqq6/So0cP+vfvj8vlYv78+QCkp6eTl5cHwKWXXsp7773H2LFjWbp0abm/F6WUqm404FOqhjEuV6m7ZhiXq0LlXnXVVRQUFLBkyZLzum/y5Mn069ePvXv38s9//pPrr7+eH3/8kSFDhjBmzBg6duyI0+m0ulJXrVrFzp07rSVUpk2bVmq5t912GwcPHmTs2LFW2n/+8x/efPNNayLIK6+8Yl3zj+GLiIjA6/VaAZxfbGwszz33HL/5zW8IDw/HZrPxxhtvEBUVxeuvv86ECRNwOp20bduWzMxM676bb76ZN954g1tvvVVb+pRSNZaprntndu3aVVJSUqq6GkpVOzZjyMcX7PkV4ltuxfsL/P9sjEFE+Pvf/85zzz3Hjz/+CECnTp3o3r27NfZOKaXqEmPMVhHpWtX1KIsGfErVMDHG8AG+lj2/T/CtrXe6mv7/rJRStV11D/i0S1epGkaczlJ3zRCns0rrpZRSqvrSgE+pGuZMbi45TmfQrhk5TidnApYouVDGGGw2m3W0bdu2wmWeyzvvvMNNN91knU+bNo0GDRpc9OcqpVRdogGfUjXQmdzcoF0zKiPY8/N6vdbxzTffVFq5ZVm/fj3//e9/rfMpU6Zw4sSJi/5cpZSqSzTgU6oWeHD8eOIcDmzGEOdw8OD48ZVW9q5duzDGMG/ePAAiIyPp2bMnAH369LFaA6OiogD4/PPPcTqd2O127HY7v//97wGIj48nKioKu92OMcYqY+7cuRQUFGCz2ejYsSMTJ04kLCwMgK1bt1ozau12OzNmzLDKiomJweFwYIwhKSmp0t5XKaVqIw34lKrhHhw/niVz5/Kux0M+8K7Hw5K5cy846Avs0h04cCAdOnTgnnvu4YEHHmDAgAEUFhayceNGZs2axfr16/niiy/wer2sX78e8K13l5ycjMfjYcmSJVaQBpCbm8vhw4fZuHEjmzZtYtWqVSQnJxMaGorX62Xnzp1Bdenfvz9NmjTB6/UyduxYHnvsMetadnY2P/zwAxs3bmTHjh2cPn36gt5XKaXqAp2lq1QNF+dw8K7HU2LW7gi7nVNFW6CVl3/JldLExsZy+vRpli1bxqBBg+jcuTMnTpzgyJEjJcowxljnIsL+/fvp27cvXq+Xw4cPAxAdHU2/fv1o0qQJ8+bNs3bNmDhxonVus9lYsGABo0aNssrevXs3AwcOxOFwsH//fivdXy+llKoKOktXKXVRZXo89CqW1qsovbgL7fotKCggOzsboFzj+tLT061xgCJi7aMbGAiWdn4+QkJ+XonQGIOrggtPK6VUbaYBn1I1XKzdzoZiaRuK0v0eHD+eWJuNl+fOJc7jYT7n1/XbuXNnoqKimDRpEo899hinT5/m3nvv5ejRo2zevBnwjbcD316211//c3vjX/7yF+vz999/T0ZGBps3byY7O5uxY8dy6aWX4iklOAWIi4vjf//3fwFfy58xhiuvvLIc34pSSqkgIlItjy5duohS6twmJidLY5B1IAVFPxuDTExOLvN6K5BFRZ8vsdutsgAxxlhHkyZN5LXXXhNA9u7dKyIijRo1khYtWoiIyLXXXmvljY6OFhGRjRs3itPpFGOMABIbGysiIi1btpSoqCix2WwCSI8ePUREJDMzUxwOhxhjpEOHDjJhwgQJDQ0VEZGUlBQJCwsTY4zYbDb561//apXVrl07q97GGHn77bcv5teslFJnBaRINYifyjp0DJ9StcCD48ez8LXXyPR4iLXbGT1uHLPmzAHKHuP3IPAVlbslW0JCAtOnT2fw4MH069ePjz/+2LoWHx+P0+nk66+/DrrH4XBYs21FhEceeYRnn30WOPuYwrOJjY1l5syZ3HnnnRV7IaWUKicdw6eUuuhmzZnDKbcbrwin3G4r2IOyx/h9Tcmu34ooKChg7969DB48GIDVq1eX+97U1FRcLhfvv/8+L7zwQqXURyml1M804FOqlitrjF8rfFuyjR43rsx7b7nlFkaMGAHAVVddxSWXXALAjBkziI+PxxhD165dcTqdvP7668TGxvLmm2/So0cPAJxOJ/Hx8QAMHDiQ77//HqfTyZVXXklBQUGJ5x07dixoMoaf1+ulW7duhIeHEx4ezqRJk6xrAwcOJDw8HKfTaT3Xz+12c/nll9OrV/GQt/I4nU4ef/zxi1a+UkpVBg34lKrlRo8bV2Lv3TuAH4CRyclBrYHFDRs2jC+//BKAAwcOUFhYSG5uLh999BHXXHMNAL1798blcjE+YPLHpk2bAHC5XHz33Xd89NFHvPfeexw7dgyXy4XdbmfixIlW/qSkJMLDw7nrrruC1trze/zxx/n22285c+YMKSkpvPzyy2zfvp2nnnqKDRs2cOTIEVwuFwsXLrTuKSgooE2bNrRq1YoNG4qHvJUnLy+PNWvWXLTylVKqMmjAp1QtN2vOHEYmJzPCbicM3/p8/19yMjkiZw32AEaNGsWPP/7IkSNHsNvtXH755SxYsIDU1FRuu+02IHgWblnmz5/PiRMnqF+/Pk6nk2+++SZoeZfU1FTy8vJYu3Ytzz77LBkZGUH3r127lsGDBxMaGkqHDh1o0aIF77zzDh9++CHDhg2z9t71L/8C8OCDD9K2bdvz6lr2848pNMbgcDi45557gpaQcTgc2Gw2YmJiANi2bRvGmIsaWCqlVEVowKdUHXC2MX5nExERQXR0NJMmTaJDhw707duXJUuWcObMGWuR49DQ0HOWIyJ0794dl8uFy+WioKCATz/9tES+G264gfDwcFasWHFe71ea1q1bs3XrVjIzM8/rvm7duuHxeFi5ciUfffQRHo+nxEQTP//uHldddRUiclG7jpVSqiI04FNKnVWnTp1YunQpgwYN4u6772bdunU0atQIm+3svz6MMeTm5gJw5513snXrVtLS0gBf9/Dnn39e4p60tDRyc3Pp1q1bUPqNN97IsmXLKCgo4Ouvv+bw4cPcdtttDB06lPfff58TJ05Y5fo9/vjj9O7dm3bt2pGXl1fu9/3qq68A37Zut9xyCwBbtmwp9/1KKVUdOaq6Akqp6u2WW25h/fr13HXXXVagVzwgK023bt245JJLaNKkCd999x3Jycl07epbscAYw8yZM61xgElJSdYSLGPGjKFDhw5BZT3//PN88skn1KtXD4AJEybQqVMnOnXqxIYNG2jWrBk2m41f/epXfPHFF9Z9S5cu5brrriMhIYEDBw7gcFzYr7ziS8NU1+WslFKqLLoOn1JKBejevTtbtmxh3bp1FBQUMGDAADp37sz27duZNWsWvXr14le/+hXGGLxeL8YY4uPjOXjwYFVXXSlVhar7Onwa8ClVww248UY2rl1LFhAN9Ozbl5U6a7RCHA6Htd2b3W7H7XYHpQFWwBcbG2uN5fvss890HJ9SdVR1D/h0DJ9SNdiAG28kde1aPgDygQ+A1LVrGXDjjVVcs5rN7XZb2xG53e4SaSKC1+sFIDMz00rTYE8pVV1pwKdUDbZx7VoWAdcDIUU/FxWlK6WUUn4a8ClVg2VBqdumZVVBXZRSSlVfGvApVYNFQ6nbpkVXQV1qG2MMNpsNm82G0+nk8OHDFS7z2muvpX79+oBvFnOfPn0qpX4Oh4N33333gsp55513rCV2pk2bZi1ifb46duwY9B2FhYWxY8eOCyqrXr16GGOCFudu2LAhxhhrF5fz1aJFC8aOHVtqXZWqCzTgU6oG69m3b4lt00YVpauK83q9eL1ebDYbN998c6WWvWXLllIXnz4fX3zxBV6vl2bNmnHvvfdWuE5Tpkyx1jQ8X7t27SI9Pd06z8/Pp1OnThdcF2MMycnJgG+bvJMnT15wWQCHDx9m3rx5pda1PFwuV4Wer1RV04BPqRps5Zo1JPXty1AgDBgKJOks3UqXmJhIeno6Bw8eJCYmBrvdjs1mY+TIkYCv5S48PJzQ0FCMMTRp0sS6t3fv3hhjsNvtQS1e8fHxXHnllYBvVnDTpk2x2+0YY3j44YcBX5ASERGBMYbw8HDsdjvTpk0rUb8+ffpYAcnnn3+O0+nEbrdjt9v5/e9/bz0vKirKekbPnj1LlDNx4kTCwsIA3yLW9erVs1oRhwwZAkBcXJz1/i1btgR86ygCXH311YSEhAAEtcZ16NDBKqdjx46Ar2XRGENsbCw2m42QkBCOHj1q1aVNmzbWVnUPPfQQsbGxQXU93zLr1avHb37zm1Lr2r9/f6uspk2bWs8wxnDppZdis9l44oknSnxfStUkGvApVcOtXLOG0yJ4RTgtwso1a3hw/HjiHA5sxhDncPDg+PFVXc0aacWKFcTGxrJz504iIyNp3bo1Xbt2xePxsH37dt59911r/b38/Hy2bdvGDz/8QEZGBh9++CGrV69mw4YNbNiwgePHj5OVVfroSo/HYx2jRo3i5ZdfBnyBSFhYGCLC888/b80MLm758uV07twZgH79+pGcnIzH42HJkiXMmDHDypebm8vhw4fZuHEjmzZtYtWqVWW+e+/evQkLC7NaOf2B5urVq/F4POTk5HD06FFmzJhBamoq4GtxLCwsDCrn6aefZvfu3Rw4cIB9+/aRlpbGc889Z12fPHkyXq+XkJAQ7rzzTiu9Xbt2uN1utm/fzuLFixk9enSFywRK1HXVqlV8/PHHfPbZZ5w5c4bjx48zYsQIK3/Xrl3xer387W9/K/O7UqpGCFxmoDodXbp0EaXU+ZuYnCyNQdaBFBT9bAwyMTm5qqtWowDWUb9+fTl58qTYbDYxxlgHIHPmzJHevXtLTEyMdW9ISIj87ne/k+HDh0tUVJSV3qlTJ4mLixMRkZYtW0q7du2sZz366KMiIvLJJ5+I71ezSFhYmCQH/HczxsjUqVOte/x1AGTv3r1B6YF13L9/v7Rs2VKaN29ulRUVFSXDhw+Xt99+W4wxIiIyYcIECQ0NtZ71z3/+s8T30rVr16Cy+/fvbz1348aNQd/fxo0bpWPHjkHPbd68uXTq1Enefvtt6z1FRNq2bSutW7cWEZHo6GgZNGiQJCQkSJcuXcQYI/n5+ZVSZvG6jhgxIui/0dVXXy2NGjWy8uXm5pb4DpQqDZAi1SB+KuuolBY+Y8wAY8xeY8x+Y8wfy8hzqzFmtzEmzRizqDKeq5QqaeFrr5W6VMvC116r0npdbPUiIogxBpsxxBhDvYiICpdpjAHg5MmT1K9fH6/Xy6uvvkqjRo2s7dXGjx/PF198wenTpzHGWAsyFxQUsHPnTrKzs630I0eOADB16lQOHTrEnj17rC7U6GjfVJvw8PBy1++LL74gKysLp9NJ7969rfT09HSrZU5EaNOmTdD7FH+/8nrrrbdISUlh27ZteL1eoqOjrf2SL0Tg8202W9DC1gDPPPMMW7dupWHDhoSGhlZKmefL6XRW6H6lqosKB3zGGDvwMjAQSATuMMYkFstzBTAZuEZE2gMPV/S5SqnSZXo8pS7VklnBv/jK65prrrngmZ7FlTcgqRcRQaTLFbQAdaTLVeGgb+rUqYCvJ8TfpfnHP/7RCvbGjh3LggULrKBCRLDb7db5vn37APjyyy+56aabOHXqFOCbDWuz2WjXrh0RZ6ljq1atWLx4MQCzZ88udQ/fqKgoli5dyrFjx9i6dSsRERFcf/311vW//OUv1ufvv/+ejIwMNm/eTHZ2tjVrtTSNGzfmscces863b9/OoUOHMMaQmJjIp59+GtRFbYzh0KFDJcq59dZbOXLkCIcPH+bgwYMcOXKE22+/vcznBhoxYgTdunUL6pauaJnF63r//feTnZ3Npk2bcLlcbN68OSh4VqrWqGgTIdATWBVwPhmYXCzPC8B951OudukqdWEusdtlHYgEHOtALrHbz3pfYWGhuFyuX6iW5UNA99zZ1Ct6x+LvXK+c95f17GnTpll18H+OjY21ulH93Z+AOJ1OERFp1KiRADJs2LCgbmH/ERcXJ4A0bdpU2rVrJ+vWrRPA6qrduHGj9cxvv/1WwsPDBZCwsLCgblaKdaHWr19fWrVqJRs3bhSn02l1ucbGxoqIrws5KipKbDabANKjRw8RkTK7dPfv3y9RUVFW9+2QIUNExNc1CojD4ZCwsDDp3bu3iIgkJSVZ6cXr1759e6ucDh06lHiuiEi7du2kZcuW1jP83a/F/5tURpnF69qvXz+rrCZNmgQ9T6nyopp36VZGwDcSeD3g/LfA7GJ5PigK+j4HNgEDyihrHJACpLRo0aLCX75SddH5jOGbPXu29Rdf8WCmXr16Vr6EhAQr3V4UOC5btswKHgC5/fbbRcQ3biosLEy++OILAawgcteuXQLI8ePH5a9//WvQ+DN/sDNz5kwrzR9YlIcpetfAgK8AxFTwL+zSAj4RkUsvvVRsNpuVD5DFixcHXfMHcqXxfw8ictZ8ubm58sMPP4iIyL/+9S8B5OTJkxf0LoFjBpVSla+6B3y/1CxdB3AF0Ae4A5hnjIktnklEXhORriLStWHDhr9Q1ZSqXWbNmcPI5GRG2O2EASPsdkYmJzNrzpwy77njjjvo168fOTk5FBYW4nK5yMrKYuTIkfzpT39i7969rFixAhHh448/BmDo0KGMHDkSEeHVV1+1uh79evbsic1m47bbbgNg9OjRhISE0KBBAx577DGeffZZRIQxY8ZY3aWTJk0iISEBESEmJqbc73yxFqBu3779Bd/r71Zt3bo1AIWFhUEzTf3LgwTOCC3u+++/p1mzZthsNu68807+53/+h7i4uAuuk1KqDqtoxEj5unRfAe4OOF8LdDtbudqlq9TF52/hExGJiIgo0f3Ypk0badq0adAsRr/ieQFJTU21WvhERDp06GB1EdpsNunfv7+kpqaWeq+/TH/L1+bNm8vdwhftdJbaqhld1M1aEYEtkZxHC5+IyJQpU4LuDQ8PL5EeGhqqXYdK1QLUgRa+LcAVxphWxphQ4Hbgw2J5PsDXuocxpgHQFvi2Ep6tlKokIkLLli2DfkHs37//rPccPHgwKH/xnRUWLlxIQUEB69atw+v1smTJEvLz863nBR5+DofjvOt+JjeXHKczaAHqHKeTMxWYQernn+kaWM+MjIyg2Z8iYrVkBl6bOnVq0L3+xZED0/Pz84PeXymlLoYKB3wi4gYmAquAr4G3RSTNGPOUMWZwUbZVwEljzG58uz89JiIV2ydHKVWpevfuzaFDh6zdIN58802WLl3KmDFjyM7OZvXq1QCsW7cOALvdTvfu3a37J02aVKLMTp06YbPZ6NevH2FhYURFRdGtWzcAunTpAoDb7bZ2ljDGcM011wBYu1iU15nc3KAFqCsj2FNKqdqiUsbwichyEWkrIm1E5JmitCdF5MOizyIij4hIooh0FJHFZy9RKfVLW7VqFU2aNCEpKQljDGPGjGH//v08++yztG3bln79+mGMoX///gB89NFHnDp1ylpj7pVXXim13M6dO+PxeIL2op05cyZfffUVxhhCQkKYP38+AC+99BK7d+/GGMNPP/108V9aKaXqCFNduxK6du0qKSkpVV0NpZRSSqlzMsZsFZGuVV2PsuheukoppZRStZwGfEoppZRStZwGfEqpWmPAjTcG7ac74MYbq7pKSilVLWjAp5SqFQbceCOpa9cG7aebunatBn1KKYUGfEqpWmLj2rUsAq4HlgAPAseK0ls2a1aldVNKqaqmAZ9SqlbIAnoB/waeAGYBefha+grS0zXoU0rVaRrwKaVqBf9+us8A/8DX0hdS9HMRkJmeXnWVU0qpKqYBn1I1UFL79kGTE5Lat6/qKlW5nn37Mgrfdj+9il3rha8FUCml6ioN+JSqYZLat+fY7t1BkxOO7d5d54O+lffk0/AAACAASURBVGvWkNS3L1H4WvoCbcDXAqiUUnWVBnxK1TDf7d5tTU4I7LL8bvfuKq1XdbByzRpimzZlFL5NuwuLfo4CYps2rdK6KaVUVdKAT6kaxj85IZB2Wf7s0NGjhDZtylAgDBgKhDZtyqGjR6u4ZkopVXUcVV0BpdT58U9OuD4gTbssg2lwp5RSwbSFT6kaJj4xsdQuy/jExCqtl1JKqepLAz6lapjUtDQaJSYGdVk2SkwkNS2timumlFKqutIuXaVqIA3ulFJKnQ9t4VNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKKaWUquU04FNKVSpjjHXYbDY2bNhw3vdXZl1Wrlx5Qff27NmTJ554otLqopRSVUkDPqVUpRMRRAS73c6QIUOqujoXJCUl5YKDRaWUqm404FNKXTRNmzYlJycHgBkzZmCz2azWv2nTpgEwa9YsK61evXrWvcOGDcNm+/lXVHh4OAkJCYCv5S4mJqZEWatXr7aeERoaGlQXu91u5W/VqpWVbowhOjrauvbee+8xevRo3G4327ZtwxjDjBkzLs4XpJRSvxAN+JRSF0V2djZHjhyhV69eADz22GM8++yziAhjxoyxgrRJkyaRkJCAiBATE1Pu8qOiohARmjdvzjPPPAPA4MGDiYmJQUTo2bNnUP7Vq1cjIuzbt4/vvvuO1atXW9c6d+6MiBAZGcl9993HwoULcTgcXHXVVYgIjz76aEW/DqWUqlIa8CmlKp2/1czr9fLBBx+wY8cORITJkydjjGH+/PmICODr/vWP83vvvffK/YznnnsOgBtuuAG32w1AXl4eU6ZMAWD9+vVB+YcPH44xhiuuuAKAf/3rX9a1Tz75BIDLL7+c7OzsC3llpZSq1iol4DPGDDDG7DXG7DfG/PEs+UYYY8QY07UynquUqp5EhD179gCQkJBAfn6+lR54+DkcjhJlFO+S9Xq9QeeNGjWy8gWWVZphw4Zx+vRp9uzZY+XNysoq8Xy73X7OspRSqiaqcMBnjLEDLwMDgUTgDmNMYin5ooFJwJcVfaZSqvpLSEhg4sSJpKenW61qXbp0AcDtdvPwww8DvtbAa665BoCRI0da9/fr1w8R4ciRI2zYsIHCwsJzPjM8PJynnnoK8LX8+f34448YY0hISLC6ks/F4XAEBYVKKVWTVUYLX3dgv4h8KyIFwGKgtGl5TwN/AfIq4ZlKqRrAPyHj6quvZubMmXz11VcYYwgJCWH+/PkAvPTSS+zevRtjDD/99JN177333ktkZCTNmzfn2muvDZrAUZYPP/yQzMxMjDFBy8G8/fbbgC+49AeE5zJ06FD27dunkzaUUrWCqWj3hTFmJDBARO4rOv8t8GsRmRiQ5yrgCREZYYz5FPi9iKSUUtY4YBxAixYtuhw6dKhCdVNKKaWU+iUYY7aKSLUdsnbRJ20YY2zA34BzTnMTkddEpKuIdG3YsOHFrppSSimlVJ1QGQHfUaB5wPllRWl+0UAH4FNjzHdAD+BDnbihlFJKKfXLqIyAbwtwhTGmlTEmFLgd+NB/UUROi0gDEYkXkXhgEzC4tC5dpZRSSilV+Soc8ImIG5gIrAK+Bt4WkTRjzFPGmMEVLV8ppZRSSlVMycWvLoCILAeWF0t7soy8fSrjmUoppZRSqnx0pw2llFJKqVpOAz6llFJKqVpOAz6lFEnt2xNjDDZjiDGGpPbtq7pKSimlKpEGfErVcUnt23Ns924+APKBD4Bju3eXGfTNmzcPm82GMcY6GjRogDHmotfVGMOKFSsu+nOUUqq20YBPqTruu927WQRcD4QAs4ETwM7duzlz5gxxcXE4nU4mTZpEYWEh48aNIzY2FhFBRBg1ahQ5OTklyv373//Oq6++Wql1FREGDhxYqWUqpVRdoAGfUnVcFtAr4PwjYH3RZ/8etC6Xi5deeolrr70WgFOnTln5Fy5cSJcuXQCsFj+bzcaSJUtYtmwZTqczKL2wsNDKGxoaal3r06cPANu3by/Rgjhs2DDrnrfeeounnnrKKs+fZ9euXQAMGTLESgsJCflFWh6VUqq604BPqTpsyJAhCBAJDAMSgQKgD75fDsnJyfz00084nU7WrVvH5s2bAXA6nXTt6tss5+uvv7bSQ0JCmDFjBiLC559/zooVK3C73cyaNQv/vt1JSUnW8/2thB06dGD9el+Y2atXL4wxiAh33333Wet/++23IyIYY+jfvz8AH374If369bPSlVJKacCnVJ21YMECVq1aRWLbtsQA/wEm4/ulEAO0T0xk+vTpNGrUCJfLRfPmzfF6vdjtdlwuF++++y4AAwcOpHlz3+6K69evZ/LkydhsNkJCQhg4cCAtW7bkwQcftIK4Q4cOWXW47rrrAPjd735npeXk5NCjRw8A/u///u+s77Bo0SLAF4CeOnXKGt+3atUqwBcQKqWU0oBPqTrrvffeo3v37qTt3UvTxEQ8wJ2AF4hr04bUtLSg/P6gzuPx8Ic//IH69esDcPjwYb799lsAbrjhBtxuN+BrvTtz5gwHDhzghRdeQEQICwvD4/FYZUZHRwMQGRlZoXfxB5NKKaVKpwGfUorUtDSu7tWL4cOHY7fb+fA//ymRJzw8nBMnTgAwY8YMKwD0B3LgG+sXGNDl5eUBvlbATZs2kZ+ff866REZGsmnTJgDGjRt3Xu/hn9Bxyy23ALB48eLzul8ppWorDfiUqqNGjBjBli1bOHHiBMeOHWPLli2MHDmyzPwZGRlkZGQwe/ZsvF4vmZmZ1hg5r9dr5XvrrbcAsNlsOJ1OQkND6dixIz179izXmLpPPvkEr9eLMYZ58+YB0KZNm3K/180338zy5csxxgTVSyml6jT/oOnqdnTp0kWUUhfX4MGDJSwsTMLCwmTo0KEiImK322XPnj0iIvLiiy9Ko0aNRETkq6++ksjISAkPD5ewsDC57777RERkz5490rx5cwkPD5fQ0FC58sorRURk5cqVEh4eLuHh4TJr1qxy1ykzM9N6/h//+EcB5NChQ+W+f+PGjdbniIgIsdvt5b5XKaUuFJAi1SB+KuswUk3HvXTt2lVSUlKquhpKqV/YZ599Zi3/AtCpUydSU1PLfX/Lli05fPiwdb5gwQJGjx5dqXVUSqnijDFbRaRrVdejLI6qroBSSgXq3bt3hSZgBM4CVkop5aNj+JRSSimlajkN+JRSSimlajkN+JRSSimlajkN+JRSSimlajkN+JRSSimlajkN+JRSSimlajkN+JSqYR4cP544hwObMcQ5HDw4fvw572nZrBkxxmAzhhhjaNms2S9QU6WUUtWFBnxK1SAPjh/PkrlzedfjIR941+Nhydy5Zw36WjZrRkF6Oh8A+cAHQEF6eoWDvvJsk9akSRP27t1boedciA0bNtCiRYvzvi82NpZGjRpdhBoppVTV0oBPqRpk4WuvsQi4Hggp+rmoKL0smenppd6TmZ5+0eubkZHBwYMHz+ue7Ozss56XR2pqKt9///1536eUUrWVBnxK1SCZHg+9iqX1KkovS1ZRnuL3ZFVSnYYNG4YxBrvdjjEGh8OB2+2mWVEL4sCBA7HZfL9q+vfvjzHGyr9jxw7A11oYFRWFMYZrrrmm1PNx48YBsHLlSqt1MSEhwXquMYZLLrkEgEmTJlnlRkZGAtCoUSMrX0xMjFX/+vXrW+kXElwqpVRNoAGfUjVIrN3OhmJpG4rSyxJdlKf4PdGVWzX+8Y9/4HK58Hg8jB49mqNHjwKwYsUKvF4vK1eu5OOPP2bXrl2ICHFxcVx33XXW/U6nExGx9s0tfl4Wr9fLxx9/zL59+8jMzGTcuHG89NJLAIgIOTk59O/fn8zMTAoLC3G5XGRlZTFy5EiSk5M5deoUe/bsYfPmzXjOEjgrpVRxFzp8pCroXrpK1SCjx41j1Ny5LMLXSrcBGFWUXpbYpk0ZVdStG3hPbNOmlVYvYwx33XUXAKGhoezatatEnmeffRaADh06WGkhISHW56effjoof/HzsjgcDm666SYA6tWrx7Jly0hKSgrKs2HDBgoLC4Oet337dnJycqhXrx4JCQmAL8hUSqnyqknDR7SFT6kaZNacOYxMTmaE3U4YMMJuZ2RyMrPmzCnznkNHjxLatClDgTBgKBDatCmHilrgKpsxBrfbXSJdRLDb7YiIdRQUFFjX4+Pjg/IHngeWmZGRcc7nl/bsli1bBj17//795/FWSqna6uWXX8YYQ0hIiDUs5eabb7aGetxzzz3ExsYSGhpqpfmHrAQOHwF+VXVvcW4a8ClVw8yaM4dTbjdeEU653WcN9vwOHT3KaRG8IpwWuWjBXmn27dsHwOOPP47H42H69OkApKWl8cwzz5SrDIfDwdq1awGYPHly0DW32826devYv38/Z86cYdCgQbRt2zYoT+/evTl06JA1ZvDNN99k6dKlDBs2jDNnzrB//362bduGy+Wq0LuWJrBVUSlVfT388MPWsJQ1a9ZQWFhIz549mT9/PgCFhYV8//33rFixgvT0dE6cOBE0fAT4qupqf24a8CmlLpqmTZvy0EMPYbPZGDRoEIMGDeIPf/gDxhg6dOjA8uXLy1XOE088weHDhzHGkJOTE3TNZrNx4403csUVVxAbG8trr73GTTfdZE3miIyMZNmyZVx66aUkJSVhjGHMmDHs37+fOXPmEBcXxxVXXEGXLl2wn2Us5IUqrbVTKVX9TJ8+nfDwcBwOB23btsXhcHDrrbfi9Xo5ffo0AK1bt2bAgAEANG7cuCqre9404FNKXZCif9Hy/vvv4/V6rfS8vDxr7b2jR48iItb1ZcuWBXWrfv7551ZZ/l+ipZ1PmTLFuufMmTPWs8HXgub1ehERfvrpJyv9wQcfBCA3N5ewsDBOnjxpXbPZbDz66KMAnDlzxkr3eDzWL3ZjDH369AFg3rx5VlfxqVOnsNlsVteOfxbw888/b6UZY+jcubMVQPrTlFI1Q1hYGPBzC31oaCgNGzYMGoZS02jAp5Sq1Xr06MHll1+O2+0mPz+frKwsvF4vSUlJ9O3bF7fbzRdffMGyZcvKVV6LFi2ssYhpaWnk5uYyadIkpkyZgsPhQEQ4efIkf/nLX6xZv/5gVSlVs0ycOBGAgoICjh8/jjHGWmbK4/FYY42L/kHXxRgjxpg3/fcbY0YaY7zFyw1UlGdVwPmTxpjjlf0uGvAppWqsvXv3kpeXd9Y8Gzdu5MCBA4DvX+3R0b4Fab799ls+//xzbDYbPXv2ZNCgQeVqhcvJycHtdmOMoX379gAsX76cdu3a4Xa7cTgcjB49mv79+1fw7aqXwNZLYwyjR48Gfh7wXtwbb7wRtEai0+kMmnBzzz33BJUXGhrK3r17SUhIIDw8HPC1FoeEhBAaGqpd46pKzJ49G8D6MwkE9WiMD97lyAt8Bvz2PB/Tp+gAQESeEpGG51nGOWnAp5SqEyIiIoK6k7Oyzr30tD/IKL7sQps2bYLK2rdvH6mpqfy///f/CA8PZ+XKlcTGxl6U96hK/ve98cYbWbRoUZn53G43d999t/U9ZWVl4fF4aNOmDQDvvfce//znPxk9erRVZseOHUlLSwsqo169elY3vsOhq4ipi2PChAlBLfCFhYVs3brVugYwc+bMoHvsdjt2u53Dhw8HJtuA3gDGmC+LP8cYE2aM+cEY4zHGeAPyJAOhRWk7jTGzjTH5Rfd8Z4w5bYxxF7Uebg8o70BRmtsYk2OMSSv+zOKVU0qpWq1Nmzbk5uaybt06AKZOncqsWbPo1asXXq+XlJQUVq1aVaLb9auvfJPuXnjhBSstMjKSAwcOWL/o77rrLv773//y+9//nuTkZLKzswkPDw8aG1jsL4Uazz+OsSxDhgwB4JtvvgEgKiqKzz//nNzcXPbu3cvdd99NvXr1WLBggXXP1q1bGT58uHUeHR2Nx+PhxIkTQa0rSlXUd999F7Qe6PmKiooCCJrk1b17d4B8ETFAA6CrMebWYreuB3JExA40DMgzFygQEZuIdCzlkZHApUAPIMkYE22MmQq0KkpvB0Scq94a8Cmlar19+/YRFhZG3759McYwbdo00tLSWLNmDQ6Hg27dugVNEgEYNWoUubm5GGMoLCy00jMyMrDZbLRs2RJjDPPnz+fUqVPMmzfP2qYtLy/PmjQSmLem83e/btu2zdrqrjTbt28vsYh1t27dAHjttdfIzs62usNLk5+fT15eHj/++ON5tZT6u4ZLq3N57/cP1o+MjLTuHTBgwAX/9zPGMGjQoKDz1q1bX1BZqvravHkzQFjReL0T+OKr4uM6OgFtivIcLyNPaQ6KyEkR+RIQfK2Iw4HDInJMRPYDx85ViAZ8Sqla6cUXXwxqscvLywvqhn3llVcAX/eNPy3wX+wLFy600j0ej1VWVFSUde4/hg4dyunTp4PS/OtzBeat6fzvMXr0aF555ZWLNq7OH1z5g+bzERic+5VnB5WDBw8iIuTn5wO+2d3ge+eVK1dW6L/fpk2brM8iwrfffnvBZanK4fF4GDt2LO3bt6dfv364XC7mzZtHt27dSEpKYsSIEdafAfh5p6DiZQBBe3MDBigE/gr8GpgRcM0O/AisAPYBrwJjy1Hd4lODz9maVxoN+JRSSp0Xf1fsf/7zn1KvJyUllVjEesuWLQCMGzeOqKiooPF6xYWGhjJq1CgWL17Mn/70p3LX66GHHgJg/fr1gG+GNmBNsvEfs2bNAoInorRp0wZjjDXRxM8YQ48ePay0gwcPlpjAUrws/3qR/n9AnDx5EmMMixcvxhhDgwYNAIKW9yn+zMDj9ttvL/d3oMpn3759TJgwgbS0NGJjY3n33XcZPnw4W7ZsITU1lSuvvJJ//OMfAPzmN79h2rRpwM+TOPx/xvxLPv3qV9YmG22BO4CBwG+ApwBjjGkC7MfXBTsZSMS32+U44Ad8weD5eA9oYYxpZIxpBTQ65x2B/yKtTkeXLl1EKaVU9eD768Jn6tSpAkhWVpbMnj076JqISGFhoQCSkJAgIiJZWVkSEhIiTqdTRETeffddAWTMmDHWPd26dZN3331X2rZtK2FhYSIi0qtXLwHk1VdfPa962mw267P/8JfhPy/+ubR7/Z9//etfl7hn586dIiLy0EMPBf3053E4HNbn+vXrB12rX7++3HLLLdZ3sHPnzhLPDry/+PerKubgwYNy+eWXW+fPP/+8PP300/Lpp59Kr169pEOHDhIfHy/333+/iIjExcVJQUGBAGKMsX7GxcXJ3LlzRUQkPz9fAA++mbr+z4nAyKLzwUBfIK8oj7coz3wgGl+roBfYCczGNx4Q4DsgTX7uJfACIwOuCeAGXMBGOUtcpS18StUxSe3bE2MMNmOIMYaks4ylUiqQv8Vp6tSpXHvttdbg9cBrxhh+/etf8/rrr/PNN99gjCE6Ohq73W5N4hg+fDijR49m/vz51j2pqaklxvV99tlnXHHFFdx///2sXr263PUMXDbD7/777y91HN7VV19d7nL9QkNDrUH//q772bNnB7XUnau7e8WKFYBv+Rp/WYH1zszMBErfG7ouiI2NpVGjshutznUdoGfPnjzxxBOlXvOP1QTf5Au3281dd93F7Nmz2blzJ1OmTAla8skYg4iQmZlJZGQkXq+XO++80xozWvQzE9925X8HxorIbhFZAvhnJ3mAL8U3OcOGrzv3lIhkiUhIUXpHEZkoImEAIhIvItb/GEV5lhSdXi++SSLxQChw1sVENeBTqg5Jat+eY7t38wGQD3wAHNu9W4M+dU7FWwv8XVr+JS0Cj61bt3LvvfdaO6CICC6Xi8suu8wqb8GCBUH35Ofnk5CQUGJtxW+++QYR4aabbipXPf3LZzRs6FvGzL+cS/E6+l155ZUV+2KApUuX4vV6uf766yttrKZ/B5e6GvBVhpSUFFauXFnu/FlZWTRp0oTCwkIWLlxopV9zzTUsXrwYICi9d+/evPXWW3g8Ho4fPw4QBWzGtxbfbcYYuzGmIXBtUTpAd2NMK2OMDbgN2FCBV9xZNAHkMHBAREoONAygAZ9Sdch3u3ezCLgeCCn6uagoXanawD/R48SJEwDWAtGdOnUCfLOEKxrkFRQUsGvXLgAmTZrEl1/6llMbN26cNT4wUPH9n8E3Lgzgvvvu4+DBgwBnXeqmLvDPcjfGkJ2dDcCMGTOCxjr6x9KBb3mgmGLjHf15Ro8ejdvtZtu2bRhjmDFjhlVWq1atSEtLCyoL4Omnn+bXv/4111xzDe3atbPSX3rpJV5++WU6duzI0aNHrfRhw4bRqVMnkpKSuOGGGwCOiEgG8D6wA0gF1gF/KEoH2IKvy/Zr4GBR3gsiIlH+1kIRaVueGyp8AAOAvfgGJP6xlOuPALuLvoC1QMtzlalj+JSqfAakAEQCjgIQo2OEVC1is9kEkCuuuEJERMLDw4PG89ntdhHxjY+79957rfsoxxg+/3i7wMOfv7R0f10A+fe//x00ps8/Hiwwv7+s4u9S2z3wwAMCyJ49e2Tz5s0CSMOGDcUYI88995yIiIwZM0aMMSLi++4MyLqi31/3gTQGia1Xz8rjcDjkqquusp5RVlmVBUiRs8dKfYCPzpbnYh5GKtj8bIyxA98ANwFH8EWvd4jI7oA81+Prt841xiQDfUTktrOV27VrV0lJSalQ3ZRSwWKM4QN8LXt+n+AbdHK6krqilFLqfDVp0oTc3FxOnz4N+HbGCQ8P56effiqRV0QwxhCLbyXjpFLKExFCQkLo1KkTW7duZceOHSQllcxZ0RgokDFmq4h0Pcv1PsDvRWRQWXkupspoP+4O7BeRb0WkAFgMDAnMICKfiIh/QZtNwGUopX5x8YmJjMIX5BUW/RxVlF4dNLzkkqAJJQ0vuaSqq6TKqaylRS6mpk2blrpEyvlISEiwxsup6sUfjJXSUgb4hqXk+/PiW6zOUHoQ519fsayyfgki8mlVBXtQOQFfMyBwo8kjRWlluRffooMlGGPGGWNSjDEpRQMglVKVKDUtjUaJiQwFwvC17DVKTCT1LGui/VIaXnIJjszMoAkljsxMDfpqmMcff/wXe9YPP/xAQkICIsKRI0fo3LnzeZexf//+oAV2VdUYNmwYZ86cYf/+/Wzbtg2Xy0VISAgAXbp0AXwznx9++GHrnkKgW9HnLvhmPzjByuNwOKw9s/07vZRVVp1Q0T5hfGvMvB5w/ltgdhl5/wdfC1/YucrVMXxK1S31isbjBI4vXAdSrw6MX6oNKGNMm4iI0+kMuj5+/PgS9wASGhoadL5jxw4REYmPjw9Kv+eee6z7A9e/ExE5efKkAPLII4+UOC9evn+smP9wOByyePHioDT/moCBY/EIWI/Nf0ybNs2qQ2hoqHTr1u0iftu1U1xcXNA4y4YNG8rMmTODvuvY2FgRCR7DN6PYnyN/nlCH4+c/WyARYWGlllVZOMcYvqo+KiPg6wmsCjifDEwuJd+N+GalNCpPuRrwKVW36ISSmg2QOXPmCCB///vfgwK+5557TnJzc0UkeBJCYGCXkJAggERERFjXwsLCZMeOHUEBnKPoL/HA+/0B2Jo1a4LSRUSio6OD8o8aNUpERKZPn16iPv48/kkd06dPF0AWL15s5Xv99dflkUcesZ5Z/Hnql9MpMVHqFf2OqAfSKTHRutYgNlYaFwWEBUU/G4OEFk3EuRjqQsDnAL4FWuFb+C8VaF8sz6+AA8AV5S1XAz6l6hZt4bs4zvaXYmUqLQjzpyUmJpaaHvjZ3xLnDwz91zp06FDi3sDgqk+fPkGtNiIiI0eODHqGf4ePwODwz3/+c6n1Le1o1qyZFfDZ7fagazk5OUGtf+np6WKMkVatWlnlB7YsTpgwQUREFixYUKK1cPHixRflv01ddLbfJ/379r0oz6zuAV+Fx/CJiBuYCKzC14L3toikGWOeMsYMLso2Hd+ChO8YY7YbYz6s6HOVUrVLaGxsqRNKQmNjq7ReNVlVLLRdfG0zgN27dxMeHo6IcEkZYzLj4uIAcDqdQen+3SeK/+Xl98knn+D1etm3bx8ATz75JO+88w6A9Sz/NRGhWbNmiAhPP/00Tz75pFVO8XL9C0O3aNGC9PT0MvMFLghtjKFJkyYl3s3/7rGxscyZMweAO++8E6fTiYiUOntUVUwWvo1qA/UqSt+4du0vX6HqoKojzrIObeFTqu5pEBsb1BrVoJLH2NQ1v2SrKcW6Rf2H/zw+Pj7oWvHPZZXhH2fXuHFjERHZsWOHJCUliYhvbKC/RbBLly4CyJEjR0REglrV/Pr06RNUftu2ba3WtZCQELnuuuvKbOXzt+L17t3buj8qKkpsNpt1zf/3VvEWvhdffFFERG6++eagd58yZUpQfbSFr/KU9Wf/Ei7eMBFqewufUkpVluM//cRpEbwinBbheClrcKnyO1srx8U0fvz4oPOwsDC+++67C1o2pVu3brRt25aMjAyMMXTq1IkdO3YA4HK5iIiIwBjD1q1biYqKolkz3yIR/nVc/bMzAT799NOg5Vu2b9/O1KlTASgsLGT9+vXWVmyBJKBFr379+tbnIUOG4PV6rVbItWW0HPlb/fz7rqoL8++FC+kQH4/dZqNDfDz/DtjmrLiyegyuA6J/kdpWPxrwKaVULRVNyY06N3Bx/sILDIpefvnloK7PvLy8Urtk/Z8dResuGnyLgzuKNqr359u7d2/QvWV18/qX4ICfty7bvHmzlVY8v9PppH79+uzZswcRITExEbfbDUC/fv0QEdLT0xk7diwej6fE+y5YsADwdeXabDZiYmLKW6TgOQAAIABJREFU/X3ZbDamT58O/LxUiCrbvxcu5Ilx45h16BD/P3tnHt5Ulf7x77k3SZN0pZSCBdpC2VugrMMiIhYFBRUFRwEXEEFRBNSxozD+EFdwGR3ZFBdcgEHFEdxgkOowVFDAEdQiCliqUJC1UGjplu/vj/QecpsUCl2StufzPPdpcnLOue9NmuSb97znfU+TmJOVhekTJpQr+g4dO4ajmiZTUA0H0AvuNCG9U1JqzvAAQgk+hUKhqKMEeqJtALAIgUaAKc6wUWn7hSKEQFZWFho1anTOvq+//jratWsHIQS2b9+OYcOGYeTIkVizZg2EEIiJicGHH5Zf7rRFixYgib59+56XjW+88Qby8vIghMB3330HAOjevdwiDfWeJ6dPx+t5eaY64K/n5eHJ6dPLHVNQUoLeKSkIBZADd1HbzikpWL12bY3YHGhUurRadaFKqykUCkXl6ZyYiD3btyMXbs9efIAk2jaornJ/otRLWBZd16HrOgoLCyGEgMPh8Evi5d27dyM8PBxRUVEYNGgQ1qxZ49NehRtd03CahNWjrQiAXQiUlHp8/c25Sqv5G+XhUygUijrMtowMU1xkIIk9oObjDEtKSlBYWFhNs1ecBQsWoFGjRhBCYM2aNbjmmmvOPage0z421md4QvvYWH+YUytRgk+hUCgUfuNC4ww1TYOmub/CPDdiNG7cWN72rLGbmpoq24xxnnTo0MHUf/LkyZW4qnPz3HPPmeIJV65cWa3nqy46Jyaa6l9XV8qf6U8+iXFOpyk8YZzTielPPlkt56uLKMGnUCgUCr9xCvAZZ3jqHOPi4+O9lkB37NiBQ4cOSUEXGhoq+zz33HPlzrVjxw789NNPeOSRR0ASQUFBmDNnzgVeUf2hJvM8jhw9Gk8uXIh74+JgFwL3xsXhyYULMXL06Co/V11FxfApFAqFwq9YhEAwIOMMTwEoPsd3044dO9C+fXssWrQIY8eOBQDExcUhKysLffr0wYYNG5Cfnw+73Q673Y6CggKQlB48l8slY/g6duxo2s1rEKjfj4FCdcVf1lZUDJ/ignnsscd85q0SQuC666477/nsdjtCQkKqwjSFQqGoMopL4wuNOMNziT0AaNeuHQDg9ttvB+Cu0PHbb78BOJMPz263A4DPJVxPzlXNQ+Ebf+V5VFwYSvApFAqFotZCEg0bNsSkSZOkSDOEXkV59dVXAbiXiQG397BPnz5VamddpCbzPCoqjxJ8tQBN0+QyxNatW02PxcbGysc8s7h36dLFFIBclqioKAgh8OOPP1a7/QqFIjAJ0nVTwH2QrvvbpPMiKioKgDuX3jPPPAMAF1TNIzk5GZ07d0ZWVhaEEGjfvr3PJV6FmdqQ51FxBiX4agGXXXaZjD25+OIzDvR58+bh999/x+LFi3Ho0CEUFRWhbdu2WLJkCbZu3YoZM2aApFfS0IiICBw5cgS7du1CUlJSTV+OQqEIAIJ0HZEulyngPtLlqjWiL65pUxQePgwB4NZhwxDXtKlXFQ6DvLw8U3UPzz5GDr6tW7ealnONihuK8tmWkYHoDh1kNYthAKIDLM+j4gxq00YA89hjj0nRBgC9e/fG119/DQAYNmwYfvrpJ/zyyy/ywys6OhpHjx5FmzZtsGPHDtluYAQuA0B2dras76hQKOoftTngPq5pUxRmZ2Mp3DFj6XB7lmwxMcjat8+/xinqLWrThiKgMJY7XnnlFT9bolAo/EltDrjPKRV7nmW2lpa2KxQK3yjBVwsYNGgQAOCbb74x7bKdMmUKSOLdd9/F8ePHcejQIbRq1QrTp08HSTz66KMAgBUrVsgxTqcTAwYMwMyZMzFv3rwavQ6FQhE41OaA+9osVhUKf6EEXy3g888/l3Uh169fL9snTpyI5s2b46abbkJERASsVit27NiB0aNHo1OnTpg5c6bPFC5ffPEFunTpgkmTJuHdd9+t6ctRKBQBwGlN8xlwf/ocKUwCgdosVhUKf6Fi+BQKhaKeEqTrsLtcMuHxaU1DQUmJv806JyqGTxGIBHoMn8XfBigUCoXCP9QGceeLrH37ENe0KYZlZ0uxGqHEnkJxVpTgUygUCkWtQ4k7heL8CPxgDYVCoVDUGZw2mynZs9MjYXx9Jz09HbGxsf42Q1FHUYJPoVAoFDWC02ZDeFGRKdlzeFGREn2lbNu2Db///ru/zajz3HHHHejUqZPPx863Usull16KoUOHVoVZ1Y4SfAqFQlEFRERE4O233wbgToKelZV13nO8+OKLdTpHprWoyGf+PGtRkV/tChSmTJkCwC06goODERwcLMtjJicn+9k6RW1HCb4qxLN2raZpSE8vmzigcpxrzqVLl0KvJWWRFIpAxuVyVaq01sGDBxEXF3fe41asWIGPP/74gs8b6Kj8eWfnH//4BwB3ybcOHTogPz8fubm5+OCDD7Bt2zYsW7bMzxbWDi666CI4nU7Y7XbcfPPNAIDbb78dNpsNISEhpu/R//73vwgNDYXdbjeVLgWAIUOGIDg4GA6HA/3795ftl19+OWw2G8LCwpCZmVkzF1UFKMFXxRh1GHVdx7XXXlulc7tcLq9/SE9GjRqFklq6606h8Dfp6emw2WxISEiA0+nE5Zdf7vVhn56ejqCgIMTHxyMoKAjNmjXD4cOHAQCFhYW4/fbbAQAWiwU///wzAGDChAlwOBxwOBxo2bIlAOCRRx5BSEgInE4nIiMj8eOPPyI9PR3r16/HqlWr4HA4MHfuXPz0009o2rSp9PYsWLDAD89M1aHy51WcjIwMJCQkICQkBNdffz10XcecOXP8bVatID09HXl5ecjOzsby5cuxZcsWvP3229i2bRsOHDiAbI+KLH/+858xcuRInD59GjExMbL96aefxq+//orc3Fzk5uZix44dmDNnDhYvXoz169cjOzsb27dvN80V6CjBV03ExMTg1KlTAIDnn38emqZJ79/MmTMBAP/617+g67psHz16NAAgMTFRtgUFBeHkyZMA3B7E1atXIyQkBM2bN5fnioiIQHR0NObNmyfjD9q2bQtd1+V5PSt0tGrVSs5vs9lgt9tr5DlRKAKdoqIiPPzww5gxYwYOHDjg9WEPuIXdX//6VxQUFMDpdGLMmDEAgOLiYq8fXCtXrsSbb76JrVu3Ij8/H59//jkAt7fhxIkTyMvLw9VXX40xY8bg4osvRr9+/XDllVciPz8fkyZNwpVXXonp06fj1KlTWLt2LaZOnVqjz0dVU2S1+kz2XGS1+tUuRd1i3LhxcDgcaNq0KQoKCvC3v/0NcXFxaN++PUJCQnD55ZfLvgcPHsSLL74IAPj73/8u2z/44APs3LkTwcHBCA0NxdGjR/Htt9/igw8+QM+ePREVFYVmzZqVGwsYkBgeqUA7unXrxtqG++kkY6KjCYAAGFb69+mnnyZJ3nbbbRRCkCR1XWeTJk1Ikrm5udywYQMfeeQRCiF46NAhkmRQUBDbtm0r51+1ahUnTZokz2W0v/baa5w7d65sb9OmDQFww4YNzMzMlH2WLFlCAExLS+OhQ4cIgEFBQTXzBCkUAYamadR1nSTle5YkrVarvO95rF+/3me753jj2LFjB4cPH84+ffp4nXf58uWMjIxkUFAQrVYrGzZsSJLs378/hwwZIvsJIWi32+WhaRr3799f3U9LteKwWhkGUJR+PjqsVn+bFDCsWbNG/j/16NGDQgjm5uby448/JgC+//77frYw8HnhhRcYGhoqv0PDw8P50EMPsWXLlrLP9ddfz44dO5J0v8fy8/NJkr///rt8/rt168ZRo0Z5zT9s2DD269dP3u/atat8zwLYwgDQT+UdysNXxQghkH3wIADgKIDnS9sffvhhCCHw1ltvgaXVTUpKSrB582YAQEhICHr37o033ngDJNGoUSMIIVBQUIADBw6YzmF4GpYuXSp/8Y8bN87LFqvVit69eyM+Ph5CCKxZswZz5syBxWLBZZddhqioKJMLW6Gob7Rv397LK7d7924UeWwiuO2220AS8+fPN/X77rvvcPXVV8sx1lIvFUtDOs7GuHHjcNddd+H06dN45plnzhovuH//fuTn5yM/Px8lJSVo0qTJeV1joJFXWIjjJFwkjpPIKyz0t0kBw+WXXy5XfTIyMuBwOBAaGoqrr74anTt3xogRI/xtYsBz8OBBOBwOREVF4bPPPsPx48dx8uRJZGVlYefOncjLy8PatWtl/+joaNx3330AgAcffFC2Dx8+HCtXrpTfv1u2bEFGRgZGjBiBzZs34+jRo8jOzsYPP/xQsxdYCZTgq2LCALxZejsJQGePdk+lXR4kERISYuqbk5PjfZ6wMDz44IN4/fXXyxVtWpmamIXqg1WhMLFy5UoAwHvvvSfbDBFn8Oabb2LLli245JJLTO3JyckyTq+82KpbbrkFmzdvxs6dOwG4hSHgfi+2b98eADBv3jzZPzQ0FLm5Z7YwxMbGylAPAKr2dT2guLgYJHHq1CmcOnVKfg9s3brV36bVCh566CG4XC4EBQVh3LhxCA8PR0JCAm699VYkJiYiOjra9J353nvvYfHixbDb7aaUOA8//DCuuOIKxMfHw263Y8CAAfjjjz8wevRoXHzxxWjSpAnatWtXu5wm/nYxlnfU1iVdAbAQ4KTSZZ1jZZZ9ioqKOGXKFJK+l3SnTZtGAFyxYgVJMi0tjQsWLJDzr1q1iiT58MMPy3n/+c9/kqTXkq7nUq0QgsOGDeNbb71FAFy3bh2PHTumlnQV9R4A1HVdvp80TSMA2mw2+f4ICQlhWlqaaUnXZrMxJiaGADh27Fi5DEy639s7duwgSd5xxx0MCgqi3W5nQkICSff712Kx0OFwsHv37gwPDydJrl69Wi7fzpkzhzt27GDz5s1pt9tps9nYvn17/zxJCoXinCDAl3T9bkB5R20VfGEAvwDIUvHXHuA9ZWJ7IiIiSJIffPCB/HIBwNGjR5MkO3fubOp/yy23yPkNwWfcN+IByYoJPpKMj4+Xc1ssFoaEhFTvE6NQBDBCCAJg06ZNOXPmTPnemD17NgHwtttuI0nOnz9fCj4j7o+kFHxz5swxxdYqFIr6RaALPuG2MfDo3r07t2zZ4m8zzpu4pk1RmJ2NpXDnl0qHexeaLYAKe3///ffo1KkTcnJyEBkZiR49euCbb77xt1kKhV9o3rw59u7di9WrV2PQoEEQQkDXdRQXF6Nx48Y4WBqTCwDr169Hv3795OOAO2537NixeOONN0xZ+gP1s1WhUFQPQohvSXb3tx3loQRfNRDXtClysrORC3d+qYgAEnsAEBwcjLy8PACA3W5Hbm4uLBaLn61SKBQKhaL2EuiCT33LVwOBJO58YeQHVCgUCoVCUT9Qu3SrGSNpskKhUCgUCoW/UIKvkmRkZJiqaCQnJ8vKFkII9O3bFwMHDjTV2FUoFApF/UYIgfDw8Ar3v/32200xogrF+aLURyUZM2YMLBaL3AVj1Lp0OBwgiSeffBJpaWl46aWXQNKU8FGhUCgUtZ/OiYkIFwKaEAgXAp0TE/1tkkLhhRJ8lWTYsGEoKipCSEgIbrjhBvTu3RsA8PjjjwMAZs+eDV3Xce+99wIALrvsMr/ZqlAoFPWNf//733KFRQiB0NBQCCFk3dRu3bpJz9mnn35q6hsREQEAuOiii0ztZef9fvt2EEABgFcBfL99u6leOQAUFBR4zWHQo0cP02PTpk3zal+0aFH1P1mKOo0SfJVk+vTpSEtLQ1xcHJYvX44GDRoAAOLj4/1rmEKhUChw4403AjiTc/bll18ut+/QoUNNfZ9//nk888wzOHDggFeJvcGDB8PpdCIMQCqAXABfALgFgIC7utJ3332HoqIijBkzBo0bNwbgLv2Vmpoqz7l7925s2bIFs2fPBumutPT0008DcJfzatmyZb1I8XPv3Xcj0mKBJgQiLRbce/fd/jap7uHvRIDlHTWVeLlThw6mQt6dOnQ4r/FLlixhZmYmSbJfv34ykbKRIDkqKooA+NJLL5F0V844F7NmzfJZtFmhUJiZNHEiG+g6BcAGus5JEyf62yRFgDFs2DCZTNuodAKAzz//PEmya9euMmE2AA4fPtw0vmHDhj4TagPwOq700QZAVmSx2+2m8WFhYbzssst8jlm0aJHpvJdcckmdTew9aeJENiktWlBY+rcJUOvezwjwxMv12sPXOTERB7dvxwq4XfErABzcvv284i8WLVqEFi1aQAiB9evXY9y4cabHg4OD4XA4MHnyZAghMHDgwHPO+emnn2L16tXndzEKRT3j3rvvxvIFC/BBSQkKAHxQUoLlCxYoz4DCxIcffoj58+fD6XRi9+7d0HUdAHD69GkAMNUuPl+uueYahMHt2SOAz0rbI2Gun77vLKm6XC4X4NHXOOoTSxYuxFIAAwBYS/8uLW1XVCFVoRoBDAbwM4BdAB7y8XgQgHdLH/8GQPy55qwJD59nGTTj+KLU01cZEhISZPklp9PJdu3a8Z133qHVaqUQgrqu8+WXXyZJDhkyhEII2X7kyBH5C08IwcGDB1fFpSoUdYpQh6Pc928Dj7JnCkVqaio3bdpE8syKCwBGR0eTPOOpK3ubJF977TWfJfY8+7ZNSGATgO1KvVMWj3bS7WFcvXo1w8PDCYAHDx6UtdDDwsKYkZFBALJO8q5du3j55ZfLc7Ru3dqnbXUJowa953u5sHTlrTaBAPfwVYXY0wHsBtASgA3ANgAdyvS5G8DLpbdvAvDuueatCcFXHf9kjz32GIUQzMrK4vbt2wmA7dq1o8Vi4aJFi0iS06ZNo8VicdsgBD/99FOS5LZt20i6l4YjIyMrd3EKRR0l1OFgE4BaHfmSUFQfDofDa6m0a9eu8ke550GSn3zyiaktPDycJBkdHe3V16ir7HkIgEE+lmcXL17M06dPe7WHhYWRJPv06WNqt9lsJMnu3bv7tLOu0UDX68SPt/og+HoD+LfH/YcBPFymz78B9C69bQFwGKVl3co7aquHLykpic2bN5f3o6Oj2bJlS+mxMw7jjRsREUGLxcJevXrxm2++IakEn0JxNoz3baLy8CnOwhVXXEEA3LVrF0ly2bJlnDt3boXGHjp0qNLnr6virDpQMXw1c1RFDF9TAL973N9b2uazD8liAMcBNCw7kRBighBiixBiy6FDh6rAtLMT36EDRgH4EkBR6d9Rpe1ViRGj4XK55OH+3wCOHTuG1NRU7Nu3D3/605+wadOmKj23QlHXyAVwMYDpAMbB+/07esIE/xmnCBh2794NAEhISADg3q17zz33YMSIEaYUKEbuVCEELBYLhBBo3rw5hBCy5vjmzZshhMD+/fvx8MMPm8ZPmjQJAPDoo4/KtqCgID9cce1lzvz5GDFxIobrOoIADNd1jJg4EXNKd0QrqojKKkYAIwC85nH/FgBzy/T5EUAzj/u7AUSdbd7asku3LI8//jiFENy7dy9//vlnuaSraRqHDBlCkiwuLubs2bNJku+8844cq2kaZ82axWuuuYYhISGVskOhqKt4euaXlnr6tNL3b23zCCiqj40bN8plUE3T2K9fP2ZnZ8slVpKMjY2lEIIkZT8DAOzTp4/Pfg888ABJ8qqrrjLF/zVp0oSke+UGysNX70CAe/iqQvDV2iXd6sLXpo1ly5bJTRsAGB8fT5IMCgqSy7xRUVEsLi7mli1bqGma2rShUPjAiOEru/wT6nD42zRFgJGTk8O+ffvSZrPJmDxDBHoepFuw9evXT45t1qyZFIAA2K1bN5OI9DU+OzubJPnFF18owVcPCXTBZ6kCJ+FmAK2FEC0A7IN7U8aoMn0+AnAbgI1wewS/KH1y6iS7du3y2W4kAPXESA3gSbdu3VBSUlLldikUdYETeXkIczoxLD8fuQBCAdDhwInS5TeFwiA8PBzp6ekAgJYtWyIzMxMAUN7XT8OGZyKN3n//ffTu3RsrVqwAAKSlpeF///vfWcdbrdYqs12hqGoqHcNHd0zeJLi9eD8BeI9khhDiMSHENaXdXgfQUAixC8D9AB6q7HkVCkX95UReHo6TcJE4Tiqxp/BiypQp+Mtf/iLv79u3D5rm/sobNGgQAGD//v0YP368z/G9evUCAFx//fXQNA3h4eEYMGAAALd4BICioiKMHj1ajuncuTMAYPjw4VV8NQpF5amSxMskPyPZhmQCySdL2/6P5Eelt0+TvIFkK5I9Sf5aFeetDKrYtUKhUNRd9u3bh+eff15upCgsLMTSpUsxcuRIrFmzBkIIxMTE4MMPPyx3jhYtWoAk+vbtK9tmzJiBzMxMWSf3X//6l2zPzs6GEAKnTp2q9utTKM4XEagrq927d+eWLVuqZW6jwsZSuHf7pcO9Bh3doQO2ZWRUyzkVCoVCoVDUXYQQ35Ls7m87yqNellbbUyr2ypZx2bN9u1/tUtQthBD4+uuv5e2qZPjw4Zg7d+45+8XHx0MIAU3TIIRAZGQkCgsLAQBhYWG4+uqrz/vcl1xyiSnWSaGoDsKcTtMqTJjT6W+TFIpaTb0UfEYeL08uLm1XKGoDn3/+Of79739XqG+7du3gcrlQUFCAEydOYOrUqdVsnUJROcKcTgTn55vqnAfn5yvRp1BUgnop+ELhXsb1JL20XaG4EOx2OzRNg6Zp+NOf/lRuv5KSEjRp0kT2vfLKK+Vjbdu2le0xMTEAgN69e0PXdWiaBofDgd9++w2pqanIzc3FJ598Ak3TsHjxYixevBg2mw2apsFiseCVV17xOvfRo0dBEnFxcV6PTZgwQZ47PDwcRuLzGTNmyPPruo5ffvnFNG7UqFHQdV16MhWKqkDk5/tchRH5+X61S6Go1fg7L0x5R3Xm4evUoYPPPF6VTbqsqL9s2bKFJLl3714KIfjNN98QADdu3EiSMifXNddcQ4vFwry8PH755ZcEwDVr1vDmm2+mpmnMysoyzWeU2yPdecE6d+5MkgwNDeXQoUPlY+XVao6Li5Nl/QDQ6XTKMcYc+/fvJwC+9tprJMmQkBAmJSXxyJEjBMD/+7//I0lu376dubm5svTfsGHDqOs6v/vuu6p9MhX1nuqoc65QVDcI8Dx89dLDty0jA9EdOmAYgCAAw6A2bCgqx/Dhw6FpGpo3bw6S5e7827hxI/70pz/B4XDg0ksvRXBwMBYuXIg1a9agZ8+eiI2NBeDOxQgA77zzDiwWCzRNw969e5GVleU15+7du1FcXIzbb78dmqbh6aefRnFxsXzcWNLNyclBSUkJBg8ebBr/9ttvQ9d1jBs3DgAwcuRI7Ny5E2+++SY0TcPMmTMBAO3bt0dISAgAt7fwo48+wg8//IDk5ORKPnsKhRm1CqNQVD31UvABbtHnmcfroosuMgUIDx440N8mKmoJkyZNwt69e7Fnzx64XC7ouo7c3KqJCJ07dy6eeuopuFwu9OvXz2dCbmMThq9azZ6Eh4cjKSmpSuo122w2kMR7771XqXnuu+8+nxtahBC49NJLz3s+i8Xis46pkZrDM+7R2Mjy6quvnvd5FNULHQ6fdc7pcPjVLoWiNlNvBZ8ngwcOxLa0NFOA8La0NCX6FBXiwIED0HUdsbGxeOWVV85aJaVPnz7YtGkT8vPz8dVXX+HUqVOYOHEiBg8ejE2bNuG3334DAHz77bdyTK9evXD8+HFs3LhRttlsNuTk5ABwe940TcPQoUMBuOMEn3nmGa9zl5SUYMeOHWjSpImp/dZbb0VJSQnefvttAMCyZcvQpk0bjBkzBi6XCzNmzAAA/PLLLzh58iQAICQkBAsWLMCjjz6KOXPmnPdz5i9uuukmAMDJkyfLrZag8D8n8vJwyuEwrcKcUtVUFIrK4e815fKOmqyl61mM3Ti+KC3GrlCci4MHD9JqtRIAg4KCqOs677nnHp8xfMXFxWzcuLGsn+xZK7l169Yy3q5p06YkyW7duplqdoaGhpIkH3jgAdnWpEkTLlu2zBSrZ9Rq9ozhE0LQ6XRy7969JM1xgOPHj5d9wsLCePDgQZLk//3f/8m6zpqmcdeuXTKGjySfeuopAuA777xzQc/d1KlTveqSbtiwgQDYv39/kmRkZKR8zLO4fcuWLb3qmeq6TpvNRpIMDg4mAG7evJkAaLFYZL/27dvLcQsXLpSvkXEEBwfL83g+rwA4f/58kuS8efNM9lQXxnmFENV+rkAgNiaGYaXxemEAY2Ni/G2SQlEhEOAxfH43oLyjJgWfChD2jadQsNls3LZtW5XOX1VzxsXFsV27dlVgUeAyYsQIn8XYAbBt27bydm3DEHzG+x2ALHTfv39/PvHEE1JkGZtLYmJiOH/+fALg5MmTSZJvv/02yTOCz263EwAzMjLkvNHR0QQgxXHHjh1Ngs+YwxCIxiYYAGzQoAFJmkRjea9JZmZmlT0/991333m/rlV5/pomNibG54a62ib6QkJC/G2Cwg8owVcLBJ/y8PnG84smJCSECQkJfrSmfOqT4PM8PvvsM5+Cb+bMmQTAsLAwf5pcIQzBZ+Dpeevfvz9jYmK8/g+FEGzWrJlPIaTruhxv7Hgmzwg+q9VKp9NJAMzNzTUJPs+xANirVy85dv/+/STJu+++W5637OtR1tvoeV/XdZMtnseDDz5IknQ4HGedIyYmhkOGDDG1GV5QX+evjajPYkVtJtAFn4rhA9A7JcVngHDvlBS/2hVItGvXDkePHgWAcnO+/ec//zHlo0tNTQUApKSkyFxuDRo0QH5pLi2jEkVMTAy6du0qzxUfH4/27dsDcMenGWM988e1atUKQghYLBaZM64+YOwCBoAhQ4Z4PZ6amooZM2YgMjISx48fr2nzAornn3/eq23y5MnIy8uDEELuOAaAjh07oqSkBBs2bJDPrxGvWB4jRowA4P7R3LhxYwDA9ddfL8c/++yzIIm1a9eipKQEs2fPlmNDQ0Nlv+eeew4AkJ+fj/Hjx4Mk5s2b53WOffv24dNPP8WQZxK4AAAgAElEQVQVV1wBkkhISMC6detMNu3fv1/OWxupK0nxjY1IW7duRXh4OBwOB+x2e4Wq4ygU1Ya/FWd5R016+EhyUEqKKW5kUEpKjZ4/EEHpr+q8vDza7XbeeuutJMvP+eZwOJiUlCTH/PDDD5w/fz5tNhtzcnJIkg0aNDB5JTZu3MinnnrKywOycuVKjh07luHh4SwuLmZBQQFtNhsnT57Mxx57jEIIZmVlcfv27QRQbzx8BsbyJMp4+AB3TF9twfDw9ezZkyS9lnSffvpp6YUz8gJWZEm3S5cuBMAnnnhCzhsdHU2SdDqdvPPOO2X7woULTZ7EUaNGEYD8XwbAhg0bknT/7xuxdJ6vSePGjb08a8brYRwtWrSQ7fn5+STJoKAgL2+eruucNWuW1znef//9s3oTrVZrVbwkfqWuePiM12Xo0KEcOHAgSbKgoID79u3zp1mKagYB7uHzuwHlHTUt+BTeAGeC1XVdZ15eHnft2iXbjcPzS8cI9jdITk429QfAuLg42d9zU8Pq1as5e/ZsGQhvfIl6ju3Tpw+TkpLYvHlzeY7o6Oh6Jfjy8/PPKvjKCo9ApjKbNuLj472u2XPTRkJCghR0noLPE+PxjIwMLzs8BZ+vTRtnE3zG/c8++0zO4Xl07NiRuq7LzT4Ghs0AOGvWLNM5Fi1aRAC02+1MTU01XQMAOhyOyr0YAUBdieEzXrOXXnqJFouF/fv357Jly/xslaK6UYJPCb5ai/GhlZWVRV3X2alTJ+lRK69/WcHXuXNnxpTzYe0p+Jo3b85OnTqxadOm7NSpE0m3kOvRo4fXuNom+NavX8+goCB5f8iQIezfvz+vv/56ucHAuJ4//viDrVu3ZnBwMB0OBx9++GGSZ8SNEILh4eEmEeIp+IyNDfVlR2dNUN7/+6xZs8oVfMYOYZJyA4mxsxkAU1NTCZh3Dg8ZMsR0zs6dO3t5dg1x17VrV+bn5zM5OVm21wXBR9aNXbqer9m3337LW265hXa7nePHj/ejVYrqRgk+JfhqLZ4fWkb6jdzcXGqaJr+ciouLOXv2bJK+l3TnzJlDAFy3bh1Jd8mw5cuXy/kNwffSSy9R0zQC7lJjJDl27FiZCoQkP/30U65bt46PP/44hRDcu3cvf/7554Bf0i1P8GmaxuPHj5Mk9+zZQ5Ls3bs377rrLtlmtVr5xx9/eHmzDO9RWcFHUpZ1q4mUIfWBs3lMfS2vkmdeA8+jcePGckzZDSK+5srPz/farOPpaSx7lN30AYAzZ84kSd50003l2qqoeoznNz09nQUFBSTdHmFDoCvqJkrwKcFXayn7pWCz2ThgwAAuW7aMVqvVK+fbl19+yaCgILkEayw7DR482LQE/NBDD8n5DcFHkkIIGQ9o0LFjRzlO0zSZ781Y+tJ1nU6ns1YKvoYNG7Jp06a866675C5Qp9PJoKAg2u122u126rrOjz/+mOPGjWOrVq38dQmKSvDJJ5+YhJZnLN/zzz8v+xlLxQcPHiQAXnLJJbKfp3g3PHy+7gNn4g09xwFncguWXUZWVD3G83vHHXcwKCiIDoeDYWFh8oevom4S6ILPAoWiHNz/v2coKCiQt2+88Uav/pdeeilOnz7t1b5q1aoKze9yubz6fP/99z7H7tq1y2d7IBIUFGS6VuM5ys7Oxty5c7FkyRLExcXh+PHjIN11eK+88krTHCtWrIBDlZWqlQwZMgQHDx5Ez549sWfPHmRmZvp8LadMmYK7775b3v/f//4nbxt1jivCvn37ALiroXjuNN6xYwcA9y57X+9fRdVhvN9fffVVVbpPETCotCyKesnggQNrrHZyYmIiioqKsHPnTpw4cQJff/01XC4XNm3ahPvvvx/r169HcXEx/vjjDyQnJ+Ovf/2rFL///Oc/q80uRfkIIfD3v//9rH10Xa/wfI0aNUJmZibsdjsAeP0wGjNmDIqKivDaa69JsVBcXCwfb9euXYXPZdQS1jT18a5QKM6gPhEU9Y6arp3sdDpx3XXXoUOHDmjevDmaNGmCkpISXHXVVbDb7YiMjMTll1+OuLg4fPLJJygqKoLT6YTdbsfUqVOrxSZF5fHlkfbFtddeK/PplcXwYBse6xtuuAGPPfbYOec0cmKeD0Zuy1tuueW8xyoUijqAv9eUyztUDJ+iuqgrub4U5TN27FhTnJqu6z43PBg7wo3NRZ6HEV/n2WakfCnbl6Ss4GEcxo71sn2NuY0dusb4sn3sdrts94z1M3ZpA+D69eu9YvgMwsLC5P3hw4erTRsKRTWDAI/hUx4+Rb2jrmTzV1w4JNGyZUvpYbv33nuhaRpIolGjRqa+8+fPB0msX78ehYWFmDt3rlx2NT5Ir7rqKuTl5WHXrl3ysbi4OEybNg0AMHv2bGRkZMg577//fhQVFcnxnnMZh1GRhiTuv/9+OTYnJ0f2ufjii0ES3377rexrYMSEAsA999yD06dPg6SpwohCoag/KMGnqHeEAkgv05Ze2q44N40aNDDFPzZq0MDfJp0XsbGxAIDdu3cDAN58800AwMiRIwEABw8eNPW/5557IIRAv379AAAvv/yy15yff/45gDMl/wB3nJ7RNzU1FR06dDivuL+qZPz48bDb7RBC4OTJk/jzn//sFzsUCoX/UIJPUe9QtZMvnEYNGsCSk2OKf7Tk5ASc6AsLCzPd9/R8ld3MYLVay52nRYsWIIkvvvhCznHq1Kly+3t66Coa41cTGJ5H43j33Xf9bZJCoahhlOBTmMjJycH8+fOrZe6PPvoIs2bNuqCxZe3Kzs4uNxD+XKxeuxadU1IwDEAQgGEAOqekYPXatRc0X32iMCcHSwEMAGAt/bu0tD2QuPfeewEAc+fOxdatW03ia8+ePQgXQnrixo8dC+DMjujGjRvLvsePHwcADBgwAEOHDvU6j+ElvOyyywAATz31FABgwYIFmDBhAu666y5px+7du1FSUlJ1F6lQKBTng7+DCMs71KaNC2dQSoqpNNGglJQKj83MzGRiYmKF+7tcLpaUlFyImefF+doVCDisVtPr4KgDxe0F3DVOPTe8FJZeY6BRtpqFr00bLeGu1aqVs7Hiu+++82o3Eo17tpFkSEiIqc34fzVKq5XtfyHX8+ijj1bNk6PwYtLEiWyg6xQAG+g6J02c6G+TFLUMqE0bipqksilHHnroIezevRvJycl48MEH8eyzz6JHjx7o1KkTZsyYAcDtIWnbti1uvfVWJCUlYf369WjXrh3GjBmDNm3aYPTo0Vi7di369u2L1q1bY9OmTQDcsVKTJk0C4M47NnnyZPTp0wctW7bE8uXLAQAnT55ESkoKunbtio4dO2LlypU+7dqzZw+SkpLkvNdffz0GDx6M1q1bIzU1VV7P66+/jjZt2qBnz54YP368PH9147TZEF5UZHodwouK4LTZauT81UVtin8sLi72ucTaCmeU1264PZQh8P7xe//99yM5OdmrPTMzEyjTHwByc3NNbT/++CMAID8/32sORWBx7913Y/mCBfigpAQFAD4oKcHyBQtwr0cibIWi1uNvxVneoTx8F4avlCOTSj0Yo0aNOud4T0/av//9b44fP1568YYMGcJ169YxMzOTQghZFi0zM5O6rvP7779nSUkJu3btyrFjx9LlcnHFihW89tprSZKLFi3iPffcQ5K87bbbOGLECJaUlDAjI4MJCQkkyaKiIllf9tChQ0xISKDL5fLy8HneX7RoEVu0aMGcnBzm5+czNjaWv/32G/ft28e4uDgeOXKEhYWFvPjii+X5q5u6mvolKiKCTUqvpbD0bxOAURER/jatQgBglwDzUHqWI+zZsyfz8vIYGhoq24z61IaHr6CggKGhoWzWrJnfbK5rNNB1n+/XBrrub9MUtQgoD5+iJvGVcmQNABeAJUuWnHO8Z3b/NWvWYM2aNejSpQu6du2KHTt2YOfOnQDcPxR69+4t46BatGiBjh07QtM0JCYmIiUlBUIIdOzYEXv27PF5rmHDhkHTNHTo0AF//PGHnHfatGno1KkTBg4ciH379snHzkZKSgrCw8Nht9vRoUMHZGVlYdOmTejfvz8iIyNhtVpxww03nHMeT+69+25EWizQhECkxXJev/brauqXQ8eOoTgiwhT/WBwRgUPHjvnZsooRBuD5Mm3+9lB+9dVXcLlc+P3337F582Y899xzyM/Ph8vlgsvlMr1vT58+jQYNGiA6Ohq///67H62uW+SUlPh8v+aomEtFHULV0q1jGEtuA0rv3wXgV7h35zz//PNYv349fv31VzidTixcuBCdOnXCo48+it27d+PXX39Fw4YN5Vwk8fDDD+POO+80ncMQcFOmTMGLL76IPXv2yHJOgHsXpGd5J08R6YnnGJYucy1ZsgSHDh3Ct99+C6vVivj4eJ/1ec82l67r5Z6zosglHrg/+NNLSjBqwQIAwJwKbGop+zoA/hcWVUVtEXe+KLJaMaqoCEtR+rrCvUO7qMxO3TCnEyI/H7lwv2Z0OHAiL69abBo+fDh+++03AO73wcmTJ1FcXIyoqCgMGTLEVIt19uzZaNmyZa2qJV0biNB1pJeUeL1fI/yURkehqA6Uh6+OUTblyI1we/f69+uHPXv2oEuXLvj+++/x1FNP4dZbb5Xjtm/fjrVr12LRokXIzXX7oQYNGoQ33ngDFosFonRXo8ViQYsWLQAA//jHP2CxVO1vhuPHjyM6OhpWqxVffvklsrKyAAChoaHSrorSo0cPrFu3DseOHUNxcTE++OCDCo9dsnChz92oSxYurND4IqvVZ+qXssJCUbPkFRbiuNVq8lAet1qRV1go+4Q5nQjOzzfFXwbn5yPM6axyeyZNmoS9e/diz549cLlc0HUdp06dwq5du9C3b1+8++67pkTQERERyMrKwoEDB6rclvrM6AkTfL5fR0+Y4Fe7FIqqRAm+OsbqtWtxAMBlAGylfy1WK97717+Qnp4u62hedtllOHLkCE6cOAEAuOaaa+BwONCwYUP07dsXQgh8/vnnyM7ORklJCWJjY9GmTRuUlJQgOTkZgNtjWFxcjBEjRpiWbd955x3ccMMNEEKgRYsW+Pnnnyts/+jRo7FlyxZ07NgRb7/9tiwab9iVlJSEBx98sEJzNW3aFNOmTUPPnj3Rt29fxMfHIzw8HJqmoWXLlmcdW9klnooIC4V/yCssxHESLhLHSa/XROTn+xT7orTyRVVy4MAB6LqO2NhYvPLKKygpKUFWVhYKCwuxcuVKzJo1y/RDZ8qUKWjbti3i4uJw8uTJKrenvjJn/nyMmDgRw3UdQQCG6zpGTJxYIW++QlFr8HcQYXmH2rRx4cAjAN0I/D506BCTk5O5e/du+VizZs14/Phxzpgxg88++6zPOYzUFp7tRptR3zMoKIjBwcGyj+f9Bx54gACYnZ1d9RdaAXJzc0m6N4MMHTqU//rXvyiEYIsWLc46zjOI+5AK4q5X1GTqmYMHD9JqtRIAg4KCqOs6O3bsKN+3QgjefPPNJM1pWWJjY+l0OllQUFDlNikUigsDAb5pw+8GlHcowXfheAq0Bg0aEADbtGljygP24IMPMjk5mRs3bjS1G7tljTmM3GWJiYlcvHixqW+rVq04c+ZMU1v//v29BCAAvvnmmyTJ4OBgr0L0mzZtMs0RGxsrxxlfhgCYmppKkl7nfPrpp6WtnrnXWrRowQceeIBJSUleedAMwVe2qPz8+fPluY0jxGM3qsrNVfepqzusFQpF9aIEnxJ8NY4h1g4dOiQTzi5ZsoQHDhzgtddeK0XRtm3baLVaKYSQHr5PPvlEzvH6669L0fPFF1/I27169TIlkD2bh2/EiBGyX9euXSmEYGFhIU+dOkUAvOSSS5iQkEAhhBy/Zs0aaYPD4SBJNmnSRM4DgI0bNyZJRkdHe4nTU6dOcfz48bI9ISGBAJiTk8O5c+dKwZednU0AXLx4MUm318Sww7g+lYi15gh1OEyJqkNLX3t/2OEr9Yy/7FEoFLUDJfiU4KtxPL1Tuq4zKyuLt912m89s/wD4wAMPlDtHv379vCoWeB4bN270Kfg8+yQnJ5MkLRaL1/hGjRpx6tSp0pvXv39/kw3Tp08nSSnOjHZjidizXQjBpk2bmsaT7qWw6OhoU3uLFi04duxYn9dk9OnXr1/lXghFhQk0kRUo4lOhUNQeAl3wqU0bdRTjBS4uLkZsbCzeeusttGvXDiTRt2/fCs0hhMDXX38tKxYAwI4dO0z/QL169fI5Njg4GCTRrl07bN26VW7ciIyMNI0/ePAgXnjhBXz44Ydo1KgR1q1bB4fDIec5313AtvOoZGFUXij7pjDwTFGjqF5qcqNERTiRl2fa2FFdKVkUCoWiplCCrx6RmJgIANiwYYNss1qtePHFF+X9Tz/9VN4+duwYioqKEBwcDMAtALt16yYfHzVqlJyjqKjI5zl/+uknAECfPn3QqVMnHD16FF9//TUA4MUXX8Rbb72FefPmoX379ti3bx86dOhgyrv37LPPAgC6d+9umte436VLl3Ned3x8PA4ePIjjx49jQWkuPQB47LHHALjTzwDA/v37MX78+HPOp6h66mqiaoVCoQgY/O1iLO9QS7oXDnwEl7dt21YuWRpLqyS5ceNGU1H5sps2du3aRQCMiIjgsmXLTH01TSNJuSSLcjZtGBsjTp06xbCwMNPy6UMPPcQuXbqY2oYNGyZt8LVp44knnih304bn7lvjGgoLC6lpms9NGyNHjjS1N2zYUI417Ag0MjMzuWTJEn+bUaXU140S69evZ/PmzUmSw4YNM8WyVhdGHKtCoahaEOBLuoIeS1iBRPfu3bllyxZ/m6HwI0IILFu2DDfeeKO/TalWjDejplXM4f6f//wHzz33HD755JNqtqzmMJIdl62AcaoaK1wEAvPmzcOkSZNAEtdddx1WrlwpQw3K4+TJkwgJCamScyoUiqpDCPEtye7n7ukfKrWkK4SIFEJ8LoTYWfq3gY8+yUKIjUKIDCHE90KIuv3t7Sf+uWQJkuLjoWsakuLj8c8K1M1V+I89e/agbdu2uPXWW5GUlIRx48YhKSkJHTt2xLvvvgvALQQffPBBr/aHHnoI69evR3JyMl544QV/XkaVcSIvD6ccDlOi6rou9gB3ImXA/eNmxYoVIAld12VVG6NEoBACISEhEEKgb9++GDRokKx+o+s6vv/+ewBAgwYNZLvNZpPjJ06cKNtTU1P9c7EKhcK/VMY9COAZAA+V3n4IwGwffdoAaF16OwbAfgAR55pbLelWnKWLF7OF02na4djC6eTS0nQjisAjMzOTQghu3LiRy5cv58CBA1lcXMwDBw6wefPmzM7OLrf9yy+/5JAhQ/x9CfWWQSkpph28g1JSLnguz+XVYcOGEQAXLVrE/Px8AuCf//xnku4Qg6ioKJLkqlWrCIA//vgjSTIqKooREREkybS0NDm3ruu8+OKL5fjhw4eTJJ1Op1rSVSiqAQT4km5lN21cC+Ct0ttvwf3DvKyg/IXkztLb2QAOAmhUtp/iwnly+nS8npdn2uH4el4enpw+3c+WKc5GXFwcevXqhfT0dIwcORK6rqNx48bo378/Nm/eXG67wn8MHjgQ29LSTHV2t6WlYfDAgVUyvxACY8aMgd1uh81mw48//igfe/zxxwEATz31FAAgKSkJQggcPnwYp06dAgD89a9/lZ68kpIS/PTTT0hPTwcALF++HABw1113VYmtCoWidlFZwdeY5P7S2wcAND5bZyFET7hLvO6u5HkVHvz0228+dzj+9Ntv/jBHUUGM3c+K2sPGtDSf6WM2pqVV2TnefPNNCCFQWFiI7du3QwgBwL0su2vXLrns6/nLvbCwEHv27MGWLVvw2muvgSTCw8Plkm5lMWy47777TOmKrrzySsTHx1/QfJdeemmV2KZQKCrGOQWfEGKtEOJHH8e1nv1K3ZnlRgELIS4C8A6AsSR9RiULISYIIbYIIbYcOnToPC+lbnE+MXntY2ORXqatGYAQEpoQCBcCYU5ntdqruHD69euHd999FyUlJTh06BD++9//omfPnuW2h4aGIjdXJSzxB1WdPqZNmzZebWPGjAFJBAUFATiTW3LVqlVo1aoV7rjjDpSUlMiURRkZGXjyySexd+9eAED//v3x/fff4/jx4277LnZbbGx+euWVV87bTpZu8Fi7di2OHj0q2zds2ICsrKzznk+hUPiByqwHA/gZwEWlty8C8HM5/cIA/A/AiIrOXZ9j+M43Jq9s/+jSKgWBUrVA4U1mZiYTExNJki6Xi3/5y1+YmJjIpKQkLlu27KzthYWFHDBgADt16sS///3vfruG+kh1pI/xrGTjmZbFqFhjs9lMqYOEEIyIiDC1GfF9nm3GQfKslXY8Uy3169fPK+2S0a/s/J4pk1Ca0mjdunWmtpiYGJLk5s2bvc7vWVVHoagLIMBj+Cor+J6FedPGMz762ACkAZh6PnPXZ8GXGBfn80slMS6u3DFLFy9mYlwcNSHqbU4zhaK6GZSS4vPHVGU2bpwNXddNgu/IkSMkyY8//pj79+8nSbZp00YKMiPP5Ycffsht27YRAKdNm8bLLrvMtFHjvffeI0mTsDME5tNPP82MjAy5gcToR5JJSUmmeYzzGWiaJjeQvPfeewTAdevWyWsgyaFDhyrBp6iT1HXB17BUzO0EsBZAZGl7dwCvld6+GUARgK0eR/K55q5vgi82Jkbu/AsB+HYZwVYIUKtgUlZR2r/seKEEn0JRaapyl+658BR8YWFhsn3atGk+PXZlBRgAdunShcuWLZPevJYtWzI3N1c+bnzWtm/f3mvs4MGD5W3y3ILPl4fxxhtvJAC2b9/e1E8JPkVdI9AFX6U2bZA8QjKFZGuSA0keLW3fQvKO0tuLSVpJJnscWytz3rpGXNOmKMzOljv/PgLwNwD/9OiTDnesXkUILe3vSXppu0KhqByr16411dldvXZtjZzXbrfL20899RSCgoJAElOnTj3ruKKiItx4441IS0tDdHQ0fv31V4SGhspE30YSZ13XvcbmX0At41tuucX0JbNs2bLznkOhUFQ9qpZuAJCTne218+9NAI/A7Rr9EsA4pxPTn3yyQvPR4cCo0nHG+FGl7QqFom4QEREBAJgzZw4A4PTp0+Xuyn311VeRl5eHAwcOYNgwr+xZFSY6Otp031HmM0XTNLzzzjsoLCwEcKZOtc1mk3W1r7vuugs+v0KhuHCU4AsAytv59ysAuxC4Ny4OTy5ciJGjR1dovvpatUChCGRuuOEGmSPPOKKiouTt1q1bo3Xr1qbHS0pK5PiDBw/KdgD4448/TH0cDgcKCgoAwNQPAD788ENcffXVsqLHhbJo0SI5v91ulzuFjWv55ptvAABBQUEQQmDNmjUAgK+++kr2q8z5FQrFhaMEXwBwtiXYEpcLP+7ZU2GxZ3AiL8+07BRoYs9ms2G6SgytqId0794dx44dAwAcOXIE+fn5CA4Oxq5duzB9+nRkZmaCpMx3V1BQAIvFAgCYNWsW1pYuIbdt2xYjRowAALRo0QKkO/ce4F6K9TzHnj17AACfffaZFG2AO4b7P//5DwDghx9+MGKzvR4z2mNjY+VS7enTp03Lt4cPH0b37t19xXr7bDfmVigUNYPwfIMHEt27d+eWLVv8bUaNYMTwlS0cb4uJQda+ff41TqHwM+Hh4Thx4gR8fVYJIUASJ0+eRFhYmKnP5MmT8Y9//ANWqxVFRUXVYpvFYoGu6ygoKDB51IzHjPMaHrDDhw9LG416twUFBbj22mvx0Ucf4bbbbsNbb71lmsdms8HlcqG4uNg0VgiB4cOHY/ny5aZ2XzaWlJSYBJgx3uXymRK1zjJp0iQ4nU4888wzAIAePXogODhYiU9FlSCE+JZkd3/bUR7KwxcAZO3bB1tMjGkJti6JvYyMDGiaJr9kkpOTIYTAhAkT0LdvX69lLsBc7F3TNBX4rTgrcXFxIInc3FyQxCOPPII+ffoAwHlXmygsLMTJkycvyI7o6GiQxLZt21BcXIwpU6actb+xUcJIrvzWW29JEWt4784XTdNMnrTqEru1kU8++QQff/yxvL9582Yl9hT1BiX4AoSsfftMS7B1RewB7soBFotFfgEtWLBAPvbVV1/Jdl3X0axZMxw+fBgvv/wyPv74Y5BEp06dcOutt/rxChTViafYN3aMhoWFybYTJ07IvlOmTJHtnrtKT548aRr/2GOP4cYbb5R9PH9MGPFlRhwa4C4ZZrQFBQXh0Ucfhc1m83kuq9VqirErLCxEeBnP2s8//wwAaNWqldf1NmzYUNriaxcsSQghZO3bss/V/PnzAfiu0hEZGQmXyyXHTpgwAdOmTUP79u0BAGlpaVi6dKnXOH+RkpICXdehaRoaNGiA6dOnQ9M0HDhwAJmZmdA0DS+++CLy8/PRpEkTaJoGTdOQnJwMwF1f2GKxQNM02Gw2fP755wDcXs1GjRrJ/o888gjef/99ZGVlYceOHdA0DVOnTkV8fLx8bmbPni1tsdvt2Lp1q5wrJiYGuq5DCHHOXdEKRcDi77ww5R31LQ9fXeaJJ54gAAYHB3PEiBEkSQAcP3687BMVFUWLxUKSTE1N9crlJSqYg7A68UxunRgXV27lE8X58fbbb5M8U41hxowZBMD58+dz586dXpUhwsPDSZIOh0O2P/LII6b/lSuuuELOD488cUYC4nXr1jEtLY0AeMUVV3Dq1KkEwF69epEkW7VqRQAsKChgbm4uAbBTp05yfMvYWIaXns9WmoC57P9s2UoVUVFR8rH58+ebqlEYufZCQ0O95ilbacPz/TBixAjT9ZE0Vc4AwCFDhtBisQTce2r+/Pm02WzMyckhSTZo0ID9+/dns2bN2LhxY0ZGRrJVq1YkyW7dutFut8v8gVu2bGFOTg41TWN6ejpJcvDgwTJXoa7r8v9k8uTJ8lrj4uLYrohbx4gAACAASURBVF07aYPnfSEEJ02aRJJs3ry5rF6i6zqjo6NJkqNGjZKfUwpFWRDgefj8bkB5hxJ8dYu0tDR26NCBAGRZKEPwXXnllQTAnTt3kiT/8pe/eH2J+ZvzLXenqDiepcWAMyW7DIwKEKRbOGVlZZF0V5vw7JeRkcH4+HgpeNq2bSvHeM7lKcQ0TWNQUJAUfAZlRRMAhoSEyPFGNRtRKvhYJgGyISZ79Oghbejfvz9jYmJM5wkJCfFKnny25+ZCsVgs7Nq16wWPrw6Sk5Ol8DSe77i4OB45coRCCGqaxry8PJKk3W7n7bffbhr/wgsvmMYLIaQY03Wdd955p+wLgNu2bStX8P3www+m5/edd96RIlHXdT7wwAMkyS+//DLgPpsUgUOgCz61pKuodpYuXYqWLVsiIyMD/fr1My3Rvfzyy1i1ahXmzp0rl78eeOABAMDNN98MADh8+DCmTZtW84Z78OT06Xg9L8+UK/H1vDw8qXYaV4qOHTuipKQEGzZscP8CBdCoUaOzjjHi3crSoUMHZGZmwuVywWaz4ZdffqmUbU6n0/RhmZubKx/zlUrJQAiBlJQUAO4YMWP59r///a+MJzSWn8vGCgohsGvXLvm4Z1oW43Fd1+WSsdVqhcPhkEvM3bp1A+COW7Tb7TIGtuw8gQBJxMTEwOVyweVygST27NmDb7/9FiThcrlw4MCBcse7XC5omibHu1wuU7yikVi6vPvnQ2ioO229Z/JrhaLW4W/FWd6hPHx1h4EDB5q8FOPHj5d/yxaBN35VT5482dTeokULv16DJoTPcnUVLXen8I2nx2vUqFEEwMjISALuOq5ZWVleS7pGrVan0ynbU1JS+MQTT5CkXIK12WxyjOEVTElJIQBu2LCB69at81rSNTCWdNPS0kiSM2bM4EsvvSTH2wHO8ljSLevh279/v7xvhDSg1Eto3Dau03gOAHDy5MnS5qioKFlaLSEhQc4VGhrKmJgYkm7PnaZpzM/Pl+XWSLJHjx4UQjA3N5dLliwhgIDz8M2ZM0cur5PuZdrly5fTZrNx4MCBTEhIYGRkJEmye/fuXku6R44cIQCmpqaSJHNycvjSSy+RdHvljP+T++67T36utG/fnrGxsdKGsku6U6ZMke2NGjWScz366KMkyY0bNyoPn6JcEOAePr8bUN6hBJ8ikEiMi+MXZQTfFwAT4+L8bVqtJiMjw2vpNCkpyWcsG2n+IaBpmmxv2bKlV/9t27aRpOxn9PWMiQsKCiJJL8FHnllKNg5jibBsPFx5MXy+hKqmaVLoGddgPB4cHGwat2DBAin4rrvuOvM5S8WsxWJhYmKitNk4n91ul/FvxnMQaIKPdMfdeS7JxsfH0263kyTz8vKoaRrvvPNO5ubmMjo6Wv4oTE5OJknOmjWLuq7L8UYMpq7rjIqKku1/+9vfSJKvvfaabJsyZYpJ8M2aNYuaplEIwaCgIH733XdyLiX4FBVBCT4l+BR1ABXDpyhLpw4dGAZ3HF8owHCAzwFsZ3iry8TdXXXVVSaBt2rVKvmY4fH2vF/2ccP71KZNGxmrVjY2r7YJvurCU6QpFDVFoAs+FcOnUFSAkaNH48mFC3FvXNwFlbtT1D22ZWTIVEoCwD0AhgP4CUA/uNUeAFl+bM2aNWjcuLHPuYQQ+OOPP856vgEDBuDw4cPYuXPnOW3r2LEjdu/ejZMnT2LZsmX1LsGyQqHwxuJvAxSK2sLI0aOVwFP4JBfANwCe8mgbB+B1AKmpqUhNTYXVasWmTZt8ju/WrRs++ugjCCF8CrrY2FgMGzYMgHvjwLmSSW/YsAEhISEIDQ31qqtbHzjfZNsKRX1ACT6FQqGoJKEApgNY69H2JYD3ARwnvfqzTNvmzZvP+nhWVpbP85atomGMs1gsOH36dEVMVygU9QS1pKtQKBSVxBYRgVFwi7yi0r+jStsVCoUiEFCCT6FQKM6C02ZDuBDQhEC4EHD6yAN46NgxFEdEmOphF0dE4NCxYzVtrkKhUPhECT6FQqEoB6fNhvCiIqwAUABgBYDwoqJyRZ9nPWwl9hQKRSChBJ9CoVCUg7WoCEsBU4WVpaXtCoVCUZtQgk/hk86JiaZlrM6Jif42SRFANGrQwPT/0ahBA3+bVC34KqF2cWm7QqFQ1CaU4FN40TkxEQe3bzctYx3cvl2JPgUAt9iz5OSY/j8sOTl1UvSFAkgv05Ze2l5fGTxwoEnsDx440N8mKRSKCqAEn8KLPdu3+1zG2rN9u1/tUgQGhTk5Pv8/CnNy/GpXdVBktfrcfVtktfrVLn8xeOBAbEtLM4n9bWlpSvQpFLUAJfgUXlTVMpaR8LW+JX2t69SWZU6LxYKff/4ZWVlZuOmmm3z28QxdcAiBxo0amR7PKyzEcavVtPv2uNWKvMLCarc/ENmYlvb/7d17dNT1ue/x9zOZxFwIAcUNIoW0CDZBQCVQc7xAjRzcXrZQ2mNFCwpbllVYtrZeWrpad7dYt1bbHqRdUlBRRDm2NVg3BSEoiGIRqiAXK6IWKBfxEoSEEJN5zh8zmZ2BCQkkM5PL57XWrJnfd76/3zzPzC+ZZ76/W9xif3VZWUrjEpHGqeCTo7TkZqwrr7wy5iSy77zzTjMik9YgGZs5Q6HQCV0twcx46KGHYtp27NhBaWnpUX2P3HXh68CnH3981K4LldXVMUffdtRiD9pOsS8iR1PBJ0fJLyyMuxkrv7CwycuoG9X785//HDPSV1BQwIgRI2LaLr300ug89W/1LwulUcLWI1EnGS4rKyMjI4O+ffuSnZ1NQUEBOTk5ZGZmMnz48Gi/YDDI8OHDyc7OJjMzk0WLFgFEL0l2xx13cOaZZ0b7jxs3jsOHD5OVlcXQoUO58847Ofnkk3l/82Z+DowC1gIvArXAhs2bufXWW5uVS3ulfRpF2jB3b5W3IUOGuKTOoMJC7wxu4J3BBxUWHvcyAJ8yZYoTvo68/+lPf4q2m5m7u5uZh1dDj/arqKjwYDDogH/lK1/xffv2OeC9e/duuQSlWbp16RKzfnTr0uWElnPDDTdEP/d4N3f3w4cPH9Xeo0cPd3c/55xzjnruwQcf9J/97GcO+DvvvOOvvPKKn3TSSTGvm5+f74CfCj4F3MEngZ8VyUniG1VS4j3Al4NXR+57gI8qKUl1aCIpB6z1VlA/NXTTCJ/EtX7TppjNWOs3bWr2MseMGRN9vHbt2pj7OmZGdnY2w4YNA2Dbtm1069YNgF27djU7BmkZLX2S4QEDBvDKK68QCARwdz766CMA0tLSyMvLA8Lrhkd2D9izZw8Ab775JnD0tWfvvvvuY44KL1++HIAQMKNeewUarTqWxcuWMbikJGafxsElJSxetqyROUUk1YKpDkCkvrS0NADSO+hRkB3Vxo0bWbVqFaFQKKZQy8zMpKKiAjOjsIFdCurWlczMTKqqqpr0en/7298A+BRYCowAdgE7oMHXkTAVdyJtk0b4JCWKiopi7kXuueceAGbPns2CBQsAqK2tZePGjQDk5OQcc/7+/ftHH//85z+Pjvp179495gCQqqoqxo8fz8yZM8nIyOBfCY9WlQGB9PQWGc0WEWltVPBJ0pWUlODu0U10o0aNSnVI0grUbbL/1re+FT2a+/Dhw1x00UXHnO+LyGXOdu7cCYQP2pg/f350tLhfv3706tWLzMxMhg4dymWXXcZXv/pVbr75Zt7/4APSMjJ4/s9/ZvN77xFMTycrK0sHbYhIu6NNupIwdSMsM2bMiGlf1sAmofr7Yb388ssNPift04YNGzCz6D57EN5M++mnn2JmXH311QCce+650U2yM2bMYOrUqTGbge+//35uu+22mGV/+OGHcV+zZ8+eHD58ODpdUVHRUumIiLQq1lq/SIuKivzIHfpFREREWiMzW+furXY/JW3SFRERSYJnn32WQCD2azc/P5+CgoIURdSwsWPH8vDDDx/3fDpnauulgk9ERERiLF26lCVLlqQ6DGlBKvhERERSLBgM0rNnT9LS0jAzvve97wFw6NAhevToQVpaGoFAIHqO0oEDB0b3d33ooYcIBAJs376d/Px8OnXqFF1OcXFx9DUKCgqiy+nTp0+0fcSIEQQCAQKBAJ06deKOO+7gwIEDvPDCCwQCAebNm8e8efPIyMggEAgQDAZ55JFHAFiwYEF0mV/60peS9XbJCVDBJyIi0grU1tZSW1vLuHHjmDlzJhAuxnJycqitrWXv3r2sW7eOBQsWsG7dOioqKhgzZgy33347t912G7179wagsrKS7du3s3r1al5//XWWLFnCxIkT2b17N9XV1VRVVbFnzx5uvfVWZsyYwYoVK3jttdcIhUKsWLGC+++/n9zcXK644gpCoRDXXXcdN9xwA7NmzSIUCnHnnXcyZcoUAMaPH8+FF15IKBTi1FNPTdl7J41TwSciIpIER+6/V6duv7fvfOc7ANx4443Rc0du2LCB999/n0AgQPfu3QmFQixZsoSMjAyeeOIJSktL6datG7/85S+jy+vVqxenn3465513Hp06dWLWrFksWrSI/fv3k56eTmZmJtXV1axdu5Y5c+ZE+wIMGTLkqPi2bdtGTU0NEydOJBAI8Itf/CIaX3V1NQsXLgTgqaeeaqF3ShJBBZ+IiEgSnH322UedYurgwYPRkbHc3PCF/TIzM2P6TJgwgVAoRCgUwt159NFHAXjxxRejy6jvyAMn6s55OnTo0JjlvPrqq02Ku7q6GiA6b938derOeSmtmwo+iZFuRp4Zgch9uo64kkZMvflmTg4GCZhxcjDI1JtvTnVIIq1S3759CQQC3HTTTQCsW7eOTz75hBtvvLHBeQYNGsQzzzzD/v37AZgzZw4ffPABGzduZO7cucyePZvq6mquvPLK6Dw7duxgz549rFmzhoMHD3LjjTdy+eWXs27dOrZt2wbAokWLWLlyJZMmTeKf//wna9asicYEkJGRQXl5ORDe9y8QCHDFFVcA4U3P999/f7TfVVddBfzPCKW0Uu7eKm9DhgxxSa4geA/w5eDVkfse4EFIdWjSyuRmZXlncAPvHmedmfLd76Y6RJFWacaMGR4MBt3M3My8pKTE3d3T0tL87rvvdnf31atXO5H/u4cPH/bTTz892j8YDPrbb7/tnTt39kGDBrm7+8KFCx3wFStWeJ8+fbxTp04eCAQc8PPOOy/62gMHDowuJxAI+JNPPunu7hdddFG0PTc3193df/CDHzjgZuZPPvmkP/PMM56enu5m5oDn5+e7u/szzzzjgUDAzcx79eoVjbsjAtZ6K6ifGrrpxMsSlWdGKfD1em0vAaOB/a10PZHk65ydTc6hQ8wHpgIzOHqdGZuWxqf1rl8rIsmRn59PVlYWW7ZsSXUoHU67PvGymZ1sZkvNbGvkvusx+nY2s51mdvxncpSkOABccETbBZF2kToWKfa+Dmwh/jpTXlub9LhERKRhzd2H7y6gzN37AWWR6Yb8J7Cyma8nCZQLrDqibVWkXaRO/R8GBcRfZ7poJ26RlPjwww81uidxNbfguwqYG3k8l/DWv6OY2RCgO/BiM19PEqgSGEd4k9wXkftxkXaROvV/GEwDJnH0OnPt5MmpCU5EROIKNnP+7u6+O/J4D+GiLoaZBYAHgeuAS461MDObDEwGoieQlOT5wp10M0YTHsXJJVzsfaH996Qez8piXGSz7jeBTcAY4HPCI3vXTp7MjN/+NqUxiohIrEYLPjNbBvSI89S0+hPuXnf0zpFuBha5+87GLqrs7rOAWRA+aKOx2KTlqbiTxnxeWUnn7GxGHzoU/WHgWVmEKjUWLCLSWjVa8Ll7g6NyZrbXzE5z991mdhrwUZxuxcCFZnYz0AnIMLOD7n6s/f1EpBX7XMWdiEib0txNus8DE4D7IvcLj+zg7tfWPTaz64EiFXsiIiIiydPcgzbuA0aa2VbC++fdB2BmRWY2u7nBiYiIiEjzNavgc/dP3L3E3fu5+yXu/mmkfa27/3uc/o+7+5TmvKaIiEgyXXrJJTGXnLz0kmMefyjSKulauiIiIg249JJLWF9WRilwGCgF1peVqeiTNkcFn4iISANWl5VFryyTHrmfH2kXaUtU8ImIiDRAl5yU9kIFn4iISAN0yUlpL1TwiYhImzBw4EAaO4F/fXl5eUyfPr1Zr1lcUhL3kpPFJSXNWq5IsqngExGRdunzzz9n6dKlTepbXl7Ob+NcEnDxsmUMLilhNHAS4QvGDy4pYfGyZS0aq0iiqeATEZGE+/73v4+ZxdxWr15Nenp6TFt1dTUAkydPjmm/6aabYpZ30UUXRdtLSkpi+t5333107x6+tPuKFSswM9auXXvM+Boq+CBc9O13J+TOfncVe9ImqeATEZGkGTJkCB65ZveIESN49NFHcfdoW9++fQH4/e9/T48ePXB31q9fz4QJE6LLGDZsGK+88grdu3dn5cqVLF++nIcffpicnBwyMzP50Y9+xMCBA4HwZt0vf/nL7Nq1C4DHH3+cq666ihEjRtCvXz/+4z/+A4C77rqLbdu2cfbZZ3P77bczfvx4SktLo6957bXXsnDhUReTEmkzVPCJiEjS1I20FRQUUF1dze233x4dmQPYu3cvS5YsAWD37t0ADBo0iOLi4ugy3njjDfr378+OHTs488wzAZgyZQoVFRVUVVUBkJsbPqxi8ODBPPfcc/z0pz+Nzr9mzRr++Mc/smHDBp599lnWrl3LfffdR9++fXnrrbd44IEHmDRpEo8//jgA+/fv57XXXuPyyy9P4Dsjklgq+EREJGX27t3Lj3/8Y9ydtLQ0QqFQk+Z77733GDp0KMuXLwfgZz/7GRkZGYRCIdw9OsJnZgwcOJAPP/wwOu/IkSM55ZRTyMrK4hvf+AarVh15HC4MHz6crVu3sm/fPp5++mnGjh1LMNjcy8+LpI4KPhERSZqvfe1rAGzZsoWMjAwAxowZw/Lly6mtrQVg1KhRAJx++ukAbN68mdWrV0eXMWbMGEKhELW1taxbtw6Av/zlL6Snp7Njxw6++c1vEgiEv94+/vhjAoEANTU10fmPPNK3oSN/x48fz7x583jssceYOHFis3MXSSUVfCIikjRr1qyJFlgvv/wygUCAoUOHUnLEaU4mTpzIrl27MDMGDBjAvHnzos/dc8895OXlsXHjRiZMmMDFF1/MmjVrqKiooE+fPixevBiATp06sWnTJswspuBbunQpn376KYcOHaK0tJTzzz+f3NxcDhyIPZ3y9ddfz69//WsACgsLE/J+iCSLCj4REUmaugM03J3i4mJqa2tj2uoKszlz5sS0z5w5k7fffht3p7CwkEceeYTBgwdz4MABysvLWb16NTk5Obg7Bw8eBMKbeevmr785dtiwYYwdO5ZBgwYxduxYioqKOOWUUzj//PM566yzuP322wHo3r07BQUF3HDDDcl/o0RamHZIEBGRNufqq6/m6quvjmmrK/QA7r777gaf69WrV8wRuHXmz58fM11ZWcnWrVu55pprWiBikdTSCJ+IiCTcr371q+ipV9qCZcuWUVBQwNSpU8nLy0t1OCLNZq31D7CoqMgbO1GmiIiISGtgZuvcvSjVcTREI3wiIiIi7ZwKPhERkSOsWrWK3r17H9c88+fPJy0tLUERiTSPCj4REZEjrF+/nh07dhzVXl5e3uA848aNi55LUKS1UcEnIiJyhFtvvRUgetk3MyMQCNC1a1fy8vI47bTTon1zc3Pp2bMnM2fOjJ5j8MwzzyQtLY1AIICZ0alTp2j/M844I7rMjIwMMjMzk5ucdEgq+EREpEGndu1KnhkBM/LMOLVr11SHlBS/+c1vgPB5A0ePHh1tc3fuvfde9uzZA0BVVRUHDx5k7ty5Ry0jFArx6quv8sEHH1BRUcGcOXOYP38+27Zto6ysjH379vHFF18kLynp0FTwiYhIXKd27UqwvJxS4DBQCgTLyztM0VefmTF16lQAbrnlluj0FVdcQSAQYOTIkUfNk56eTnFxMfn5+ZgZL774IjNmzCAYDHLxxRfTrVs3evbsmexUpINSwSciInFVl5czH/g6kB65nx9pPxFPP/UUZ+XnkxYIcFZ+Pk8/9VTLBZtkgwYNYvbs2bz88stceOGFcfvUXc+3TnV1dTJCE4lLBZ+IiMR1ALjgiLYLIu3H6+mnnmLa5MnM+Mc/qHJnxj/+wbTJk1tt0de/f/9jPr948WKqqqqora3lhRdeaPJyv/vd71JTU8PKlSspLy9n165dzQ1VpElU8ImISFy5wKoj2lZF2o/X9GnTmFNZGTNaOKeykunTpjUzysQYOXIkaWlpmFncy7D16NGDjIwMcnJyYg7IaMz48ePJz89n+PDhdO3alWAwSHp6ekuGLhKXCj4REYkro0sXxgEvAV9E7sdF2o/Xlu3b444Wbtm+vblhJkxNTQ3ujrsTCoVinquqqqK6upqf/OQn0bZbbrklevm4v//971RVVUWfC4VCPPfccwAsXLgQd+ezzz6jtraWwsLCJGQjHZ0KPhERiWvfZ59R06ULo4GTgNFATZcu7Pvss+NeVkHv3nFHCwuO8+TGrcG0adPIysoiKyuLu+6667jnLy4uxszo2rUrJ510Eq+++moCohSJpYJPREQatO+zz9jvTsid/e4nVOwBTJs+nUnZ2TGjhZOys5k2fXpLhpsU06dPx92prKw8ofkrKiqiI4eHDh0iGAy2cIQiR9NaJiIiCXfNtdcCMHXaNLZs305B795Mnz492i4iiWV1+xu0NkVFRb527dpUhyEiIiLSKDNb5+5FqY6jIdqkKyIiItLOqeATERERaedU8ImIiLQR+/btO+Z0PIcPH+bzzz9PVEjSRqjgExERaSVycnIws+ht3759MdPnn3/+UdPBYDA6XXfE78SJE6NtmZmZ/PCHP0xxZpJqKvhERKRdGTxgAHlmBMzIM2PwgAGpDqlJLrvsMiorK3nvvfeiJ3Du06cPALm5ubg77777bsx0Xl4etbW1LF68mBdeeIHa2lqKi4ujyxwwYADuzqxZs5KfkLQqKvhERKTdGDxgAB9t3kwpcBgoBT7avLlNFH1Lly4F4IwzzsDMAKJX69i2bVtM37rpN998E4BRo0Zx+eWXA/DGG29E+23cuDGxQUuboYJPRETajQ83b2Y+xFyzd36kva2oOylz/Uu6nXrqqTF9jpwWaYwKPhERaTcOQNxr9h5IQSzH6+KLLwbg3nvvBeB3v/sdkydPPuY85557LgAvvfQSS5YsAWDo0KEJjFLaKp14WURE2o08M0oJj+zVeYnwdYD3t9Lvu/pyc3M5ePBgdHrAgAFs2rSJ+t/VZhYzHQwGqa2tBSAtLY2amhomTpzIY489Rmv9jm+PdOJlERGRJMkvLGQcxFyzd1ykvS04cOBAzCbdjRs3HlW0HTldU1MT7V9TUwPAo48+qmJPYjSr4DOzk81sqZltjdx3baBfbzN70cy2mNlmM8tvzuuKiIjEs37TJv6lsJDRwEmER/b+pbCQ9Zs2pTgykdRq7gjfXUCZu/cDyiLT8TwBPODuBcAw4KNmvq6IiEhc6zdtYr87IXf2u6vYE6H5Bd9VwNzI47mEf0zFMLNCIOjuSwHc/aC7VzbzdUVERESkiZpb8HV3992Rx3uA7nH69AfKzexPZvammT1gZmnxFmZmk81srZmtbcrlYkRERESkccHGOpjZMqBHnKem1Z9wdzezeHuIBoELgXOA7cAC4HpgzpEd3X0WMAvCR+k2FpuIiIiINK7Rgs/dL2noOTPba2anuftuMzuN+Pvm7QTecvf3I/OUAucRp+ATERERkZbX3E26zwMTIo8nAAvj9HkD6GJmdacFvxhoO6c8FxEREWnjmlvw3QeMNLOtwCWRacysyMxmA7h7LfBDoMzM3gYM+H0zX1dEREREmqjRTbrH4u6fACVx2tcC/15veikwqDmvJSIi0hp1zs7GDh3iAJALeFYWn1fqZBTSuuhKGyIiIieoc3Y2OYcOUQocBkqBnEOH6JydneLIRGKp4BMRETlBdugQ8wlfuzc9cj8/0i7SmqjgExEROUEHgAuOaLsg0i7SmqjgExEROUG5wKoj2lZF2kVaExV8IiIiJ8izshgHvAR8EbkfF2kXaU2adZSuiIhIR/Z5ZSWds7MZraN0pZVTwSciItIMKu6kLdAmXREREZF2TgWfiIiISDungk9ERESknVPBJyIiItLOqeATERERaedU8ImIiIi0cyr4RERERNo5FXwiIiIi7ZwKPhEREZF2TgWfiIiISDungk9ERESknVPBJyIiItLOmbunOoa4zGwf8I8ELb4b8HGClt2aKe+Op6Pmrrw7FuXd8bTG3Pu4+6mpDqIhrbbgSyQzW+vuRamOI9mUd8fTUXNX3h2L8u54OnLuJ0qbdEVERETaORV8IiIiIu1cRy34ZqU6gBRR3h1PR81deXcsyrvj6ci5n5AOuQ+fiIiISEfSUUf4RERERDoMFXwiIiIi7VyHKPjM7GQzW2pmWyP3XRvod7+ZbTKzLWb2f83Mkh1rSzqOvHub2YuRvDebWX5yI215Tc090rezme00s4eTGWMiNCVvMzvbzFZH1vUNZnZ1KmJtCWZ2qZn93czeM7O74jx/kpktiDz/1/awbkOT8r4t8re8wczKzKxPKuJsaY3lXa/fWDNzM2sXp+1oSt5m9n8in/kmM5uf7BgToQnreW8ze8nM3oys65elIs62okMUfMBdQJm79wPKItMxzOx/AecDg4CzgKHA8GQGmQCN5h3xBPCAuxcAw4CPkhRfIjU1d4D/BFYmJarEa0relcB4dx8AXAr82sy6JDHGFmFmacBM4F+BQuAaMys8otsk4DN3PwP4FfBfyY2y5TUx7zeBIncfBPwBuD+5Uba8JuaNmeUCtwJ/TW6EidGUvM2sH/Aj4PzI3/X3kh5oC2vi5/0T4P+5+znAt4HfJjfKtqWjFHxXQWeU6AAABDFJREFUAXMjj+cCo+P0cSATyABOAtKBvUmJLnEazTvyBxR096UA7n7Q3SuTF2LCNOUzx8yGAN2BF5MUV6I1mre7v+vuWyOPdxEu8Fvt2eGPYRjwnru/7+7VwDOE86+v/vvxB6CkrY/c04S83f2len/HrwO9khxjIjTl84bwD7j/AqqSGVwCNSXvG4GZ7v4ZgLu3hx/tTcnbgc6Rx3nAriTG1+Z0lIKvu7vvjjzeQ/gLPoa7rwZeAnZHbkvcfUvyQkyIRvMG+gPlZvanyLD4A5FfVm1do7mbWQB4EPhhMgNLsKZ85lFmNozwj5xtiQ4sAU4HdtSb3hlpi9vH3WuA/cApSYkucZqSd32TgL8kNKLkaDRvMzsX+JK7/3cyA0uwpnze/YH+Zvaqmb1uZpcmLbrEaUredwPXmdlOYBEwNTmhtU3BVAfQUsxsGdAjzlPT6k+4u5vZUeeiMbMzgAL+55fwUjO70N1fafFgW1Bz8ya8DlwInANsBxYA1wNzWjbSltcCud8MLHL3nW1p0KcF8q5bzmnAk8AEdw+1bJTSGpjZdUARbX/3lEZFfsA9RPj/V0cTBPoBIwh/h600s4HuXp7SqBLvGuBxd3/QzIqBJ83sLP0/i6/dFHzufklDz5nZXjM7zd13R77k4g13jwFed/eDkXn+AhQDrbrga4G8dwJvufv7kXlKgfNoAwVfC+ReDFxoZjcDnYAMMzvo7sfa3y/lWiBvzKwz8N/ANHd/PUGhJto/gS/Vm+4VaYvXZ6eZBQlv9vkkOeElTFPyxswuIfwjYLi7H05SbInUWN65hPe/fjnyA64H8LyZ/Zu7r01alC2vKZ/3TuCv7v4F8IGZvUu4AHwjOSEmRFPynkR4P2TcfbWZZQLdaB/7obe4jrJJ93lgQuTxBGBhnD7bgeFmFjSzdMK/iNv6Jt2m5P0G0MXM6vbhuhjYnITYEq3R3N39Wnfv7e75hDfrPtHai70maDRvM8sAniOc7x+SGFtLewPoZ2ZfjuT0bcL511f//fgmsNzb/tnmG83bzM4BHgH+rZ3szwWN5O3u+929m7vnR/6mXyecf1su9qBp63kp4dE9zKwb4U287yczyARoSt7bgRIAMysgvB/+vqRG2Za4e7u/Ed5npwzYCiwDTo60FwGzI4/TCP+D3EK44Hko1XEnI+/I9EhgA/A28DiQkerYk5V7vf7XAw+nOu5k5A1cB3wBvFXvdnaqYz/BfC8D3iW8D+K0SNvPCX/RQ/gL4FngPWAN8JVUx5ykvJcRPuis7vN9PtUxJyPvI/q+TPhI5ZTHnYTP2whvzt4c+T/+7VTHnKS8C4FXgfWR9fx/pzrm1nzTpdVERERE2rmOsklXREREpMNSwSciIiLSzqngExEREWnnVPCJiIiItHMq+ERERETaORV8IiIiIu2cCj4RERGRdu7/A6lMvDvUqb0UAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 720x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OPEyIk10o8pB",
        "outputId": "56f5e569-38e1-45bc-b9a8-457e08fab40d"
      },
      "source": [
        "num_features = 300\n",
        "testDataVecs = getAvgFeatureVecs(test, model, num_features)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Block id 0 of 1994\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:16: DeprecationWarning: Call to deprecated `__getitem__` (Method will be removed in 4.0.0, use self.wv.__getitem__() instead).\n",
            "  app.launch_new_instance()\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xoP78152Jcf8"
      },
      "source": [
        "# **Random Forest**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BOLGPTnF-sb8"
      },
      "source": [
        "**Split the HDFS.1**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hkqHlT1N-ARn"
      },
      "source": [
        "X = trainDataVecs\n",
        "y = log.iloc[:, 2]"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VMXk-FqA9_hb"
      },
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ruqSwtAUX9mF"
      },
      "source": [
        "np.save(\"X_train.npy\", X_train)\n",
        "np.save(\"X_test.npy\", X_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7rv0Od2wORK6",
        "outputId": "b1bc08a4-5d5c-43ba-bd3c-508b49db9c4b"
      },
      "source": [
        "y_train.value_counts()"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Normal     446525\n",
              "Anomaly     13523\n",
              "Name: Label, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cQ5bIZ1kVKuQ",
        "outputId": "5e9fa4a1-4598-4fe6-9e45-d032a3ac9805"
      },
      "source": [
        "y_test.value_counts()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Normal     111698\n",
              "Anomaly      3315\n",
              "Name: Label, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BFB61tXd-7L7"
      },
      "source": [
        "# Train the model\n",
        "classifier = RandomForestClassifier(n_estimators=1000, random_state=0)\n",
        "classifier.fit(X_train, y_train)\n",
        "\n",
        "# Prediction\n",
        "y_pred = classifier.predict(X_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yt1wId_u_Lha",
        "outputId": "6def8017-9b65-47e2-d5df-444012d4384a"
      },
      "source": [
        "# Evaluate model\n",
        "from sklearn.metrics import classification_report, confusion_matrix, accuracy_score\n",
        "print(confusion_matrix(y_test,y_pred))\n",
        "print(classification_report(y_test,y_pred))\n",
        "print(accuracy_score(y_test, y_pred))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[  3311      4]\n",
            " [     5 111693]]\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "     Anomaly       1.00      1.00      1.00      3315\n",
            "      Normal       1.00      1.00      1.00    111698\n",
            "\n",
            "    accuracy                           1.00    115013\n",
            "   macro avg       1.00      1.00      1.00    115013\n",
            "weighted avg       1.00      1.00      1.00    115013\n",
            "\n",
            "0.9999217479763157\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "erWYS7O1R_Nv"
      },
      "source": [
        "# Save model\n",
        "\n",
        "with open('RF_W2V_AVG_classifier', 'wb') as picklefile:\n",
        "    pickle.dump(classifier,picklefile)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eRV1ZcJHSQ82"
      },
      "source": [
        "# Load model\n",
        "\n",
        "with open('RF_BW_classifier', 'rb') as training_model:\n",
        "    model = pickle.load(training_model)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V7rNlPdTrmtu"
      },
      "source": [
        "# **Logistic Regression**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "A7RSjy2_rpwr",
        "outputId": "7a959774-a2f2-4bad-bcd5-1799b047fbaa"
      },
      "source": [
        "# Train the model\n",
        "classifier = LogisticRegression(verbose=1, solver='liblinear',random_state=0, C=5, penalty='l2',max_iter=1000)\n",
        "classifier.fit(X_train, y_train)\n",
        "\n",
        "# Prediction\n",
        "y_pred = classifier.predict(X_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[LibLinear]"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EUdU4kmfsBpV",
        "outputId": "5973657f-836f-4229-9168-d47512e2fe57"
      },
      "source": [
        "# Evaluate model\n",
        "from sklearn.metrics import classification_report, confusion_matrix, accuracy_score\n",
        "print(confusion_matrix(y_test,y_pred))\n",
        "print(classification_report(y_test,y_pred))\n",
        "print(accuracy_score(y_test, y_pred))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[  1507   1808]\n",
            " [    70 111628]]\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "     Anomaly       0.96      0.45      0.62      3315\n",
            "      Normal       0.98      1.00      0.99    111698\n",
            "\n",
            "    accuracy                           0.98    115013\n",
            "   macro avg       0.97      0.73      0.80    115013\n",
            "weighted avg       0.98      0.98      0.98    115013\n",
            "\n",
            "0.9836714110578804\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EXkh35nfLSuH"
      },
      "source": [
        "# **Word2vec Training - Clustering**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3MHMSyZ5LXzc",
        "outputId": "daeeac18-02e5-4cec-93ee-2008a2bb1233"
      },
      "source": [
        "from sklearn.cluster import KMeans\n",
        "\n",
        "# Set \"k\" (num_clusters) to be 1/5th of the vocabulary size, or an average of 5 words per cluster\n",
        "word_vectors = model.wv.syn0\n",
        "num_clusters = word_vectors.shape[0] // 5\n",
        "\n",
        "# Initalize a k-means object and use it to extract centroids\n",
        "kmeans_clustering = KMeans(n_clusters = num_clusters)\n",
        "idx = kmeans_clustering.fit_predict(word_vectors)\n"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:4: DeprecationWarning: Call to deprecated `syn0` (Attribute will be removed in 4.0.0, use self.wv.vectors instead).\n",
            "  after removing the cwd from sys.path.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5E7sUUPDNbLw"
      },
      "source": [
        "                                                                                           \n",
        "word_centroid_map = dict(zip(model.wv.index2word, idx))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BXYTyCKbNgaG",
        "outputId": "97697d54-829d-401f-8544-e1eefe965a3b"
      },
      "source": [
        "# For the first 10 clusters\n",
        "for cluster in range(0,10):\n",
        "  \n",
        "    print (\"\\nCluster %d\" % cluster)\n",
        "\n",
        "    words = []\n",
        "    for i in range(0,len(word_centroid_map.values())):\n",
        "        if(list(word_centroid_map.values())[i] == cluster):\n",
        "          words.append(list(word_centroid_map.keys())[i])\n",
        "    print(words)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Cluster 0\n",
            "['to', 'Served', 'exception', 'Got', 'while', 'serving', 'Verification']\n",
            "\n",
            "Cluster 1\n",
            "['blk', 'block', 'for', 'src', 'dest', 'Receiving', 'of', 'Received', 'PacketResponder', 'terminating', 'from', 'size', 'allocateBlock', 'user', 'root']\n",
            "\n",
            "Cluster 2\n",
            "['Exception', 'in', 'receiveBlock', 'Changing', 'offset', 'net', 'Interrupted']\n",
            "\n",
            "Cluster 3\n",
            "['received', 'Redundant']\n",
            "\n",
            "Cluster 4\n",
            "['transfer', 'ask', 'Starting', 'thread', 'Transmitted']\n",
            "\n",
            "Cluster 5\n",
            "['BLOCK', 'mnt', 'hadoop', 'file', 'Deleting', 'delete', 'added', 'succeeded']\n",
            "\n",
            "Cluster 6\n",
            "['NameSystem', 'is', 'blockMap', 'updated']\n",
            "\n",
            "Cluster 7\n",
            "['trying', 'empty', 'packet']\n",
            "\n",
            "Cluster 8\n",
            "['rand', 'randtxt', 'sortrand', 'wcnt', 'mapred']\n",
            "\n",
            "Cluster 9\n",
            "['PendingReplicationMonitor', 'timed', 'out']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Uu6xIhVTFD4n"
      },
      "source": [
        "# **Bag of Words**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NsfzDNTMWMqm"
      },
      "source": [
        "normal = log[log['Label'] == 'Normal']\n",
        "anomaly = log[log['Label'] == 'Anomaly']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bfEuVvv8TNTy"
      },
      "source": [
        "items = (log['Event_Description'].unique())"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8HKOPTLj1GM7"
      },
      "source": [
        "data_dtm=data_dtm.transpose()\n",
        "\n",
        "top_dict = {}\n",
        "for i in data_dtm.columns:\n",
        "  top = data_dtm[i].sort_values(ascending=False).head(20)\n",
        "  top_dict[i]=list(zip(top.index,top.values))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LkhqtKUV0_Bf"
      },
      "source": [
        "words =[]\n",
        "\n",
        "for x in data_dtm.columns:\n",
        "  top = [word for (word, count) in top_dict[x]]\n",
        "  for t in top:\n",
        "    words.append(t)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i2vonVh-1Wlq"
      },
      "source": [
        "stop_words = [word for word,count in Counter(words).most_common() if count>575061*0.7]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iJpYRABRd8yx"
      },
      "source": [
        "# **Random Forest - Using Bag of Words**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fyfo0_z6pq6j"
      },
      "source": [
        "data_dtm = pd.read_pickle(\"/content/data586_Project/MyDrive/DATA586_Project/dtm.pkl\")"
      ],
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "id": "3l5iaqpQeZbd",
        "outputId": "0b0fb90e-8efa-4b0e-94bc-6b2837baa835"
      },
      "source": [
        "data_dtm.reset_index(inplace=True)\n",
        "data_dtm.head(10)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>BlockId</th>\n",
              "      <th>Label</th>\n",
              "      <th>added</th>\n",
              "      <th>adding</th>\n",
              "      <th>addstoredblock</th>\n",
              "      <th>allocateblock</th>\n",
              "      <th>ask</th>\n",
              "      <th>belong</th>\n",
              "      <th>blk</th>\n",
              "      <th>block</th>\n",
              "      <th>blockinfo</th>\n",
              "      <th>blockmap</th>\n",
              "      <th>broken</th>\n",
              "      <th>ch</th>\n",
              "      <th>changing</th>\n",
              "      <th>channel</th>\n",
              "      <th>channels</th>\n",
              "      <th>closed</th>\n",
              "      <th>closedbyinterruptexception</th>\n",
              "      <th>conf</th>\n",
              "      <th>connected</th>\n",
              "      <th>connection</th>\n",
              "      <th>current</th>\n",
              "      <th>data</th>\n",
              "      <th>datanode</th>\n",
              "      <th>delete</th>\n",
              "      <th>deleting</th>\n",
              "      <th>dest</th>\n",
              "      <th>dfs</th>\n",
              "      <th>does</th>\n",
              "      <th>ec</th>\n",
              "      <th>eofexception</th>\n",
              "      <th>error</th>\n",
              "      <th>exception</th>\n",
              "      <th>existing</th>\n",
              "      <th>failed</th>\n",
              "      <th>file</th>\n",
              "      <th>got</th>\n",
              "      <th>grep</th>\n",
              "      <th>grepa</th>\n",
              "      <th>...</th>\n",
              "      <th>served</th>\n",
              "      <th>serving</th>\n",
              "      <th>size</th>\n",
              "      <th>socketchannel</th>\n",
              "      <th>sockettimeoutexception</th>\n",
              "      <th>sort</th>\n",
              "      <th>sorter</th>\n",
              "      <th>sortrand</th>\n",
              "      <th>split</th>\n",
              "      <th>src</th>\n",
              "      <th>starting</th>\n",
              "      <th>stream</th>\n",
              "      <th>subdir</th>\n",
              "      <th>succeeded</th>\n",
              "      <th>task</th>\n",
              "      <th>temp</th>\n",
              "      <th>temporary</th>\n",
              "      <th>terminating</th>\n",
              "      <th>text</th>\n",
              "      <th>thread</th>\n",
              "      <th>timed</th>\n",
              "      <th>timeout</th>\n",
              "      <th>transfer</th>\n",
              "      <th>transmitted</th>\n",
              "      <th>trying</th>\n",
              "      <th>unexpected</th>\n",
              "      <th>updated</th>\n",
              "      <th>user</th>\n",
              "      <th>valid</th>\n",
              "      <th>verification</th>\n",
              "      <th>volumemap</th>\n",
              "      <th>waiting</th>\n",
              "      <th>wcnt</th>\n",
              "      <th>wordcount</th>\n",
              "      <th>write</th>\n",
              "      <th>writeblock</th>\n",
              "      <th>writer</th>\n",
              "      <th>writing</th>\n",
              "      <th>written</th>\n",
              "      <th>xml</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>24807389905763</td>\n",
              "      <td>Normal</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>22</td>\n",
              "      <td>19</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>27644297472372</td>\n",
              "      <td>Normal</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>22</td>\n",
              "      <td>19</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>33380551139732</td>\n",
              "      <td>Normal</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>34</td>\n",
              "      <td>27</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>8</td>\n",
              "      <td>4</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>35641716531335</td>\n",
              "      <td>Normal</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>22</td>\n",
              "      <td>19</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>40083153975957</td>\n",
              "      <td>Normal</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>22</td>\n",
              "      <td>19</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>72472557232301</td>\n",
              "      <td>Normal</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>22</td>\n",
              "      <td>19</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>72859758335309</td>\n",
              "      <td>Normal</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>22</td>\n",
              "      <td>19</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>78222097481044</td>\n",
              "      <td>Normal</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>13</td>\n",
              "      <td>13</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>97457323857639</td>\n",
              "      <td>Normal</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>13</td>\n",
              "      <td>13</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>99286907018543</td>\n",
              "      <td>Normal</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>13</td>\n",
              "      <td>13</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>10 rows × 132 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "          BlockId   Label  added  adding  ...  writer  writing  written  xml\n",
              "0  24807389905763  Normal      6       0  ...       0        0        0    0\n",
              "1  27644297472372  Normal      6       0  ...       0        0        0    0\n",
              "2  33380551139732  Normal      6       0  ...       0        0        0    0\n",
              "3  35641716531335  Normal      6       0  ...       0        0        0    0\n",
              "4  40083153975957  Normal      6       0  ...       0        0        0    0\n",
              "5  72472557232301  Normal      6       0  ...       0        0        0    0\n",
              "6  72859758335309  Normal      6       0  ...       0        0        0    0\n",
              "7  78222097481044  Normal      3       0  ...       0        0        0    0\n",
              "8  97457323857639  Normal      3       0  ...       0        0        0    0\n",
              "9  99286907018543  Normal      3       0  ...       0        0        0    0\n",
              "\n",
              "[10 rows x 132 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JPYwd5vrkhzJ"
      },
      "source": [
        "label_col = {'Normal': 1, 'Anomaly': 0}\n",
        "\n",
        "data_dtm.Label = [label_col[item] for item in data_dtm.Label]"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n7poL3l-uCwZ"
      },
      "source": [
        "y = data_dtm.iloc[:, 2]\n",
        "X = data_dtm.iloc[:, 3:]"
      ],
      "execution_count": 30,
      "outputs": []
    }
  ]
}