Classificação de Tráfego.ipynb

{"cells":[{"cell_type":"markdown","metadata":{"id":"cifnL38ZNTEi"},"source":["# SBRC2022"]},{"cell_type":"markdown","metadata":{"id":"urtTT5qRNTEk"},"source":["<div align=\"justify\"> O objetivo deste notebook é apresentar exemplos de utilização de modelos de Aprendizado de Máquina aplicado aos problemas de redes de computadores. Iremos tratar especificamente de Classificação de Tráfego, indentificando os diveros tipos  de tráfego baseado nas medições da rede. Para isto iremos utilizar a linguagem Python com as bibliotecas NumPy, Pandas e Scikit learn. </div>\n","\n","Esse Notebook aboradará os seguintes passos:\n","    <ul>\n","    <li>Preparação dos Dados</li>\n","    <li>Definição das Classes</li>\n","    <li>Classificação com Modelos AM</li>\n","    <li>Conclusão</li>\n","\n"]},{"cell_type":"markdown","metadata":{"id":"dy_6BmGMtrxt"},"source":["## Conectar Google Drive"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"p6IGAVyh4UdH","executionInfo":{"status":"ok","timestamp":1653643095927,"user_tz":180,"elapsed":19156,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"baa4ed22-4780-4e4b-dea5-6af131a0f97d"},"source":["from google.colab import drive\n","drive.mount('/content/drive')"],"execution_count":1,"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/drive\n"]}]},{"cell_type":"code","source":["!ls 'drive/MyDrive/Doutorado/Minicurso'"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"np6Q9awYSN62","executionInfo":{"status":"ok","timestamp":1653643100597,"user_tz":180,"elapsed":682,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"d9166f64-1424-4767-ad44-fd5636f2a2fe"},"execution_count":2,"outputs":[{"output_type":"stream","name":"stdout","text":["'Classificação de Tráfego.ipynb'\n","'Classifications_Models-Rolling Windows.ipynb'\n","'Cópia de Classificação de Tráfego.ipynb'\n"," dataset_classificacao.csv\n"," dataset_sdn.csv\n"," DDoS_Detection.ipynb\n"," filtered_dataset_classificacao.csv\n"," modelo.mlp\n"," Untitled0.ipynb\n"]}]},{"cell_type":"code","source":["path_dataset = 'drive/MyDrive/Doutorado/Minicurso/dataset_classificacao.csv'\n","path_dataset_filtered = 'drive/MyDrive/Doutorado/Minicurso/filtered_dataset_classificacao.csv'\n","path_model_mlp = 'drive/MyDrive/Doutorado/Minicurso/modelo.mlp'"],"metadata":{"id":"R8FbkhoFSri7","executionInfo":{"status":"ok","timestamp":1653643104845,"user_tz":180,"elapsed":266,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":3,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"z6yBunvrNTEl"},"source":["## Importar Bibliotecas"]},{"cell_type":"code","execution_count":4,"metadata":{"id":"EM2BSvK0NTEm","executionInfo":{"status":"ok","timestamp":1653643108457,"user_tz":180,"elapsed":1790,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[],"source":["#libraries for data manipulation\n","import numpy as np\n","import pandas as pd\n","\n","#Models for Machine Learning\n","from sklearn.neighbors import KNeighborsClassifier\n","from sklearn.naive_bayes import GaussianNB\n","from sklearn import svm\n","from sklearn.ensemble import RandomForestClassifier\n","from sklearn.ensemble import AdaBoostClassifier\n","\n","\n","#libraries for selecting features and parameters\n","from sklearn.feature_selection import RFE\n","from sklearn.model_selection import GridSearchCV\n","from sklearn.model_selection import train_test_split\n","\n","#libraries for validation\n","from sklearn.model_selection import cross_val_score\n","from sklearn import metrics\n","from sklearn.metrics import confusion_matrix, classification_report,accuracy_score, make_scorer\n","from sklearn.metrics import balanced_accuracy_score\n","\n","#libraries for graphics\n","import seaborn as sn\n","import pandas as pd\n","import matplotlib.pyplot as plt\n","\n","import time\n"]},{"cell_type":"markdown","metadata":{"id":"Ke7wGlZgNTEn"},"source":["## Preparação dos Dados"]},{"cell_type":"markdown","source":["### Carregar Dataset"],"metadata":{"id":"7LsFXLiwdLEg"}},{"cell_type":"code","execution_count":5,"metadata":{"id":"9dg5Mb0VNTEn","executionInfo":{"status":"ok","timestamp":1653643167354,"user_tz":180,"elapsed":55387,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[],"source":["df = pd.read_csv(path_dataset, sep=',',dtype={'ProtocolName':'category'})"]},{"cell_type":"code","execution_count":6,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":835},"id":"1-0_haFHNTEo","executionInfo":{"status":"ok","timestamp":1653643184759,"user_tz":180,"elapsed":1824,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"bcacc5c0-9fae-4e8d-e99f-e1f85a6e542a"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["                                       Flow.ID      Source.IP  Source.Port  \\\n","0          172.19.1.46-10.200.7.7-52422-3128-6    172.19.1.46        52422   \n","1          172.19.1.46-10.200.7.7-52422-3128-6     10.200.7.7         3128   \n","2         10.200.7.217-50.31.185.39-38848-80-6   50.31.185.39           80   \n","3         10.200.7.217-50.31.185.39-38848-80-6   50.31.185.39           80   \n","4        192.168.72.43-10.200.7.7-55961-3128-6  192.168.72.43        55961   \n","...                                        ...            ...          ...   \n","3577291  10.200.7.199-98.138.79.73-42135-443-6   98.138.79.73          443   \n","3577292  10.200.7.217-98.138.79.73-51546-443-6   98.138.79.73          443   \n","3577293  10.200.7.218-98.138.79.73-44366-443-6   98.138.79.73          443   \n","3577294  10.200.7.195-98.138.79.73-52341-443-6   98.138.79.73          443   \n","3577295  10.200.7.196-98.138.79.73-34188-443-6   98.138.79.73          443   \n","\n","        Destination.IP  Destination.Port  Protocol           Timestamp  \\\n","0           10.200.7.7              3128         6  26/04/201711:11:17   \n","1          172.19.1.46             52422         6  26/04/201711:11:17   \n","2         10.200.7.217             38848         6  26/04/201711:11:17   \n","3         10.200.7.217             38848         6  26/04/201711:11:17   \n","4           10.200.7.7              3128         6  26/04/201711:11:17   \n","...                ...               ...       ...                 ...   \n","3577291   10.200.7.199             42135         6  15/05/201705:43:40   \n","3577292   10.200.7.217             51546         6  15/05/201705:46:10   \n","3577293   10.200.7.218             44366         6  15/05/201705:45:39   \n","3577294   10.200.7.195             52341         6  15/05/201705:45:59   \n","3577295   10.200.7.196             34188         6  15/05/201705:46:05   \n","\n","         Flow.Duration  Total.Fwd.Packets  Total.Backward.Packets  ...  \\\n","0                45523                 22                      55  ...   \n","1                    1                  2                       0  ...   \n","2                    1                  3                       0  ...   \n","3                  217                  1                       3  ...   \n","4                78068                  5                       0  ...   \n","...                ...                ...                     ...  ...   \n","3577291        2290821                  5                       4  ...   \n","3577292             24                  5                       0  ...   \n","3577293        2591653                  6                       5  ...   \n","3577294        2622421                  4                       3  ...   \n","3577295        2009138                  3                       2  ...   \n","\n","         Active.Std  Active.Max  Active.Min  Idle.Mean  Idle.Std  Idle.Max  \\\n","0               0.0         0.0         0.0        0.0       0.0       0.0   \n","1               0.0         0.0         0.0        0.0       0.0       0.0   \n","2               0.0         0.0         0.0        0.0       0.0       0.0   \n","3               0.0         0.0         0.0        0.0       0.0       0.0   \n","4               0.0         0.0         0.0        0.0       0.0       0.0   \n","...             ...         ...         ...        ...       ...       ...   \n","3577291         0.0         0.0         0.0        0.0       0.0       0.0   \n","3577292         0.0         0.0         0.0        0.0       0.0       0.0   \n","3577293         0.0         0.0         0.0        0.0       0.0       0.0   \n","3577294         0.0         0.0         0.0        0.0       0.0       0.0   \n","3577295         0.0         0.0         0.0        0.0       0.0       0.0   \n","\n","         Idle.Min   Label  L7Protocol  ProtocolName  \n","0             0.0  BENIGN         131    HTTP_PROXY  \n","1             0.0  BENIGN         131    HTTP_PROXY  \n","2             0.0  BENIGN           7          HTTP  \n","3             0.0  BENIGN           7          HTTP  \n","4             0.0  BENIGN         131    HTTP_PROXY  \n","...           ...     ...         ...           ...  \n","3577291       0.0  BENIGN          91           SSL  \n","3577292       0.0  BENIGN          91           SSL  \n","3577293       0.0  BENIGN          91           SSL  \n","3577294       0.0  BENIGN          91           SSL  \n","3577295       0.0  BENIGN          91           SSL  \n","\n","[3577296 rows x 87 columns]"],"text/html":["\n","  <div id=\"df-bc62a74c-3aa8-4a59-aa16-481c906c8aed\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Flow.ID</th>\n","      <th>Source.IP</th>\n","      <th>Source.Port</th>\n","      <th>Destination.IP</th>\n","      <th>Destination.Port</th>\n","      <th>Protocol</th>\n","      <th>Timestamp</th>\n","      <th>Flow.Duration</th>\n","      <th>Total.Fwd.Packets</th>\n","      <th>Total.Backward.Packets</th>\n","      <th>...</th>\n","      <th>Active.Std</th>\n","      <th>Active.Max</th>\n","      <th>Active.Min</th>\n","      <th>Idle.Mean</th>\n","      <th>Idle.Std</th>\n","      <th>Idle.Max</th>\n","      <th>Idle.Min</th>\n","      <th>Label</th>\n","      <th>L7Protocol</th>\n","      <th>ProtocolName</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>172.19.1.46-10.200.7.7-52422-3128-6</td>\n","      <td>172.19.1.46</td>\n","      <td>52422</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>45523</td>\n","      <td>22</td>\n","      <td>55</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>172.19.1.46-10.200.7.7-52422-3128-6</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>172.19.1.46</td>\n","      <td>52422</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>10.200.7.217-50.31.185.39-38848-80-6</td>\n","      <td>50.31.185.39</td>\n","      <td>80</td>\n","      <td>10.200.7.217</td>\n","      <td>38848</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>7</td>\n","      <td>HTTP</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>10.200.7.217-50.31.185.39-38848-80-6</td>\n","      <td>50.31.185.39</td>\n","      <td>80</td>\n","      <td>10.200.7.217</td>\n","      <td>38848</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>217</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>7</td>\n","      <td>HTTP</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>192.168.72.43-10.200.7.7-55961-3128-6</td>\n","      <td>192.168.72.43</td>\n","      <td>55961</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>78068</td>\n","      <td>5</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>3577291</th>\n","      <td>10.200.7.199-98.138.79.73-42135-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.199</td>\n","      <td>42135</td>\n","      <td>6</td>\n","      <td>15/05/201705:43:40</td>\n","      <td>2290821</td>\n","      <td>5</td>\n","      <td>4</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","    <tr>\n","      <th>3577292</th>\n","      <td>10.200.7.217-98.138.79.73-51546-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.217</td>\n","      <td>51546</td>\n","      <td>6</td>\n","      <td>15/05/201705:46:10</td>\n","      <td>24</td>\n","      <td>5</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","    <tr>\n","      <th>3577293</th>\n","      <td>10.200.7.218-98.138.79.73-44366-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.218</td>\n","      <td>44366</td>\n","      <td>6</td>\n","      <td>15/05/201705:45:39</td>\n","      <td>2591653</td>\n","      <td>6</td>\n","      <td>5</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","    <tr>\n","      <th>3577294</th>\n","      <td>10.200.7.195-98.138.79.73-52341-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.195</td>\n","      <td>52341</td>\n","      <td>6</td>\n","      <td>15/05/201705:45:59</td>\n","      <td>2622421</td>\n","      <td>4</td>\n","      <td>3</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","    <tr>\n","      <th>3577295</th>\n","      <td>10.200.7.196-98.138.79.73-34188-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.196</td>\n","      <td>34188</td>\n","      <td>6</td>\n","      <td>15/05/201705:46:05</td>\n","      <td>2009138</td>\n","      <td>3</td>\n","      <td>2</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>3577296 rows × 87 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-bc62a74c-3aa8-4a59-aa16-481c906c8aed')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-bc62a74c-3aa8-4a59-aa16-481c906c8aed button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-bc62a74c-3aa8-4a59-aa16-481c906c8aed');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":6}],"source":["df"]},{"cell_type":"markdown","source":["### Visualizar os Dados"],"metadata":{"id":"3rX1HPPjdQaz"}},{"cell_type":"code","execution_count":7,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"IU73oMoxNTEp","executionInfo":{"status":"ok","timestamp":1653643189326,"user_tz":180,"elapsed":242,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"f45c452f-744f-4862-9b38-3d316c7e4a0a"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["Index(['Flow.ID', 'Source.IP', 'Source.Port', 'Destination.IP',\n","       'Destination.Port', 'Protocol', 'Timestamp', 'Flow.Duration',\n","       'Total.Fwd.Packets', 'Total.Backward.Packets',\n","       'Total.Length.of.Fwd.Packets', 'Total.Length.of.Bwd.Packets',\n","       'Fwd.Packet.Length.Max', 'Fwd.Packet.Length.Min',\n","       'Fwd.Packet.Length.Mean', 'Fwd.Packet.Length.Std',\n","       'Bwd.Packet.Length.Max', 'Bwd.Packet.Length.Min',\n","       'Bwd.Packet.Length.Mean', 'Bwd.Packet.Length.Std', 'Flow.Bytes.s',\n","       'Flow.Packets.s', 'Flow.IAT.Mean', 'Flow.IAT.Std', 'Flow.IAT.Max',\n","       'Flow.IAT.Min', 'Fwd.IAT.Total', 'Fwd.IAT.Mean', 'Fwd.IAT.Std',\n","       'Fwd.IAT.Max', 'Fwd.IAT.Min', 'Bwd.IAT.Total', 'Bwd.IAT.Mean',\n","       'Bwd.IAT.Std', 'Bwd.IAT.Max', 'Bwd.IAT.Min', 'Fwd.PSH.Flags',\n","       'Bwd.PSH.Flags', 'Fwd.URG.Flags', 'Bwd.URG.Flags', 'Fwd.Header.Length',\n","       'Bwd.Header.Length', 'Fwd.Packets.s', 'Bwd.Packets.s',\n","       'Min.Packet.Length', 'Max.Packet.Length', 'Packet.Length.Mean',\n","       'Packet.Length.Std', 'Packet.Length.Variance', 'FIN.Flag.Count',\n","       'SYN.Flag.Count', 'RST.Flag.Count', 'PSH.Flag.Count', 'ACK.Flag.Count',\n","       'URG.Flag.Count', 'CWE.Flag.Count', 'ECE.Flag.Count', 'Down.Up.Ratio',\n","       'Average.Packet.Size', 'Avg.Fwd.Segment.Size', 'Avg.Bwd.Segment.Size',\n","       'Fwd.Header.Length.1', 'Fwd.Avg.Bytes.Bulk', 'Fwd.Avg.Packets.Bulk',\n","       'Fwd.Avg.Bulk.Rate', 'Bwd.Avg.Bytes.Bulk', 'Bwd.Avg.Packets.Bulk',\n","       'Bwd.Avg.Bulk.Rate', 'Subflow.Fwd.Packets', 'Subflow.Fwd.Bytes',\n","       'Subflow.Bwd.Packets', 'Subflow.Bwd.Bytes', 'Init_Win_bytes_forward',\n","       'Init_Win_bytes_backward', 'act_data_pkt_fwd', 'min_seg_size_forward',\n","       'Active.Mean', 'Active.Std', 'Active.Max', 'Active.Min', 'Idle.Mean',\n","       'Idle.Std', 'Idle.Max', 'Idle.Min', 'Label', 'L7Protocol',\n","       'ProtocolName'],\n","      dtype='object')"]},"metadata":{},"execution_count":7}],"source":["#view of columns \n","df.columns"]},{"cell_type":"code","source":["df.info()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"k31KzG7uYLp_","executionInfo":{"status":"ok","timestamp":1653643192189,"user_tz":180,"elapsed":256,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"0c8983cb-89ed-4516-8cf8-6471a661abc1"},"execution_count":8,"outputs":[{"output_type":"stream","name":"stdout","text":["<class 'pandas.core.frame.DataFrame'>\n","RangeIndex: 3577296 entries, 0 to 3577295\n","Data columns (total 87 columns):\n"," #   Column                       Dtype   \n","---  ------                       -----   \n"," 0   Flow.ID                      object  \n"," 1   Source.IP                    object  \n"," 2   Source.Port                  int64   \n"," 3   Destination.IP               object  \n"," 4   Destination.Port             int64   \n"," 5   Protocol                     int64   \n"," 6   Timestamp                    object  \n"," 7   Flow.Duration                int64   \n"," 8   Total.Fwd.Packets            int64   \n"," 9   Total.Backward.Packets       int64   \n"," 10  Total.Length.of.Fwd.Packets  int64   \n"," 11  Total.Length.of.Bwd.Packets  float64 \n"," 12  Fwd.Packet.Length.Max        int64   \n"," 13  Fwd.Packet.Length.Min        int64   \n"," 14  Fwd.Packet.Length.Mean       float64 \n"," 15  Fwd.Packet.Length.Std        float64 \n"," 16  Bwd.Packet.Length.Max        int64   \n"," 17  Bwd.Packet.Length.Min        int64   \n"," 18  Bwd.Packet.Length.Mean       float64 \n"," 19  Bwd.Packet.Length.Std        float64 \n"," 20  Flow.Bytes.s                 float64 \n"," 21  Flow.Packets.s               float64 \n"," 22  Flow.IAT.Mean                float64 \n"," 23  Flow.IAT.Std                 float64 \n"," 24  Flow.IAT.Max                 float64 \n"," 25  Flow.IAT.Min                 int64   \n"," 26  Fwd.IAT.Total                float64 \n"," 27  Fwd.IAT.Mean                 float64 \n"," 28  Fwd.IAT.Std                  float64 \n"," 29  Fwd.IAT.Max                  float64 \n"," 30  Fwd.IAT.Min                  float64 \n"," 31  Bwd.IAT.Total                float64 \n"," 32  Bwd.IAT.Mean                 float64 \n"," 33  Bwd.IAT.Std                  float64 \n"," 34  Bwd.IAT.Max                  float64 \n"," 35  Bwd.IAT.Min                  float64 \n"," 36  Fwd.PSH.Flags                int64   \n"," 37  Bwd.PSH.Flags                int64   \n"," 38  Fwd.URG.Flags                int64   \n"," 39  Bwd.URG.Flags                int64   \n"," 40  Fwd.Header.Length            int64   \n"," 41  Bwd.Header.Length            int64   \n"," 42  Fwd.Packets.s                float64 \n"," 43  Bwd.Packets.s                float64 \n"," 44  Min.Packet.Length            int64   \n"," 45  Max.Packet.Length            int64   \n"," 46  Packet.Length.Mean           float64 \n"," 47  Packet.Length.Std            float64 \n"," 48  Packet.Length.Variance       float64 \n"," 49  FIN.Flag.Count               int64   \n"," 50  SYN.Flag.Count               int64   \n"," 51  RST.Flag.Count               int64   \n"," 52  PSH.Flag.Count               int64   \n"," 53  ACK.Flag.Count               int64   \n"," 54  URG.Flag.Count               int64   \n"," 55  CWE.Flag.Count               int64   \n"," 56  ECE.Flag.Count               int64   \n"," 57  Down.Up.Ratio                int64   \n"," 58  Average.Packet.Size          float64 \n"," 59  Avg.Fwd.Segment.Size         float64 \n"," 60  Avg.Bwd.Segment.Size         float64 \n"," 61  Fwd.Header.Length.1          int64   \n"," 62  Fwd.Avg.Bytes.Bulk           int64   \n"," 63  Fwd.Avg.Packets.Bulk         int64   \n"," 64  Fwd.Avg.Bulk.Rate            int64   \n"," 65  Bwd.Avg.Bytes.Bulk           int64   \n"," 66  Bwd.Avg.Packets.Bulk         int64   \n"," 67  Bwd.Avg.Bulk.Rate            int64   \n"," 68  Subflow.Fwd.Packets          int64   \n"," 69  Subflow.Fwd.Bytes            int64   \n"," 70  Subflow.Bwd.Packets          int64   \n"," 71  Subflow.Bwd.Bytes            int64   \n"," 72  Init_Win_bytes_forward       int64   \n"," 73  Init_Win_bytes_backward      int64   \n"," 74  act_data_pkt_fwd             int64   \n"," 75  min_seg_size_forward         int64   \n"," 76  Active.Mean                  float64 \n"," 77  Active.Std                   float64 \n"," 78  Active.Max                   float64 \n"," 79  Active.Min                   float64 \n"," 80  Idle.Mean                    float64 \n"," 81  Idle.Std                     float64 \n"," 82  Idle.Max                     float64 \n"," 83  Idle.Min                     float64 \n"," 84  Label                        object  \n"," 85  L7Protocol                   int64   \n"," 86  ProtocolName                 category\n","dtypes: category(1), float64(36), int64(45), object(5)\n","memory usage: 2.3+ GB\n"]}]},{"cell_type":"code","source":["df.isna().sum()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"5FCLBetSXgUt","executionInfo":{"status":"ok","timestamp":1653643197631,"user_tz":180,"elapsed":1521,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"a3fe285b-af6d-4989-8c71-b407f6586ded"},"execution_count":9,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Flow.ID             0\n","Source.IP           0\n","Source.Port         0\n","Destination.IP      0\n","Destination.Port    0\n","                   ..\n","Idle.Max            0\n","Idle.Min            0\n","Label               0\n","L7Protocol          0\n","ProtocolName        0\n","Length: 87, dtype: int64"]},"metadata":{},"execution_count":9}]},{"cell_type":"code","source":["classes = df['ProtocolName'].value_counts()\n","classes = classes.head(10)\n","classes "],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"IoUyN1pVZfNe","executionInfo":{"status":"ok","timestamp":1653643200382,"user_tz":180,"elapsed":276,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"d5a16516-bf1d-4cb3-8401-2e1a3d0bc3d8"},"execution_count":10,"outputs":[{"output_type":"execute_result","data":{"text/plain":["GOOGLE            959110\n","HTTP              683734\n","HTTP_PROXY        623210\n","SSL               404883\n","HTTP_CONNECT      317526\n","YOUTUBE           170781\n","AMAZON             86875\n","MICROSOFT          54710\n","GMAIL              40260\n","WINDOWS_UPDATE     34471\n","Name: ProtocolName, dtype: int64"]},"metadata":{},"execution_count":10}]},{"cell_type":"markdown","source":["### Dataset Filtrado"],"metadata":{"id":"cwnRNQkndYyY"}},{"cell_type":"code","source":["values= ['GOOGLE','HTTP','HTTP_PROXY','SSL','HTTP_CONNECT','YOUTUBE','AMAZON','MICROSOFT','GMAIL','WINDOWS_UPDATE']\n","filtered_df = df[df.ProtocolName.isin(values)]\n","filtered_df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":835},"id":"Uly29sS5dQFF","executionInfo":{"status":"ok","timestamp":1653643207366,"user_tz":180,"elapsed":3410,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"c70cbd1f-d511-41bc-e383-9fd2a51b6594"},"execution_count":11,"outputs":[{"output_type":"execute_result","data":{"text/plain":["                                       Flow.ID      Source.IP  Source.Port  \\\n","0          172.19.1.46-10.200.7.7-52422-3128-6    172.19.1.46        52422   \n","1          172.19.1.46-10.200.7.7-52422-3128-6     10.200.7.7         3128   \n","2         10.200.7.217-50.31.185.39-38848-80-6   50.31.185.39           80   \n","3         10.200.7.217-50.31.185.39-38848-80-6   50.31.185.39           80   \n","4        192.168.72.43-10.200.7.7-55961-3128-6  192.168.72.43        55961   \n","...                                        ...            ...          ...   \n","3577291  10.200.7.199-98.138.79.73-42135-443-6   98.138.79.73          443   \n","3577292  10.200.7.217-98.138.79.73-51546-443-6   98.138.79.73          443   \n","3577293  10.200.7.218-98.138.79.73-44366-443-6   98.138.79.73          443   \n","3577294  10.200.7.195-98.138.79.73-52341-443-6   98.138.79.73          443   \n","3577295  10.200.7.196-98.138.79.73-34188-443-6   98.138.79.73          443   \n","\n","        Destination.IP  Destination.Port  Protocol           Timestamp  \\\n","0           10.200.7.7              3128         6  26/04/201711:11:17   \n","1          172.19.1.46             52422         6  26/04/201711:11:17   \n","2         10.200.7.217             38848         6  26/04/201711:11:17   \n","3         10.200.7.217             38848         6  26/04/201711:11:17   \n","4           10.200.7.7              3128         6  26/04/201711:11:17   \n","...                ...               ...       ...                 ...   \n","3577291   10.200.7.199             42135         6  15/05/201705:43:40   \n","3577292   10.200.7.217             51546         6  15/05/201705:46:10   \n","3577293   10.200.7.218             44366         6  15/05/201705:45:39   \n","3577294   10.200.7.195             52341         6  15/05/201705:45:59   \n","3577295   10.200.7.196             34188         6  15/05/201705:46:05   \n","\n","         Flow.Duration  Total.Fwd.Packets  Total.Backward.Packets  ...  \\\n","0                45523                 22                      55  ...   \n","1                    1                  2                       0  ...   \n","2                    1                  3                       0  ...   \n","3                  217                  1                       3  ...   \n","4                78068                  5                       0  ...   \n","...                ...                ...                     ...  ...   \n","3577291        2290821                  5                       4  ...   \n","3577292             24                  5                       0  ...   \n","3577293        2591653                  6                       5  ...   \n","3577294        2622421                  4                       3  ...   \n","3577295        2009138                  3                       2  ...   \n","\n","         Active.Std  Active.Max  Active.Min  Idle.Mean  Idle.Std  Idle.Max  \\\n","0               0.0         0.0         0.0        0.0       0.0       0.0   \n","1               0.0         0.0         0.0        0.0       0.0       0.0   \n","2               0.0         0.0         0.0        0.0       0.0       0.0   \n","3               0.0         0.0         0.0        0.0       0.0       0.0   \n","4               0.0         0.0         0.0        0.0       0.0       0.0   \n","...             ...         ...         ...        ...       ...       ...   \n","3577291         0.0         0.0         0.0        0.0       0.0       0.0   \n","3577292         0.0         0.0         0.0        0.0       0.0       0.0   \n","3577293         0.0         0.0         0.0        0.0       0.0       0.0   \n","3577294         0.0         0.0         0.0        0.0       0.0       0.0   \n","3577295         0.0         0.0         0.0        0.0       0.0       0.0   \n","\n","         Idle.Min   Label  L7Protocol  ProtocolName  \n","0             0.0  BENIGN         131    HTTP_PROXY  \n","1             0.0  BENIGN         131    HTTP_PROXY  \n","2             0.0  BENIGN           7          HTTP  \n","3             0.0  BENIGN           7          HTTP  \n","4             0.0  BENIGN         131    HTTP_PROXY  \n","...           ...     ...         ...           ...  \n","3577291       0.0  BENIGN          91           SSL  \n","3577292       0.0  BENIGN          91           SSL  \n","3577293       0.0  BENIGN          91           SSL  \n","3577294       0.0  BENIGN          91           SSL  \n","3577295       0.0  BENIGN          91           SSL  \n","\n","[3375560 rows x 87 columns]"],"text/html":["\n","  <div id=\"df-56afb51b-b0e1-4755-a15d-ccc129508bfe\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Flow.ID</th>\n","      <th>Source.IP</th>\n","      <th>Source.Port</th>\n","      <th>Destination.IP</th>\n","      <th>Destination.Port</th>\n","      <th>Protocol</th>\n","      <th>Timestamp</th>\n","      <th>Flow.Duration</th>\n","      <th>Total.Fwd.Packets</th>\n","      <th>Total.Backward.Packets</th>\n","      <th>...</th>\n","      <th>Active.Std</th>\n","      <th>Active.Max</th>\n","      <th>Active.Min</th>\n","      <th>Idle.Mean</th>\n","      <th>Idle.Std</th>\n","      <th>Idle.Max</th>\n","      <th>Idle.Min</th>\n","      <th>Label</th>\n","      <th>L7Protocol</th>\n","      <th>ProtocolName</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>172.19.1.46-10.200.7.7-52422-3128-6</td>\n","      <td>172.19.1.46</td>\n","      <td>52422</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>45523</td>\n","      <td>22</td>\n","      <td>55</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>172.19.1.46-10.200.7.7-52422-3128-6</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>172.19.1.46</td>\n","      <td>52422</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>10.200.7.217-50.31.185.39-38848-80-6</td>\n","      <td>50.31.185.39</td>\n","      <td>80</td>\n","      <td>10.200.7.217</td>\n","      <td>38848</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>7</td>\n","      <td>HTTP</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>10.200.7.217-50.31.185.39-38848-80-6</td>\n","      <td>50.31.185.39</td>\n","      <td>80</td>\n","      <td>10.200.7.217</td>\n","      <td>38848</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>217</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>7</td>\n","      <td>HTTP</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>192.168.72.43-10.200.7.7-55961-3128-6</td>\n","      <td>192.168.72.43</td>\n","      <td>55961</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>78068</td>\n","      <td>5</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>3577291</th>\n","      <td>10.200.7.199-98.138.79.73-42135-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.199</td>\n","      <td>42135</td>\n","      <td>6</td>\n","      <td>15/05/201705:43:40</td>\n","      <td>2290821</td>\n","      <td>5</td>\n","      <td>4</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","    <tr>\n","      <th>3577292</th>\n","      <td>10.200.7.217-98.138.79.73-51546-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.217</td>\n","      <td>51546</td>\n","      <td>6</td>\n","      <td>15/05/201705:46:10</td>\n","      <td>24</td>\n","      <td>5</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","    <tr>\n","      <th>3577293</th>\n","      <td>10.200.7.218-98.138.79.73-44366-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.218</td>\n","      <td>44366</td>\n","      <td>6</td>\n","      <td>15/05/201705:45:39</td>\n","      <td>2591653</td>\n","      <td>6</td>\n","      <td>5</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","    <tr>\n","      <th>3577294</th>\n","      <td>10.200.7.195-98.138.79.73-52341-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.195</td>\n","      <td>52341</td>\n","      <td>6</td>\n","      <td>15/05/201705:45:59</td>\n","      <td>2622421</td>\n","      <td>4</td>\n","      <td>3</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","    <tr>\n","      <th>3577295</th>\n","      <td>10.200.7.196-98.138.79.73-34188-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.196</td>\n","      <td>34188</td>\n","      <td>6</td>\n","      <td>15/05/201705:46:05</td>\n","      <td>2009138</td>\n","      <td>3</td>\n","      <td>2</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>3375560 rows × 87 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-56afb51b-b0e1-4755-a15d-ccc129508bfe')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-56afb51b-b0e1-4755-a15d-ccc129508bfe button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-56afb51b-b0e1-4755-a15d-ccc129508bfe');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":11}]},{"cell_type":"code","source":["classes = filtered_df['ProtocolName'].value_counts()\n","classes = classes.head(10)\n","classes "],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"68Skn0P5k639","executionInfo":{"status":"ok","timestamp":1653643210832,"user_tz":180,"elapsed":229,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"7d3fc67e-17b6-4982-8e90-f6d1a4863b92"},"execution_count":12,"outputs":[{"output_type":"execute_result","data":{"text/plain":["GOOGLE            959110\n","HTTP              683734\n","HTTP_PROXY        623210\n","SSL               404883\n","HTTP_CONNECT      317526\n","YOUTUBE           170781\n","AMAZON             86875\n","MICROSOFT          54710\n","GMAIL              40260\n","WINDOWS_UPDATE     34471\n","Name: ProtocolName, dtype: int64"]},"metadata":{},"execution_count":12}]},{"cell_type":"markdown","source":["#### Ajustar as Classes"],"metadata":{"id":"QNdD_bOxxWI4"}},{"cell_type":"code","source":["filtered_df['classes'] = filtered_df['ProtocolName'].cat.codes"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"TzmZ7MmRYaeb","executionInfo":{"status":"ok","timestamp":1653643216062,"user_tz":180,"elapsed":251,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"8ea34a31-2a41-44f6-8abb-cb87f90b0f56"},"execution_count":13,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n","A value is trying to be set on a copy of a slice from a DataFrame.\n","Try using .loc[row_indexer,col_indexer] = value instead\n","\n","See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n","  \"\"\"Entry point for launching an IPython kernel.\n"]}]},{"cell_type":"code","source":["filtered_df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":835},"id":"sRIT86DwhqF1","executionInfo":{"status":"ok","timestamp":1653643219824,"user_tz":180,"elapsed":1360,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"88c00780-3344-4f64-8f44-ebaf311cfcdd"},"execution_count":14,"outputs":[{"output_type":"execute_result","data":{"text/plain":["                                       Flow.ID      Source.IP  Source.Port  \\\n","0          172.19.1.46-10.200.7.7-52422-3128-6    172.19.1.46        52422   \n","1          172.19.1.46-10.200.7.7-52422-3128-6     10.200.7.7         3128   \n","2         10.200.7.217-50.31.185.39-38848-80-6   50.31.185.39           80   \n","3         10.200.7.217-50.31.185.39-38848-80-6   50.31.185.39           80   \n","4        192.168.72.43-10.200.7.7-55961-3128-6  192.168.72.43        55961   \n","...                                        ...            ...          ...   \n","3577291  10.200.7.199-98.138.79.73-42135-443-6   98.138.79.73          443   \n","3577292  10.200.7.217-98.138.79.73-51546-443-6   98.138.79.73          443   \n","3577293  10.200.7.218-98.138.79.73-44366-443-6   98.138.79.73          443   \n","3577294  10.200.7.195-98.138.79.73-52341-443-6   98.138.79.73          443   \n","3577295  10.200.7.196-98.138.79.73-34188-443-6   98.138.79.73          443   \n","\n","        Destination.IP  Destination.Port  Protocol           Timestamp  \\\n","0           10.200.7.7              3128         6  26/04/201711:11:17   \n","1          172.19.1.46             52422         6  26/04/201711:11:17   \n","2         10.200.7.217             38848         6  26/04/201711:11:17   \n","3         10.200.7.217             38848         6  26/04/201711:11:17   \n","4           10.200.7.7              3128         6  26/04/201711:11:17   \n","...                ...               ...       ...                 ...   \n","3577291   10.200.7.199             42135         6  15/05/201705:43:40   \n","3577292   10.200.7.217             51546         6  15/05/201705:46:10   \n","3577293   10.200.7.218             44366         6  15/05/201705:45:39   \n","3577294   10.200.7.195             52341         6  15/05/201705:45:59   \n","3577295   10.200.7.196             34188         6  15/05/201705:46:05   \n","\n","         Flow.Duration  Total.Fwd.Packets  Total.Backward.Packets  ...  \\\n","0                45523                 22                      55  ...   \n","1                    1                  2                       0  ...   \n","2                    1                  3                       0  ...   \n","3                  217                  1                       3  ...   \n","4                78068                  5                       0  ...   \n","...                ...                ...                     ...  ...   \n","3577291        2290821                  5                       4  ...   \n","3577292             24                  5                       0  ...   \n","3577293        2591653                  6                       5  ...   \n","3577294        2622421                  4                       3  ...   \n","3577295        2009138                  3                       2  ...   \n","\n","         Active.Max  Active.Min  Idle.Mean  Idle.Std  Idle.Max  Idle.Min  \\\n","0               0.0         0.0        0.0       0.0       0.0       0.0   \n","1               0.0         0.0        0.0       0.0       0.0       0.0   \n","2               0.0         0.0        0.0       0.0       0.0       0.0   \n","3               0.0         0.0        0.0       0.0       0.0       0.0   \n","4               0.0         0.0        0.0       0.0       0.0       0.0   \n","...             ...         ...        ...       ...       ...       ...   \n","3577291         0.0         0.0        0.0       0.0       0.0       0.0   \n","3577292         0.0         0.0        0.0       0.0       0.0       0.0   \n","3577293         0.0         0.0        0.0       0.0       0.0       0.0   \n","3577294         0.0         0.0        0.0       0.0       0.0       0.0   \n","3577295         0.0         0.0        0.0       0.0       0.0       0.0   \n","\n","          Label  L7Protocol  ProtocolName  classes  \n","0        BENIGN         131    HTTP_PROXY        9  \n","1        BENIGN         131    HTTP_PROXY        9  \n","2        BENIGN           7          HTTP        7  \n","3        BENIGN           7          HTTP        7  \n","4        BENIGN         131    HTTP_PROXY        9  \n","...         ...         ...           ...      ...  \n","3577291  BENIGN          91           SSL       15  \n","3577292  BENIGN          91           SSL       15  \n","3577293  BENIGN          91           SSL       15  \n","3577294  BENIGN          91           SSL       15  \n","3577295  BENIGN          91           SSL       15  \n","\n","[3375560 rows x 88 columns]"],"text/html":["\n","  <div id=\"df-d8cf6e3e-c590-4947-980f-81386b5a8563\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Flow.ID</th>\n","      <th>Source.IP</th>\n","      <th>Source.Port</th>\n","      <th>Destination.IP</th>\n","      <th>Destination.Port</th>\n","      <th>Protocol</th>\n","      <th>Timestamp</th>\n","      <th>Flow.Duration</th>\n","      <th>Total.Fwd.Packets</th>\n","      <th>Total.Backward.Packets</th>\n","      <th>...</th>\n","      <th>Active.Max</th>\n","      <th>Active.Min</th>\n","      <th>Idle.Mean</th>\n","      <th>Idle.Std</th>\n","      <th>Idle.Max</th>\n","      <th>Idle.Min</th>\n","      <th>Label</th>\n","      <th>L7Protocol</th>\n","      <th>ProtocolName</th>\n","      <th>classes</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>172.19.1.46-10.200.7.7-52422-3128-6</td>\n","      <td>172.19.1.46</td>\n","      <td>52422</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>45523</td>\n","      <td>22</td>\n","      <td>55</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","      <td>9</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>172.19.1.46-10.200.7.7-52422-3128-6</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>172.19.1.46</td>\n","      <td>52422</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","      <td>9</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>10.200.7.217-50.31.185.39-38848-80-6</td>\n","      <td>50.31.185.39</td>\n","      <td>80</td>\n","      <td>10.200.7.217</td>\n","      <td>38848</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>7</td>\n","      <td>HTTP</td>\n","      <td>7</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>10.200.7.217-50.31.185.39-38848-80-6</td>\n","      <td>50.31.185.39</td>\n","      <td>80</td>\n","      <td>10.200.7.217</td>\n","      <td>38848</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>217</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>7</td>\n","      <td>HTTP</td>\n","      <td>7</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>192.168.72.43-10.200.7.7-55961-3128-6</td>\n","      <td>192.168.72.43</td>\n","      <td>55961</td>\n","      <td>10.200.7.7</td>\n","      <td>3128</td>\n","      <td>6</td>\n","      <td>26/04/201711:11:17</td>\n","      <td>78068</td>\n","      <td>5</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>131</td>\n","      <td>HTTP_PROXY</td>\n","      <td>9</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>3577291</th>\n","      <td>10.200.7.199-98.138.79.73-42135-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.199</td>\n","      <td>42135</td>\n","      <td>6</td>\n","      <td>15/05/201705:43:40</td>\n","      <td>2290821</td>\n","      <td>5</td>\n","      <td>4</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","      <td>15</td>\n","    </tr>\n","    <tr>\n","      <th>3577292</th>\n","      <td>10.200.7.217-98.138.79.73-51546-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.217</td>\n","      <td>51546</td>\n","      <td>6</td>\n","      <td>15/05/201705:46:10</td>\n","      <td>24</td>\n","      <td>5</td>\n","      <td>0</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","      <td>15</td>\n","    </tr>\n","    <tr>\n","      <th>3577293</th>\n","      <td>10.200.7.218-98.138.79.73-44366-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.218</td>\n","      <td>44366</td>\n","      <td>6</td>\n","      <td>15/05/201705:45:39</td>\n","      <td>2591653</td>\n","      <td>6</td>\n","      <td>5</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","      <td>15</td>\n","    </tr>\n","    <tr>\n","      <th>3577294</th>\n","      <td>10.200.7.195-98.138.79.73-52341-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.195</td>\n","      <td>52341</td>\n","      <td>6</td>\n","      <td>15/05/201705:45:59</td>\n","      <td>2622421</td>\n","      <td>4</td>\n","      <td>3</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","      <td>15</td>\n","    </tr>\n","    <tr>\n","      <th>3577295</th>\n","      <td>10.200.7.196-98.138.79.73-34188-443-6</td>\n","      <td>98.138.79.73</td>\n","      <td>443</td>\n","      <td>10.200.7.196</td>\n","      <td>34188</td>\n","      <td>6</td>\n","      <td>15/05/201705:46:05</td>\n","      <td>2009138</td>\n","      <td>3</td>\n","      <td>2</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>BENIGN</td>\n","      <td>91</td>\n","      <td>SSL</td>\n","      <td>15</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>3375560 rows × 88 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d8cf6e3e-c590-4947-980f-81386b5a8563')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-d8cf6e3e-c590-4947-980f-81386b5a8563 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-d8cf6e3e-c590-4947-980f-81386b5a8563');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":14}]},{"cell_type":"code","source":["classes = filtered_df['classes'].value_counts()\n","classes = classes.head(10)\n","classes "],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"yzcGtuLhleBI","executionInfo":{"status":"ok","timestamp":1653643223207,"user_tz":180,"elapsed":247,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"7b2551db-3f38-4c3a-da9e-182ac29ee9aa"},"execution_count":15,"outputs":[{"output_type":"execute_result","data":{"text/plain":["6     959110\n","7     683734\n","9     623210\n","15    404883\n","8     317526\n","21    170781\n","0      86875\n","10     54710\n","5      40260\n","19     34471\n","Name: classes, dtype: int64"]},"metadata":{},"execution_count":15}]},{"cell_type":"code","source":["dict_classes = {6: 'GOOGLE',\n","                7: 'HTTP',\n","                9: 'HTTP_PROXY',\n","                15: 'SSL',\n","                8: 'HTTP_CONNECT',\n","                21: 'YOUTUBE',\n","                0: 'AMAZON',\n","                10: 'MICROSOFT',\n","                5: 'GMAIL',\n","                19: 'WINDOWS_UPDATE'}"],"metadata":{"id":"uUcVuI-jlyym","executionInfo":{"status":"ok","timestamp":1653643225899,"user_tz":180,"elapsed":234,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":16,"outputs":[]},{"cell_type":"markdown","source":["#### Salvar Dataset"],"metadata":{"id":"OtHgRxizdeaN"}},{"cell_type":"code","source":["filtered_df.to_csv(path_dataset_filtered,index=False)"],"metadata":{"id":"tpLV8jdssxaV","executionInfo":{"status":"ok","timestamp":1653643414666,"user_tz":180,"elapsed":186181,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":17,"outputs":[]},{"cell_type":"markdown","source":["#### Carregar Dataset Filtrado"],"metadata":{"id":"uk1kXGqmdoQh"}},{"cell_type":"code","source":["filtered_df = pd.read_csv(path_dataset_filtered, sep=',',nrows=10000,dtype={'ProtocolName':'category'})"],"metadata":{"id":"HnLnUir_Ahe6","executionInfo":{"status":"ok","timestamp":1653643441930,"user_tz":180,"elapsed":705,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":18,"outputs":[]},{"cell_type":"markdown","source":["#### Visualizar  Dataset Filtrado"],"metadata":{"id":"QvF7se2Xdsh5"}},{"cell_type":"code","source":["filtered_df.columns"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"qb53PkOiuhhb","executionInfo":{"status":"ok","timestamp":1653643447091,"user_tz":180,"elapsed":243,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"9c042772-8a4b-4f68-9b93-27cca7f31d7d"},"execution_count":19,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Index(['Flow.ID', 'Source.IP', 'Source.Port', 'Destination.IP',\n","       'Destination.Port', 'Protocol', 'Timestamp', 'Flow.Duration',\n","       'Total.Fwd.Packets', 'Total.Backward.Packets',\n","       'Total.Length.of.Fwd.Packets', 'Total.Length.of.Bwd.Packets',\n","       'Fwd.Packet.Length.Max', 'Fwd.Packet.Length.Min',\n","       'Fwd.Packet.Length.Mean', 'Fwd.Packet.Length.Std',\n","       'Bwd.Packet.Length.Max', 'Bwd.Packet.Length.Min',\n","       'Bwd.Packet.Length.Mean', 'Bwd.Packet.Length.Std', 'Flow.Bytes.s',\n","       'Flow.Packets.s', 'Flow.IAT.Mean', 'Flow.IAT.Std', 'Flow.IAT.Max',\n","       'Flow.IAT.Min', 'Fwd.IAT.Total', 'Fwd.IAT.Mean', 'Fwd.IAT.Std',\n","       'Fwd.IAT.Max', 'Fwd.IAT.Min', 'Bwd.IAT.Total', 'Bwd.IAT.Mean',\n","       'Bwd.IAT.Std', 'Bwd.IAT.Max', 'Bwd.IAT.Min', 'Fwd.PSH.Flags',\n","       'Bwd.PSH.Flags', 'Fwd.URG.Flags', 'Bwd.URG.Flags', 'Fwd.Header.Length',\n","       'Bwd.Header.Length', 'Fwd.Packets.s', 'Bwd.Packets.s',\n","       'Min.Packet.Length', 'Max.Packet.Length', 'Packet.Length.Mean',\n","       'Packet.Length.Std', 'Packet.Length.Variance', 'FIN.Flag.Count',\n","       'SYN.Flag.Count', 'RST.Flag.Count', 'PSH.Flag.Count', 'ACK.Flag.Count',\n","       'URG.Flag.Count', 'CWE.Flag.Count', 'ECE.Flag.Count', 'Down.Up.Ratio',\n","       'Average.Packet.Size', 'Avg.Fwd.Segment.Size', 'Avg.Bwd.Segment.Size',\n","       'Fwd.Header.Length.1', 'Fwd.Avg.Bytes.Bulk', 'Fwd.Avg.Packets.Bulk',\n","       'Fwd.Avg.Bulk.Rate', 'Bwd.Avg.Bytes.Bulk', 'Bwd.Avg.Packets.Bulk',\n","       'Bwd.Avg.Bulk.Rate', 'Subflow.Fwd.Packets', 'Subflow.Fwd.Bytes',\n","       'Subflow.Bwd.Packets', 'Subflow.Bwd.Bytes', 'Init_Win_bytes_forward',\n","       'Init_Win_bytes_backward', 'act_data_pkt_fwd', 'min_seg_size_forward',\n","       'Active.Mean', 'Active.Std', 'Active.Max', 'Active.Min', 'Idle.Mean',\n","       'Idle.Std', 'Idle.Max', 'Idle.Min', 'Label', 'L7Protocol',\n","       'ProtocolName', 'classes'],\n","      dtype='object')"]},"metadata":{},"execution_count":19}]},{"cell_type":"markdown","source":["## Definição das Classes"],"metadata":{"id":"Uy_vgiY9xt3I"}},{"cell_type":"markdown","source":["### Dataset Análise"],"metadata":{"id":"HgDfZfICdzjR"}},{"cell_type":"code","source":["cols = ['Flow.Duration','Flow.Bytes.s','Flow.Packets.s', 'Flow.IAT.Mean', 'Flow.IAT.Std', 'Flow.IAT.Max',\n","       'Flow.IAT.Min','Total.Fwd.Packets','Total.Backward.Packets','Total.Length.of.Fwd.Packets','Total.Length.of.Bwd.Packets','classes']"],"metadata":{"id":"vp6z1jPvt6uJ","executionInfo":{"status":"ok","timestamp":1653643450605,"user_tz":180,"elapsed":236,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":20,"outputs":[]},{"cell_type":"code","source":["df_analise = filtered_df[cols]"],"metadata":{"id":"tnS210aGvHaf","executionInfo":{"status":"ok","timestamp":1653643452693,"user_tz":180,"elapsed":235,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":21,"outputs":[]},{"cell_type":"code","source":["df_analise"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":488},"id":"FeLkD2zKvsak","executionInfo":{"status":"ok","timestamp":1653643454432,"user_tz":180,"elapsed":264,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"40afbd28-5519-4c1a-de8c-c06bef07b1ba"},"execution_count":22,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      Flow.Duration  Flow.Bytes.s  Flow.Packets.s  Flow.IAT.Mean  \\\n","0             45523  2.428355e+06    1.691453e+03     598.986842   \n","1                 1  1.200000e+07    2.000000e+06       1.000000   \n","2                 1  6.740000e+08    3.000000e+06       0.500000   \n","3               217  0.000000e+00    1.843318e+04      72.333333   \n","4             78068  1.378286e+04    6.404673e+01   19517.000000   \n","...             ...           ...             ...            ...   \n","9995              1  3.240000e+08    2.000000e+06       1.000000   \n","9996            417  2.877698e+04    4.796163e+03     417.000000   \n","9997       16500339  6.361081e+02    1.696935e+00  611123.666667   \n","9998        8051824  3.648987e+03    1.751156e+01   57513.028571   \n","9999        8038513  7.263781e+02    5.846853e+00  174750.282609   \n","\n","      Flow.IAT.Std  Flow.IAT.Max  Flow.IAT.Min  Total.Fwd.Packets  \\\n","0     8.160613e+02        3880.0             1                 22   \n","1     0.000000e+00           1.0             1                  2   \n","2     7.071068e-01           1.0             0                  3   \n","3     6.266046e+01         110.0             0                  1   \n","4     2.575850e+04       54313.0             0                  5   \n","...            ...           ...           ...                ...   \n","9995  0.000000e+00           1.0             1                  2   \n","9996  0.000000e+00         417.0           417                  2   \n","9997  2.161468e+06     9954935.0             0                 14   \n","9998  3.359510e+05     2723322.0             0                 21   \n","9999  5.357455e+05     2582774.0             0                 24   \n","\n","      Total.Backward.Packets  Total.Length.of.Fwd.Packets  \\\n","0                         55                          132   \n","1                          0                           12   \n","2                          0                          674   \n","3                          3                            0   \n","4                          0                         1076   \n","...                      ...                          ...   \n","9995                       0                          324   \n","9996                       0                           12   \n","9997                      14                         7874   \n","9998                     120                         2538   \n","9999                      23                         2288   \n","\n","      Total.Length.of.Bwd.Packets  classes  \n","0                        110414.0        9  \n","1                             0.0        9  \n","2                             0.0        7  \n","3                             0.0        7  \n","4                             0.0        9  \n","...                           ...      ...  \n","9995                          0.0        7  \n","9996                          0.0        7  \n","9997                       2622.0        7  \n","9998                      26843.0        6  \n","9999                       3551.0        6  \n","\n","[10000 rows x 12 columns]"],"text/html":["\n","  <div id=\"df-d2f8cd1c-cf84-4946-91a8-10871b416c9e\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Flow.Duration</th>\n","      <th>Flow.Bytes.s</th>\n","      <th>Flow.Packets.s</th>\n","      <th>Flow.IAT.Mean</th>\n","      <th>Flow.IAT.Std</th>\n","      <th>Flow.IAT.Max</th>\n","      <th>Flow.IAT.Min</th>\n","      <th>Total.Fwd.Packets</th>\n","      <th>Total.Backward.Packets</th>\n","      <th>Total.Length.of.Fwd.Packets</th>\n","      <th>Total.Length.of.Bwd.Packets</th>\n","      <th>classes</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>45523</td>\n","      <td>2.428355e+06</td>\n","      <td>1.691453e+03</td>\n","      <td>598.986842</td>\n","      <td>8.160613e+02</td>\n","      <td>3880.0</td>\n","      <td>1</td>\n","      <td>22</td>\n","      <td>55</td>\n","      <td>132</td>\n","      <td>110414.0</td>\n","      <td>9</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>1</td>\n","      <td>1.200000e+07</td>\n","      <td>2.000000e+06</td>\n","      <td>1.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>1.0</td>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>12</td>\n","      <td>0.0</td>\n","      <td>9</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>1</td>\n","      <td>6.740000e+08</td>\n","      <td>3.000000e+06</td>\n","      <td>0.500000</td>\n","      <td>7.071068e-01</td>\n","      <td>1.0</td>\n","      <td>0</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>674</td>\n","      <td>0.0</td>\n","      <td>7</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>217</td>\n","      <td>0.000000e+00</td>\n","      <td>1.843318e+04</td>\n","      <td>72.333333</td>\n","      <td>6.266046e+01</td>\n","      <td>110.0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>0.0</td>\n","      <td>7</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>78068</td>\n","      <td>1.378286e+04</td>\n","      <td>6.404673e+01</td>\n","      <td>19517.000000</td>\n","      <td>2.575850e+04</td>\n","      <td>54313.0</td>\n","      <td>0</td>\n","      <td>5</td>\n","      <td>0</td>\n","      <td>1076</td>\n","      <td>0.0</td>\n","      <td>9</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>9995</th>\n","      <td>1</td>\n","      <td>3.240000e+08</td>\n","      <td>2.000000e+06</td>\n","      <td>1.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>1.0</td>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>324</td>\n","      <td>0.0</td>\n","      <td>7</td>\n","    </tr>\n","    <tr>\n","      <th>9996</th>\n","      <td>417</td>\n","      <td>2.877698e+04</td>\n","      <td>4.796163e+03</td>\n","      <td>417.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>417.0</td>\n","      <td>417</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>12</td>\n","      <td>0.0</td>\n","      <td>7</td>\n","    </tr>\n","    <tr>\n","      <th>9997</th>\n","      <td>16500339</td>\n","      <td>6.361081e+02</td>\n","      <td>1.696935e+00</td>\n","      <td>611123.666667</td>\n","      <td>2.161468e+06</td>\n","      <td>9954935.0</td>\n","      <td>0</td>\n","      <td>14</td>\n","      <td>14</td>\n","      <td>7874</td>\n","      <td>2622.0</td>\n","      <td>7</td>\n","    </tr>\n","    <tr>\n","      <th>9998</th>\n","      <td>8051824</td>\n","      <td>3.648987e+03</td>\n","      <td>1.751156e+01</td>\n","      <td>57513.028571</td>\n","      <td>3.359510e+05</td>\n","      <td>2723322.0</td>\n","      <td>0</td>\n","      <td>21</td>\n","      <td>120</td>\n","      <td>2538</td>\n","      <td>26843.0</td>\n","      <td>6</td>\n","    </tr>\n","    <tr>\n","      <th>9999</th>\n","      <td>8038513</td>\n","      <td>7.263781e+02</td>\n","      <td>5.846853e+00</td>\n","      <td>174750.282609</td>\n","      <td>5.357455e+05</td>\n","      <td>2582774.0</td>\n","      <td>0</td>\n","      <td>24</td>\n","      <td>23</td>\n","      <td>2288</td>\n","      <td>3551.0</td>\n","      <td>6</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>10000 rows × 12 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d2f8cd1c-cf84-4946-91a8-10871b416c9e')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-d2f8cd1c-cf84-4946-91a8-10871b416c9e button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-d2f8cd1c-cf84-4946-91a8-10871b416c9e');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":22}]},{"cell_type":"markdown","source":["### Criar conjunto alvo"],"metadata":{"id":"O1lYoLQKd5te"}},{"cell_type":"code","source":["classes_analise = df_analise['classes'].value_counts()\n","classes_analise "],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"t4b2oKbMsz-v","executionInfo":{"status":"ok","timestamp":1653643457548,"user_tz":180,"elapsed":318,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"d872715a-d98c-4bb6-aceb-60ba923ba1f3"},"execution_count":23,"outputs":[{"output_type":"execute_result","data":{"text/plain":["7     3848\n","6     1599\n","9     1281\n","15    1091\n","8     1022\n","21     699\n","19     160\n","10     137\n","0      109\n","5       54\n","Name: classes, dtype: int64"]},"metadata":{},"execution_count":23}]},{"cell_type":"code","execution_count":24,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"cU9ME_MrNTEr","executionInfo":{"status":"ok","timestamp":1653643461110,"user_tz":180,"elapsed":243,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"d91a9c08-8634-4cb7-c5a3-f80e185f4a5b"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["0    9\n","1    9\n","2    7\n","3    7\n","4    9\n","Name: classes, dtype: int64"]},"metadata":{},"execution_count":24}],"source":["#create Target Feature\n","y = df_analise['classes']\n","#view of shares\n","y.head()"]},{"cell_type":"code","execution_count":25,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"C5x3ZkI-NTEq","executionInfo":{"status":"ok","timestamp":1653643463492,"user_tz":180,"elapsed":257,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"f3afb346-94b7-4812-9885-d39d6dde11f7"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["count    10000.000000\n","mean         9.195900\n","std          4.429479\n","min          0.000000\n","25%          7.000000\n","50%          7.000000\n","75%          9.000000\n","max         21.000000\n","Name: classes, dtype: float64"]},"metadata":{},"execution_count":25}],"source":["y.describe()"]},{"cell_type":"markdown","source":["### Criar Atributos Preditivos"],"metadata":{"id":"_Pd7TtZUd9OJ"}},{"cell_type":"code","source":["x = df_analise.iloc[:,0:11]"],"metadata":{"id":"A6VRFT2Nv908","executionInfo":{"status":"ok","timestamp":1653643466043,"user_tz":180,"elapsed":239,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":26,"outputs":[]},{"cell_type":"code","source":["x"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":488},"id":"-kIlr9G3wF02","executionInfo":{"status":"ok","timestamp":1653643467831,"user_tz":180,"elapsed":304,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"1f8fce03-79a0-4654-dbaa-82a930c63012"},"execution_count":27,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      Flow.Duration  Flow.Bytes.s  Flow.Packets.s  Flow.IAT.Mean  \\\n","0             45523  2.428355e+06    1.691453e+03     598.986842   \n","1                 1  1.200000e+07    2.000000e+06       1.000000   \n","2                 1  6.740000e+08    3.000000e+06       0.500000   \n","3               217  0.000000e+00    1.843318e+04      72.333333   \n","4             78068  1.378286e+04    6.404673e+01   19517.000000   \n","...             ...           ...             ...            ...   \n","9995              1  3.240000e+08    2.000000e+06       1.000000   \n","9996            417  2.877698e+04    4.796163e+03     417.000000   \n","9997       16500339  6.361081e+02    1.696935e+00  611123.666667   \n","9998        8051824  3.648987e+03    1.751156e+01   57513.028571   \n","9999        8038513  7.263781e+02    5.846853e+00  174750.282609   \n","\n","      Flow.IAT.Std  Flow.IAT.Max  Flow.IAT.Min  Total.Fwd.Packets  \\\n","0     8.160613e+02        3880.0             1                 22   \n","1     0.000000e+00           1.0             1                  2   \n","2     7.071068e-01           1.0             0                  3   \n","3     6.266046e+01         110.0             0                  1   \n","4     2.575850e+04       54313.0             0                  5   \n","...            ...           ...           ...                ...   \n","9995  0.000000e+00           1.0             1                  2   \n","9996  0.000000e+00         417.0           417                  2   \n","9997  2.161468e+06     9954935.0             0                 14   \n","9998  3.359510e+05     2723322.0             0                 21   \n","9999  5.357455e+05     2582774.0             0                 24   \n","\n","      Total.Backward.Packets  Total.Length.of.Fwd.Packets  \\\n","0                         55                          132   \n","1                          0                           12   \n","2                          0                          674   \n","3                          3                            0   \n","4                          0                         1076   \n","...                      ...                          ...   \n","9995                       0                          324   \n","9996                       0                           12   \n","9997                      14                         7874   \n","9998                     120                         2538   \n","9999                      23                         2288   \n","\n","      Total.Length.of.Bwd.Packets  \n","0                        110414.0  \n","1                             0.0  \n","2                             0.0  \n","3                             0.0  \n","4                             0.0  \n","...                           ...  \n","9995                          0.0  \n","9996                          0.0  \n","9997                       2622.0  \n","9998                      26843.0  \n","9999                       3551.0  \n","\n","[10000 rows x 11 columns]"],"text/html":["\n","  <div id=\"df-c0edce14-f17d-4943-a184-709122f091b0\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Flow.Duration</th>\n","      <th>Flow.Bytes.s</th>\n","      <th>Flow.Packets.s</th>\n","      <th>Flow.IAT.Mean</th>\n","      <th>Flow.IAT.Std</th>\n","      <th>Flow.IAT.Max</th>\n","      <th>Flow.IAT.Min</th>\n","      <th>Total.Fwd.Packets</th>\n","      <th>Total.Backward.Packets</th>\n","      <th>Total.Length.of.Fwd.Packets</th>\n","      <th>Total.Length.of.Bwd.Packets</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>45523</td>\n","      <td>2.428355e+06</td>\n","      <td>1.691453e+03</td>\n","      <td>598.986842</td>\n","      <td>8.160613e+02</td>\n","      <td>3880.0</td>\n","      <td>1</td>\n","      <td>22</td>\n","      <td>55</td>\n","      <td>132</td>\n","      <td>110414.0</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>1</td>\n","      <td>1.200000e+07</td>\n","      <td>2.000000e+06</td>\n","      <td>1.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>1.0</td>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>12</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>1</td>\n","      <td>6.740000e+08</td>\n","      <td>3.000000e+06</td>\n","      <td>0.500000</td>\n","      <td>7.071068e-01</td>\n","      <td>1.0</td>\n","      <td>0</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>674</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>217</td>\n","      <td>0.000000e+00</td>\n","      <td>1.843318e+04</td>\n","      <td>72.333333</td>\n","      <td>6.266046e+01</td>\n","      <td>110.0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>78068</td>\n","      <td>1.378286e+04</td>\n","      <td>6.404673e+01</td>\n","      <td>19517.000000</td>\n","      <td>2.575850e+04</td>\n","      <td>54313.0</td>\n","      <td>0</td>\n","      <td>5</td>\n","      <td>0</td>\n","      <td>1076</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>9995</th>\n","      <td>1</td>\n","      <td>3.240000e+08</td>\n","      <td>2.000000e+06</td>\n","      <td>1.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>1.0</td>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>324</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>9996</th>\n","      <td>417</td>\n","      <td>2.877698e+04</td>\n","      <td>4.796163e+03</td>\n","      <td>417.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>417.0</td>\n","      <td>417</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>12</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>9997</th>\n","      <td>16500339</td>\n","      <td>6.361081e+02</td>\n","      <td>1.696935e+00</td>\n","      <td>611123.666667</td>\n","      <td>2.161468e+06</td>\n","      <td>9954935.0</td>\n","      <td>0</td>\n","      <td>14</td>\n","      <td>14</td>\n","      <td>7874</td>\n","      <td>2622.0</td>\n","    </tr>\n","    <tr>\n","      <th>9998</th>\n","      <td>8051824</td>\n","      <td>3.648987e+03</td>\n","      <td>1.751156e+01</td>\n","      <td>57513.028571</td>\n","      <td>3.359510e+05</td>\n","      <td>2723322.0</td>\n","      <td>0</td>\n","      <td>21</td>\n","      <td>120</td>\n","      <td>2538</td>\n","      <td>26843.0</td>\n","    </tr>\n","    <tr>\n","      <th>9999</th>\n","      <td>8038513</td>\n","      <td>7.263781e+02</td>\n","      <td>5.846853e+00</td>\n","      <td>174750.282609</td>\n","      <td>5.357455e+05</td>\n","      <td>2582774.0</td>\n","      <td>0</td>\n","      <td>24</td>\n","      <td>23</td>\n","      <td>2288</td>\n","      <td>3551.0</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>10000 rows × 11 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c0edce14-f17d-4943-a184-709122f091b0')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-c0edce14-f17d-4943-a184-709122f091b0 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-c0edce14-f17d-4943-a184-709122f091b0');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":27}]},{"cell_type":"code","execution_count":28,"metadata":{"id":"PrtqWFYRNTEt","executionInfo":{"status":"ok","timestamp":1653643471190,"user_tz":180,"elapsed":251,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[],"source":["x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size=0.3)\n","x_treino = x_treino.dropna()\n","y_treino = y_treino.dropna()"]},{"cell_type":"markdown","metadata":{"id":"OB-1RIeMNTEs"},"source":["###  Grid Search Parameters"]},{"cell_type":"code","execution_count":29,"metadata":{"id":"xCik5rWlNTEt","outputId":"8a646808-a87e-44f2-ed6a-0e995a5db45e","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643478260,"user_tz":180,"elapsed":398,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["{'n_neighbors': [1, 3, 5, 10, 20]}\n","{'n_estimators': [10, 20, 50, 100, 200, 400]}\n","{'C': [1, 2, 4, 8, 16, 32, 64]}\n"]}],"source":["#parameters for KNN - number of neighbors\n","lista_k = [1,3,5,10,20]\n","parametros_knn = dict(n_neighbors=lista_k)\n","\n","#parameteres for RF and AdaBoost - number of trees\n","lista_t = [10,20,50,100,200,400]\n","parametros_rf = dict(n_estimators=lista_t)\n","\n","#parameters for SVM\n","lista_c = [1,2,4,8,16,32,64]\n","parametros_rbf = dict(C=lista_c)\n","\n","print(parametros_knn)\n","print(parametros_rf)\n","print(parametros_rbf)"]},{"cell_type":"markdown","source":["## Classificação com Modelos AM"],"metadata":{"id":"eoZ-swRmyDJa"}},{"cell_type":"markdown","metadata":{"id":"PeSflTL4NTEt"},"source":["### KNN"]},{"cell_type":"code","execution_count":30,"metadata":{"id":"QvWRjD9xNTEt","executionInfo":{"status":"ok","timestamp":1653643483521,"user_tz":180,"elapsed":248,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[],"source":["knn = KNeighborsClassifier()\n","grid = GridSearchCV(knn,parametros_knn,cv=10)"]},{"cell_type":"code","execution_count":31,"metadata":{"id":"3487EsLsNTEu","executionInfo":{"status":"ok","timestamp":1653643487373,"user_tz":180,"elapsed":2023,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[],"source":["inicio_grid_knn = time.time()\n","grid.fit(x_treino,y_treino)\n","tempo_grid_knn = time.time()- inicio_grid_knn"]},{"cell_type":"code","execution_count":32,"metadata":{"id":"QLk6FTI8NTEu","outputId":"a8faaa89-fb02-42e6-c245-3906b9576c97","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643489641,"user_tz":180,"elapsed":230,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["{'n_neighbors': 10}"]},"metadata":{},"execution_count":32}],"source":["grid.best_params_"]},{"cell_type":"code","execution_count":33,"metadata":{"id":"Tcsos-PvNTEu","outputId":"7d080b20-4501-4157-df39-926ac2603f4b","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643491512,"user_tz":180,"elapsed":243,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["0.6215714285714287"]},"metadata":{},"execution_count":33}],"source":["grid.best_score_"]},{"cell_type":"markdown","metadata":{"id":"RfsJuYuBNTEv"},"source":["#### KNN Performance"]},{"cell_type":"code","source":["inicio_predicao_knn = time.time()\n","prev_knn = grid.predict(x_teste)\n","tempo_predicao_knn = time.time() - inicio_predicao_knn"],"metadata":{"id":"VHOkngwRG9-1","executionInfo":{"status":"ok","timestamp":1653643494743,"user_tz":180,"elapsed":240,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":34,"outputs":[]},{"cell_type":"code","source":["prev_knn"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"lFKUg62eJU65","executionInfo":{"status":"ok","timestamp":1653643496649,"user_tz":180,"elapsed":232,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"e83083d5-4297-4701-eb06-6e802ac8773f"},"execution_count":35,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([ 7,  7,  7, ..., 21,  6,  7])"]},"metadata":{},"execution_count":35}]},{"cell_type":"code","source":["acc_res = metrics.accuracy_score(y_teste,prev_knn)\n","prec = metrics.precision_score(y_teste,prev_knn,average='weighted')\n","rec = metrics.recall_score(y_teste,prev_knn,average='weighted')\n","f1s = metrics.f1_score(y_teste,prev_knn,average='weighted')"],"metadata":{"id":"-TgYtZS2e1im","executionInfo":{"status":"ok","timestamp":1653643498670,"user_tz":180,"elapsed":280,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":36,"outputs":[]},{"cell_type":"code","source":["print(f'Precision: {prec:.2f}')\n","print(f'Recall:    {rec:.2f}')\n","print(f'F-1 Score: {f1s:.2f}')\n","print(f'Accuracy:  {acc_res:.2f}')\n","print(\" \")\n","print(f'Tempo Grid:  {tempo_grid_knn:.2f}')\n","print(f'Tempo Execução:  {tempo_predicao_knn:.2f}')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1PBljPfiyL7P","executionInfo":{"status":"ok","timestamp":1653643501287,"user_tz":180,"elapsed":238,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"5ca21c39-1468-4087-a176-16bcb79cd87f"},"execution_count":37,"outputs":[{"output_type":"stream","name":"stdout","text":["Precision: 0.62\n","Recall:    0.62\n","F-1 Score: 0.61\n","Accuracy:  0.62\n"," \n","Tempo Grid:  1.98\n","Tempo Execução:  0.12\n"]}]},{"cell_type":"code","execution_count":38,"metadata":{"id":"-yjikdtINTEv","outputId":"fe9f7253-e26a-4c2e-ffca-811027242063","colab":{"base_uri":"https://localhost:8080/","height":299},"executionInfo":{"status":"ok","timestamp":1653643507353,"user_tz":180,"elapsed":1152,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}],"source":["ax= plt.subplot()\n","array = confusion_matrix(y_teste, prev_knn)      \n","df_cm = pd.DataFrame(array, range(10),range(10))\n","#plt.figure(figsize = (10,7))\n","sn.set(font_scale=1.4)#for label size\n","sn.heatmap(df_cm, annot=True,annot_kws={\"size\": 14}, ax = ax,fmt=\"d\")# font size\n","\n","ax.set_xlabel('Predicted labels');ax.set_ylabel('True labels'); \n","ax.set_title('Confusion Matrix for KNN'); \n","#ax.xaxis.set_ticklabels(['Não Popular', 'Popular']); ax.yaxis.set_ticklabels(['Não Popular', 'Popular']);"]},{"cell_type":"markdown","metadata":{"id":"FynN1WYONTEx"},"source":["### Random Forest"]},{"cell_type":"code","execution_count":39,"metadata":{"id":"k9CzB_37NTEx","executionInfo":{"status":"ok","timestamp":1653643520092,"user_tz":180,"elapsed":248,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[],"source":["rf = RandomForestClassifier(random_state=1986,\n","                           criterion='gini',\n","                           max_depth=10,\n","                           n_estimators=50,\n","                           n_jobs=-1)\n","\n","grid = GridSearchCV(rf,parametros_rf,cv=10)"]},{"cell_type":"code","execution_count":40,"metadata":{"id":"-_gY3_6JNTEx","executionInfo":{"status":"ok","timestamp":1653643621951,"user_tz":180,"elapsed":100284,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[],"source":["inicio_grid_rf = time.time()\n","grid.fit(x_treino,y_treino)\n","tempo_grid_rf = time.time()- inicio_grid_rf"]},{"cell_type":"code","execution_count":41,"metadata":{"id":"ZMfL2NdFNTEy","outputId":"6c6fa9b1-99ba-49f0-c2fa-e68b33a9605d","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643631070,"user_tz":180,"elapsed":243,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["{'n_estimators': 400}"]},"metadata":{},"execution_count":41}],"source":["grid.best_params_"]},{"cell_type":"code","execution_count":42,"metadata":{"id":"WS6fnv9sNTEy","outputId":"6c65792e-743b-49ba-91db-73dffd31d292","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643633407,"user_tz":180,"elapsed":3,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["0.719"]},"metadata":{},"execution_count":42}],"source":["grid.best_score_"]},{"cell_type":"markdown","metadata":{"id":"nKj9vYYoNTE1"},"source":["#### Random Forest - Performance"]},{"cell_type":"code","source":["inicio_predicao_rf = time.time()\n","prev_rf = grid.predict(x_teste)\n","tempo_predicao_rf = time.time() - inicio_predicao_rf"],"metadata":{"id":"-v37d-PODwTm","executionInfo":{"status":"ok","timestamp":1653643637914,"user_tz":180,"elapsed":637,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":43,"outputs":[]},{"cell_type":"code","source":["acc_res = metrics.accuracy_score(y_teste,prev_rf)\n","prec = metrics.precision_score(y_teste,prev_rf,average='weighted')\n","rec = metrics.recall_score(y_teste,prev_rf,average='weighted')\n","f1s = metrics.f1_score(y_teste,prev_rf,average='weighted')"],"metadata":{"executionInfo":{"status":"ok","timestamp":1653643641227,"user_tz":180,"elapsed":325,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"colab":{"base_uri":"https://localhost:8080/"},"id":"8mKKs7DPDWk8","outputId":"84a6f8a3-39d1-4402-f061-265db175d976"},"execution_count":44,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n","  _warn_prf(average, modifier, msg_start, len(result))\n"]}]},{"cell_type":"code","source":["print(f'Precision: {prec:.2f}')\n","print(f'Recall:    {rec:.2f}')\n","print(f'F-1 Score: {f1s:.2f}')\n","print(f'Accuracy:  {acc_res:.2f}')\n","print(\" \")\n","print(f'Tempo Grid:  {tempo_grid_rf:.2f}')\n","print(f'Tempo Execução:  {tempo_predicao_rf:.2f}')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643643586,"user_tz":180,"elapsed":2,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"64c066bb-6069-45d6-e1d7-9a833d8a9c20","id":"QwX94t5tDWk8"},"execution_count":45,"outputs":[{"output_type":"stream","name":"stdout","text":["Precision: 0.73\n","Recall:    0.72\n","F-1 Score: 0.71\n","Accuracy:  0.72\n"," \n","Tempo Grid:  100.04\n","Tempo Execução:  0.32\n"]}]},{"cell_type":"code","execution_count":46,"metadata":{"id":"JR-v6uS2NTE1","outputId":"52612d94-d24c-494e-99e8-227a8caa9140","colab":{"base_uri":"https://localhost:8080/","height":315},"executionInfo":{"status":"ok","timestamp":1653643647032,"user_tz":180,"elapsed":1210,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["ax= plt.subplot()\n","array = confusion_matrix(y_teste, prev_rf)      \n","df_cm = pd.DataFrame(array, range(10),range(10))\n","#plt.figure(figsize = (10,7))\n","sn.set(font_scale=1.4)#for label size\n","sn.heatmap(df_cm, annot=True,annot_kws={\"size\": 14}, ax = ax,fmt=\"d\")# font size\n","\n","ax.set_xlabel('Predicted labels');ax.set_ylabel('True labels'); \n","ax.set_title('Confusion Matrix for Random Forest'); "]},{"cell_type":"markdown","source":["### Seleção de Features"],"metadata":{"id":"2fCk_WSPp0gs"}},{"cell_type":"code","source":["x_treino"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":488},"id":"72tifC0LqruN","executionInfo":{"status":"ok","timestamp":1653643652515,"user_tz":180,"elapsed":263,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"e65f3168-bd4d-45d5-d5c6-dfe9d62ef8ce"},"execution_count":47,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      Flow.Duration  Flow.Bytes.s  Flow.Packets.s  Flow.IAT.Mean  \\\n","9844       15119395  1.836714e+02    1.190524e+00  889376.176471   \n","1347              2  6.000000e+06    1.000000e+06       2.000000   \n","4162             21  0.000000e+00    1.428571e+05      10.500000   \n","4803           6662  2.521765e+04    9.006304e+02    1332.400000   \n","5923        5908120  1.744921e+04    1.946474e+01   51825.614035   \n","...             ...           ...             ...            ...   \n","9226         360205  1.303508e+05    1.610194e+02    6319.385965   \n","8609         146471  0.000000e+00    1.365458e+01  146471.000000   \n","7117            202  0.000000e+00    9.900990e+03     202.000000   \n","4730              1  5.099000e+09    3.000000e+06       0.500000   \n","4370            246  0.000000e+00    8.130081e+03     246.000000   \n","\n","      Flow.IAT.Std  Flow.IAT.Max  Flow.IAT.Min  Total.Fwd.Packets  \\\n","9844  3.640850e+06    15017726.0             0                 14   \n","1347  0.000000e+00           2.0             2                  2   \n","4162  1.343503e+01          20.0             1                  3   \n","4803  2.488598e+03        5775.0            40                  3   \n","5923  4.806171e+05     5127291.0             0                 36   \n","...            ...           ...           ...                ...   \n","9226  2.854241e+04      201867.0             0                 11   \n","8609  0.000000e+00      146471.0        146471                  1   \n","7117  0.000000e+00         202.0           202                  2   \n","4730  7.071068e-01           1.0             0                  3   \n","4370  0.000000e+00         246.0           246                  2   \n","\n","      Total.Backward.Packets  Total.Length.of.Fwd.Packets  \\\n","9844                       4                         1636   \n","1347                       0                           12   \n","4162                       0                            0   \n","4803                       3                          168   \n","5923                      79                         5062   \n","...                      ...                          ...   \n","9226                      47                         1233   \n","8609                       1                            0   \n","7117                       0                            0   \n","4730                       0                         5099   \n","4370                       0                            0   \n","\n","      Total.Length.of.Bwd.Packets  \n","9844                       1141.0  \n","1347                          0.0  \n","4162                          0.0  \n","4803                          0.0  \n","5923                      98030.0  \n","...                           ...  \n","9226                      45720.0  \n","8609                          0.0  \n","7117                          0.0  \n","4730                          0.0  \n","4370                          0.0  \n","\n","[7000 rows x 11 columns]"],"text/html":["\n","  <div id=\"df-1da17a0f-d70d-427c-a53c-905d0c1403bf\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Flow.Duration</th>\n","      <th>Flow.Bytes.s</th>\n","      <th>Flow.Packets.s</th>\n","      <th>Flow.IAT.Mean</th>\n","      <th>Flow.IAT.Std</th>\n","      <th>Flow.IAT.Max</th>\n","      <th>Flow.IAT.Min</th>\n","      <th>Total.Fwd.Packets</th>\n","      <th>Total.Backward.Packets</th>\n","      <th>Total.Length.of.Fwd.Packets</th>\n","      <th>Total.Length.of.Bwd.Packets</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>9844</th>\n","      <td>15119395</td>\n","      <td>1.836714e+02</td>\n","      <td>1.190524e+00</td>\n","      <td>889376.176471</td>\n","      <td>3.640850e+06</td>\n","      <td>15017726.0</td>\n","      <td>0</td>\n","      <td>14</td>\n","      <td>4</td>\n","      <td>1636</td>\n","      <td>1141.0</td>\n","    </tr>\n","    <tr>\n","      <th>1347</th>\n","      <td>2</td>\n","      <td>6.000000e+06</td>\n","      <td>1.000000e+06</td>\n","      <td>2.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>2.0</td>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>12</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>4162</th>\n","      <td>21</td>\n","      <td>0.000000e+00</td>\n","      <td>1.428571e+05</td>\n","      <td>10.500000</td>\n","      <td>1.343503e+01</td>\n","      <td>20.0</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>4803</th>\n","      <td>6662</td>\n","      <td>2.521765e+04</td>\n","      <td>9.006304e+02</td>\n","      <td>1332.400000</td>\n","      <td>2.488598e+03</td>\n","      <td>5775.0</td>\n","      <td>40</td>\n","      <td>3</td>\n","      <td>3</td>\n","      <td>168</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>5923</th>\n","      <td>5908120</td>\n","      <td>1.744921e+04</td>\n","      <td>1.946474e+01</td>\n","      <td>51825.614035</td>\n","      <td>4.806171e+05</td>\n","      <td>5127291.0</td>\n","      <td>0</td>\n","      <td>36</td>\n","      <td>79</td>\n","      <td>5062</td>\n","      <td>98030.0</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>9226</th>\n","      <td>360205</td>\n","      <td>1.303508e+05</td>\n","      <td>1.610194e+02</td>\n","      <td>6319.385965</td>\n","      <td>2.854241e+04</td>\n","      <td>201867.0</td>\n","      <td>0</td>\n","      <td>11</td>\n","      <td>47</td>\n","      <td>1233</td>\n","      <td>45720.0</td>\n","    </tr>\n","    <tr>\n","      <th>8609</th>\n","      <td>146471</td>\n","      <td>0.000000e+00</td>\n","      <td>1.365458e+01</td>\n","      <td>146471.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>146471.0</td>\n","      <td>146471</td>\n","      <td>1</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>7117</th>\n","      <td>202</td>\n","      <td>0.000000e+00</td>\n","      <td>9.900990e+03</td>\n","      <td>202.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>202.0</td>\n","      <td>202</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>4730</th>\n","      <td>1</td>\n","      <td>5.099000e+09</td>\n","      <td>3.000000e+06</td>\n","      <td>0.500000</td>\n","      <td>7.071068e-01</td>\n","      <td>1.0</td>\n","      <td>0</td>\n","      <td>3</td>\n","      <td>0</td>\n","      <td>5099</td>\n","      <td>0.0</td>\n","    </tr>\n","    <tr>\n","      <th>4370</th>\n","      <td>246</td>\n","      <td>0.000000e+00</td>\n","      <td>8.130081e+03</td>\n","      <td>246.000000</td>\n","      <td>0.000000e+00</td>\n","      <td>246.0</td>\n","      <td>246</td>\n","      <td>2</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0.0</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>7000 rows × 11 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1da17a0f-d70d-427c-a53c-905d0c1403bf')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-1da17a0f-d70d-427c-a53c-905d0c1403bf button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-1da17a0f-d70d-427c-a53c-905d0c1403bf');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":47}]},{"cell_type":"code","source":["y_treino"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"6rJWZk3Eq8jy","executionInfo":{"status":"ok","timestamp":1653643656797,"user_tz":180,"elapsed":263,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"e4af4e52-901d-4ac3-ca93-cc04eeb8aa32"},"execution_count":48,"outputs":[{"output_type":"execute_result","data":{"text/plain":["9844     7\n","1347     7\n","4162     7\n","4803     6\n","5923     8\n","        ..\n","9226     6\n","8609    15\n","7117     7\n","4730    15\n","4370    15\n","Name: classes, Length: 7000, dtype: int64"]},"metadata":{},"execution_count":48}]},{"cell_type":"code","source":["# Create the RFE object and rank each pixel\n","rfe = RFE(estimator=rf, n_features_to_select=1, step=1)"],"metadata":{"id":"kBslPvzeqFcL","executionInfo":{"status":"ok","timestamp":1653643668768,"user_tz":180,"elapsed":325,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":49,"outputs":[]},{"cell_type":"code","source":["rfe.fit(x_treino, y_treino)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"OiSY7pmFyrTx","executionInfo":{"status":"ok","timestamp":1653643677274,"user_tz":180,"elapsed":7143,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"7493a9af-dedb-46fa-97b4-0276ac76bd49"},"execution_count":50,"outputs":[{"output_type":"execute_result","data":{"text/plain":["RFE(estimator=RandomForestClassifier(max_depth=10, n_estimators=50, n_jobs=-1,\n","                                     random_state=1986),\n","    n_features_to_select=1)"]},"metadata":{},"execution_count":50}]},{"cell_type":"code","source":["#===========================================================================\n","# now print out the features in order of ranking\n","#===========================================================================\n","from operator import itemgetter\n","features = x_treino.columns.to_list()\n","for w, z in (sorted(zip(rfe.ranking_ , features), key=itemgetter(0))):\n","    print(w, z)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Z_UoJVuor5t2","executionInfo":{"status":"ok","timestamp":1653643681872,"user_tz":180,"elapsed":228,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"480ad660-3460-43ed-8f40-4f2782b73e85"},"execution_count":51,"outputs":[{"output_type":"stream","name":"stdout","text":["1 Flow.IAT.Max\n","2 Flow.Duration\n","3 Total.Length.of.Fwd.Packets\n","4 Total.Backward.Packets\n","5 Flow.Packets.s\n","6 Flow.IAT.Mean\n","7 Total.Fwd.Packets\n","8 Flow.Bytes.s\n","9 Total.Length.of.Bwd.Packets\n","10 Flow.IAT.Std\n","11 Flow.IAT.Min\n"]}]},{"cell_type":"code","source":["n_features_to_select = 5\n","rfe = RFE(rf, n_features_to_select=n_features_to_select)"],"metadata":{"id":"0yMN8-4M3G1C","executionInfo":{"status":"ok","timestamp":1653643684431,"user_tz":180,"elapsed":276,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":52,"outputs":[]},{"cell_type":"markdown","source":["### Random Forest Selecionado"],"metadata":{"id":"70wF4L793jB4"}},{"cell_type":"code","source":["inicio_rfe_rf = time.time()\n","rfe.fit(x_treino,y_treino)\n","tempo_rfe_rf = time.time()- inicio_rfe_rf"],"metadata":{"id":"eSvnv8RH3idW","executionInfo":{"status":"ok","timestamp":1653643692235,"user_tz":180,"elapsed":4950,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":53,"outputs":[]},{"cell_type":"markdown","source":["#### Performance"],"metadata":{"id":"EiqNpQgS35BF"}},{"cell_type":"code","source":["inicio_predicao_rf = time.time()\n","prev_rf = rfe.predict(x_teste)\n","tempo_predicao_rf = time.time() - inicio_predicao_rf"],"metadata":{"executionInfo":{"status":"ok","timestamp":1653643694344,"user_tz":180,"elapsed":257,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"id":"mPCWNiEy324m"},"execution_count":54,"outputs":[]},{"cell_type":"code","source":["acc_res = metrics.accuracy_score(y_teste,prev_rf)\n","prec = metrics.precision_score(y_teste,prev_rf,average='weighted')\n","rec = metrics.recall_score(y_teste,prev_rf,average='weighted')\n","f1s = metrics.f1_score(y_teste,prev_rf,average='weighted')"],"metadata":{"executionInfo":{"status":"ok","timestamp":1653643696737,"user_tz":180,"elapsed":253,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"cb04a62b-7cc5-400f-a35e-f25c76e46d98","id":"dGkd0G00324m"},"execution_count":55,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n","  _warn_prf(average, modifier, msg_start, len(result))\n"]}]},{"cell_type":"code","source":["print(f'Precision: {prec:.2f}')\n","print(f'Recall:    {rec:.2f}')\n","print(f'F-1 Score: {f1s:.2f}')\n","print(f'Accuracy:  {acc_res:.2f}')\n","print(\" \")\n","print(f'Tempo RFE:  {tempo_rfe_rf:.2f}')\n","print(f'Tempo Execução:  {tempo_predicao_rf:.2f}')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643699112,"user_tz":180,"elapsed":289,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"3d74ed2c-5b6d-4cea-c220-b30cbf05bd1e","id":"Sweg0iXa324m"},"execution_count":56,"outputs":[{"output_type":"stream","name":"stdout","text":["Precision: 0.71\n","Recall:    0.69\n","F-1 Score: 0.68\n","Accuracy:  0.69\n"," \n","Tempo RFE:  4.70\n","Tempo Execução:  0.11\n"]}]},{"cell_type":"code","execution_count":58,"metadata":{"outputId":"80d49291-58a5-42df-8bf4-52f980f4f4b9","colab":{"base_uri":"https://localhost:8080/","height":315},"executionInfo":{"status":"ok","timestamp":1653643722585,"user_tz":180,"elapsed":1262,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"id":"FWJCPmOm324m"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["ax= plt.subplot()\n","array = confusion_matrix(y_teste, prev_rf)      \n","df_cm = pd.DataFrame(array, range(10),range(10))\n","#plt.figure(figsize = (10,7))\n","sn.set(font_scale=1.4)#for label size\n","sn.heatmap(df_cm, annot=True,annot_kws={\"size\": 14}, ax = ax,fmt=\"d\")# font size\n","\n","ax.set_xlabel('Predicted labels');ax.set_ylabel('True labels'); \n","ax.set_title('Confusion Matrix for Random Forest'); "]},{"cell_type":"markdown","metadata":{"id":"zZ07xXVDmYhx"},"source":["### Redes Neurais"]},{"cell_type":"code","metadata":{"id":"bJVjlHVQmYHD","executionInfo":{"status":"ok","timestamp":1653643732307,"user_tz":180,"elapsed":240,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"source":["from sklearn.neural_network import MLPClassifier\n","from sklearn.datasets import make_classification\n","from sklearn.model_selection import train_test_split"],"execution_count":59,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"xL0qwzDlZk56"},"source":["#### Construindo a Rede"]},{"cell_type":"code","source":["mlp = MLPClassifier(random_state=1, max_iter=300,hidden_layer_sizes=(10,50,100,300,300,100,50,10),activation = 'relu', batch_size= 1000, solver='adam')"],"metadata":{"id":"QcqBbZlqyhxo","executionInfo":{"status":"ok","timestamp":1653643735883,"user_tz":180,"elapsed":258,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":60,"outputs":[]},{"cell_type":"code","metadata":{"id":"X4KfhIkEmkB6","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643869613,"user_tz":180,"elapsed":131579,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"d7b91b53-dde4-42b6-86c3-1e90cf3facf1"},"source":["inicio_treinamento_mlp = time.time()\n","mlp.fit(x_treino, y_treino)\n","tempo_treinamento_mlp = time.time() - inicio_treinamento_mlp"],"execution_count":61,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:696: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (300) reached and the optimization hasn't converged yet.\n","  ConvergenceWarning,\n"]}]},{"cell_type":"code","source":["inicio_predicao_mlp = time.time()\n","prev_mlp =  mlp.predict(x_teste)\n","tempo_predicao_mlp = time.time() - inicio_predicao_mlp"],"metadata":{"id":"bgW3QbPtyf9L","executionInfo":{"status":"ok","timestamp":1653643872919,"user_tz":180,"elapsed":237,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":62,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"C3qY9NaXnCSr"},"source":["#### Performance\n"]},{"cell_type":"code","metadata":{"id":"urRZEJfhGBEZ","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643874735,"user_tz":180,"elapsed":255,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"563bbb44-ae43-41d7-d2df-fd7a61eeceb8"},"source":["acc_res = metrics.accuracy_score(y_teste,prev_mlp)\n","prec = metrics.precision_score(y_teste,prev_mlp,average='weighted')\n","rec = metrics.recall_score(y_teste,prev_mlp,average='weighted')\n","f1s = metrics.f1_score(y_teste,prev_mlp,average='weighted')"],"execution_count":63,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n","  _warn_prf(average, modifier, msg_start, len(result))\n"]}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"hq9czy0oGBEZ","executionInfo":{"status":"ok","timestamp":1653643876624,"user_tz":180,"elapsed":247,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"52c46f7c-3fba-4c98-8e6d-2e63d1c4b4f4"},"source":["print(f'Precision: {prec:.2f}')\n","print(f'Recall:    {rec:.2f}')\n","print(f'F-1 Score: {f1s:.2f}')\n","print(f'Accuracy:  {acc_res:.2f}')\n","print(\" \")\n","print(f'Tempo Grid:  {tempo_treinamento_mlp:.2f}')\n","print(f'Tempo Execução:  {tempo_predicao_mlp:.2f}')"],"execution_count":64,"outputs":[{"output_type":"stream","name":"stdout","text":["Precision: 0.15\n","Recall:    0.39\n","F-1 Score: 0.22\n","Accuracy:  0.39\n"," \n","Tempo Grid:  131.32\n","Tempo Execução:  0.07\n"]}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":315},"id":"2fvcGAeunEkT","executionInfo":{"status":"ok","timestamp":1653643896950,"user_tz":180,"elapsed":1625,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"0597a338-7b77-4d11-edc4-324281481d1b"},"source":["ax= plt.subplot()\n","array = confusion_matrix(y_teste, prev_mlp)      \n","df_cm = pd.DataFrame(array, range(10),range(10))\n","#plt.figure(figsize = (10,7))\n","sn.set(font_scale=1.4)#for label size\n","sn.heatmap(df_cm, annot=True,annot_kws={\"size\": 14}, ax = ax,fmt=\"d\")# font size\n","\n","ax.set_xlabel('Predicted labels');ax.set_ylabel('True labels'); \n","ax.set_title('Confusion Matrix for Rede MLP');"],"execution_count":66,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","metadata":{"id":"7ZMgpJSc2vCZ"},"source":["### Rede  Neural Keras\n"]},{"cell_type":"code","source":["!pip install plot_model"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"DoJgo_wcqg7g","executionInfo":{"status":"ok","timestamp":1653607127628,"user_tz":180,"elapsed":6194,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"4bf9b931-4a28-4f45-8d58-ed1f3a1047ec"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n","Collecting plot_model\n","  Downloading plot_model-0.20-py3-none-any.whl (6.9 kB)\n","Installing collected packages: plot-model\n","Successfully installed plot-model-0.20\n"]}]},{"cell_type":"code","metadata":{"id":"LdofW4in22wB","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643911738,"user_tz":180,"elapsed":8126,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"43a7f4b1-f46e-4e79-9529-9ec16d19a59a"},"source":["import tensorflow as tf\n","print(\"Versão do TensorFlow:\",tf.__version__)"],"execution_count":67,"outputs":[{"output_type":"stream","name":"stdout","text":["Versão do TensorFlow: 2.8.0\n"]}]},{"cell_type":"code","metadata":{"id":"iADesfAU3r7i","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653643929252,"user_tz":180,"elapsed":233,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"7dfa3427-9dd7-4517-dc66-5e7a90a30be1"},"source":["import keras as k\n","print(\"Versão do Keras:\",k.__version__)"],"execution_count":68,"outputs":[{"output_type":"stream","name":"stdout","text":["Versão do Keras: 2.8.0\n"]}]},{"cell_type":"code","metadata":{"id":"VUwy2W9T4R8E","executionInfo":{"status":"ok","timestamp":1653643931088,"user_tz":180,"elapsed":255,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"source":["#Imports\n","from keras.models import Sequential\n","from keras.layers import Conv1D\n","from keras.layers import Conv2D\n","from keras.layers import MaxPooling2D,MaxPooling1D\n","from keras.layers import Dense,Flatten, Input, BatchNormalization "],"execution_count":69,"outputs":[]},{"cell_type":"code","source":["from sklearn.datasets import load_iris\n","from sklearn.model_selection import train_test_split\n","from sklearn.metrics import classification_report, confusion_matrix\n","\n","import matplotlib.pyplot as plt\n","\n","from keras.utils import np_utils\n","from keras.utils.vis_utils import plot_model\n","from keras import Model\n","from IPython.core.display import Image"],"metadata":{"id":"RPw-NQzAi7JF","executionInfo":{"status":"ok","timestamp":1653643933708,"user_tz":180,"elapsed":324,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":70,"outputs":[]},{"cell_type":"markdown","source":["#### Definição de Classes One-Hot Encoding"],"metadata":{"id":"tKVOXGMPywoL"}},{"cell_type":"code","source":["dict_classes"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"7N4HQ2T4ncnx","executionInfo":{"status":"ok","timestamp":1653643936857,"user_tz":180,"elapsed":245,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"e6ce2538-907e-4dbe-b0b9-d43173f3678d"},"execution_count":71,"outputs":[{"output_type":"execute_result","data":{"text/plain":["{0: 'AMAZON',\n"," 5: 'GMAIL',\n"," 6: 'GOOGLE',\n"," 7: 'HTTP',\n"," 8: 'HTTP_CONNECT',\n"," 9: 'HTTP_PROXY',\n"," 10: 'MICROSOFT',\n"," 15: 'SSL',\n"," 19: 'WINDOWS_UPDATE',\n"," 21: 'YOUTUBE'}"]},"metadata":{},"execution_count":71}]},{"cell_type":"code","source":["dict_classes_ajustado = {0 : 0,\n","                         5 : 1,\n","                         6 : 2,\n","                         7 : 3,\n","                         8 : 4,\n","                         9 : 5,\n","                        10 : 6,\n","                        15 : 7,\n","                        19 : 8,\n","                        21 : 9}"],"metadata":{"id":"DiWn-wglnhLo","executionInfo":{"status":"ok","timestamp":1653643938851,"user_tz":180,"elapsed":267,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":72,"outputs":[]},{"cell_type":"code","source":["y_treino = y_treino.map(dict_classes_ajustado)\n","y_teste = y_teste.map(dict_classes_ajustado)"],"metadata":{"id":"KyPWM4TooD5n","executionInfo":{"status":"ok","timestamp":1653643940876,"user_tz":180,"elapsed":263,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":73,"outputs":[]},{"cell_type":"code","source":["y_treino"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"jEQVLK_SoP5x","executionInfo":{"status":"ok","timestamp":1653643943086,"user_tz":180,"elapsed":297,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"695b7255-67a0-47d6-8506-3d95a5c30fb0"},"execution_count":74,"outputs":[{"output_type":"execute_result","data":{"text/plain":["9844    3\n","1347    3\n","4162    3\n","4803    2\n","5923    4\n","       ..\n","9226    2\n","8609    7\n","7117    3\n","4730    7\n","4370    7\n","Name: classes, Length: 7000, dtype: int64"]},"metadata":{},"execution_count":74}]},{"cell_type":"code","source":["y_train_onehot = k.utils.np_utils.to_categorical(y_treino, num_classes = 10)\n","y_test_onehot = k.utils.np_utils.to_categorical(y_teste, num_classes = 10)"],"metadata":{"id":"6EX58Gsfi8Cf","executionInfo":{"status":"ok","timestamp":1653643945441,"user_tz":180,"elapsed":2,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":75,"outputs":[]},{"cell_type":"markdown","source":["#### Construindo a Rede"],"metadata":{"id":"ExbHTGVQy38v"}},{"cell_type":"code","source":["def create_model():\n","  model = Sequential()\n","  model.add(Dense(10, activation='relu', input_dim=11))\n","  model.add(Dense(8,activation='relu'))\n","  model.add(Dense(6,activation='relu'))\n","  model.add(Dense(10,activation='softmax'))\n","  model.compile('adam','categorical_crossentropy', metrics=['categorical_accuracy'])\n","\n","  return model"],"metadata":{"id":"HGMNhEOspY7g","executionInfo":{"status":"ok","timestamp":1653643947379,"user_tz":180,"elapsed":265,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":76,"outputs":[]},{"cell_type":"code","source":["model=create_model()\n","\n","k.utils.vis_utils.plot_model(\n","    model,\n","    show_shapes=True,\n","    show_layer_names=True,\n","    rankdir=\"LR\",\n","    expand_nested=True,\n","    dpi=96\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":152},"id":"2-sF2ew7pMSv","executionInfo":{"status":"ok","timestamp":1653643950670,"user_tz":180,"elapsed":1316,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"2f7bc609-c0c4-4f7d-f078-d7d35569a574"},"execution_count":77,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<IPython.core.display.Image object>"],"image/png":"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\n"},"metadata":{},"execution_count":77}]},{"cell_type":"markdown","source":["#### Treinamento"],"metadata":{"id":"ybeRZcQE95J6"}},{"cell_type":"code","source":["historico_treinamento = model.fit(x_treino, y_train_onehot, validation_split=.1, epochs=500, verbose=2)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"sCUMAhDWpV4_","executionInfo":{"status":"ok","timestamp":1653644137498,"user_tz":180,"elapsed":180148,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"d4ec2bee-0632-4ff7-cebe-1ec004b77b54"},"execution_count":78,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/500\n","197/197 - 1s - loss: 854251.5000 - categorical_accuracy: 0.3392 - val_loss: 123233.2578 - val_categorical_accuracy: 0.4243 - 1s/epoch - 7ms/step\n","Epoch 2/500\n","197/197 - 0s - loss: 13996.2695 - categorical_accuracy: 0.4344 - val_loss: 34357.8477 - val_categorical_accuracy: 0.4143 - 343ms/epoch - 2ms/step\n","Epoch 3/500\n","197/197 - 0s - loss: 1608.8820 - categorical_accuracy: 0.4116 - val_loss: 612.3534 - val_categorical_accuracy: 0.3729 - 343ms/epoch - 2ms/step\n","Epoch 4/500\n","197/197 - 0s - loss: 401.2962 - categorical_accuracy: 0.3922 - val_loss: 488.7277 - val_categorical_accuracy: 0.3729 - 340ms/epoch - 2ms/step\n","Epoch 5/500\n","197/197 - 0s - loss: 275.4845 - categorical_accuracy: 0.3865 - val_loss: 370.8303 - val_categorical_accuracy: 0.3657 - 340ms/epoch - 2ms/step\n","Epoch 6/500\n","197/197 - 0s - loss: 163.9326 - categorical_accuracy: 0.3767 - val_loss: 262.8062 - val_categorical_accuracy: 0.3243 - 374ms/epoch - 2ms/step\n","Epoch 7/500\n","197/197 - 0s - loss: 83.1792 - categorical_accuracy: 0.2141 - val_loss: 251.3870 - val_categorical_accuracy: 0.1586 - 379ms/epoch - 2ms/step\n","Epoch 8/500\n","197/197 - 0s - loss: 71.3533 - categorical_accuracy: 0.1790 - val_loss: 236.7271 - val_categorical_accuracy: 0.1543 - 369ms/epoch - 2ms/step\n","Epoch 9/500\n","197/197 - 0s - loss: 61.9600 - categorical_accuracy: 0.3840 - val_loss: 217.4087 - val_categorical_accuracy: 0.3757 - 331ms/epoch - 2ms/step\n","Epoch 10/500\n","197/197 - 0s - loss: 54.8760 - categorical_accuracy: 0.3875 - val_loss: 146.6656 - val_categorical_accuracy: 0.3757 - 378ms/epoch - 2ms/step\n","Epoch 11/500\n","197/197 - 0s - loss: 44.9804 - categorical_accuracy: 0.3860 - val_loss: 115.3736 - val_categorical_accuracy: 0.3757 - 365ms/epoch - 2ms/step\n","Epoch 12/500\n","197/197 - 0s - loss: 35.0886 - categorical_accuracy: 0.3871 - val_loss: 95.4167 - val_categorical_accuracy: 0.3757 - 338ms/epoch - 2ms/step\n","Epoch 13/500\n","197/197 - 0s - loss: 31.2360 - categorical_accuracy: 0.3870 - val_loss: 73.6152 - val_categorical_accuracy: 0.3757 - 327ms/epoch - 2ms/step\n","Epoch 14/500\n","197/197 - 0s - loss: 26.8293 - categorical_accuracy: 0.3863 - val_loss: 53.8830 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 15/500\n","197/197 - 0s - loss: 23.4945 - categorical_accuracy: 0.3859 - val_loss: 24.0032 - val_categorical_accuracy: 0.3757 - 393ms/epoch - 2ms/step\n","Epoch 16/500\n","197/197 - 0s - loss: 19.1382 - categorical_accuracy: 0.3854 - val_loss: 46.4802 - val_categorical_accuracy: 0.3757 - 361ms/epoch - 2ms/step\n","Epoch 17/500\n","197/197 - 0s - loss: 17.4106 - categorical_accuracy: 0.3843 - val_loss: 13.8918 - val_categorical_accuracy: 0.3757 - 356ms/epoch - 2ms/step\n","Epoch 18/500\n","197/197 - 0s - loss: 14.7964 - categorical_accuracy: 0.3841 - val_loss: 14.0989 - val_categorical_accuracy: 0.3743 - 337ms/epoch - 2ms/step\n","Epoch 19/500\n","197/197 - 0s - loss: 13.3063 - categorical_accuracy: 0.3840 - val_loss: 15.7009 - val_categorical_accuracy: 0.3757 - 357ms/epoch - 2ms/step\n","Epoch 20/500\n","197/197 - 0s - loss: 11.5763 - categorical_accuracy: 0.3844 - val_loss: 40.3000 - val_categorical_accuracy: 0.3757 - 367ms/epoch - 2ms/step\n","Epoch 21/500\n","197/197 - 0s - loss: 10.3319 - categorical_accuracy: 0.3849 - val_loss: 31.8207 - val_categorical_accuracy: 0.3757 - 368ms/epoch - 2ms/step\n","Epoch 22/500\n","197/197 - 0s - loss: 8.3081 - categorical_accuracy: 0.3848 - val_loss: 15.8254 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 23/500\n","197/197 - 0s - loss: 10.4826 - categorical_accuracy: 0.3846 - val_loss: 33.5591 - val_categorical_accuracy: 0.3743 - 363ms/epoch - 2ms/step\n","Epoch 24/500\n","197/197 - 0s - loss: 9.9052 - categorical_accuracy: 0.3846 - val_loss: 10.2879 - val_categorical_accuracy: 0.3757 - 337ms/epoch - 2ms/step\n","Epoch 25/500\n","197/197 - 0s - loss: 7.8367 - categorical_accuracy: 0.3838 - val_loss: 8.5530 - val_categorical_accuracy: 0.3743 - 321ms/epoch - 2ms/step\n","Epoch 26/500\n","197/197 - 0s - loss: 6.3876 - categorical_accuracy: 0.3846 - val_loss: 5.8322 - val_categorical_accuracy: 0.3757 - 368ms/epoch - 2ms/step\n","Epoch 27/500\n","197/197 - 0s - loss: 5.2622 - categorical_accuracy: 0.3843 - val_loss: 5.1344 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 28/500\n","197/197 - 0s - loss: 5.1139 - categorical_accuracy: 0.3843 - val_loss: 37.1995 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 29/500\n","197/197 - 0s - loss: 4.8641 - categorical_accuracy: 0.3844 - val_loss: 2.7333 - val_categorical_accuracy: 0.3757 - 364ms/epoch - 2ms/step\n","Epoch 30/500\n","197/197 - 0s - loss: 5.1683 - categorical_accuracy: 0.3841 - val_loss: 2.7113 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 31/500\n","197/197 - 0s - loss: 3.8389 - categorical_accuracy: 0.3841 - val_loss: 3.6620 - val_categorical_accuracy: 0.3743 - 365ms/epoch - 2ms/step\n","Epoch 32/500\n","197/197 - 0s - loss: 3.1048 - categorical_accuracy: 0.3840 - val_loss: 3.5740 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 33/500\n","197/197 - 0s - loss: 3.4959 - categorical_accuracy: 0.3848 - val_loss: 2.4028 - val_categorical_accuracy: 0.3743 - 372ms/epoch - 2ms/step\n","Epoch 34/500\n","197/197 - 0s - loss: 3.7993 - categorical_accuracy: 0.3843 - val_loss: 2.6944 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 35/500\n","197/197 - 0s - loss: 2.9924 - categorical_accuracy: 0.3846 - val_loss: 2.6317 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 36/500\n","197/197 - 0s - loss: 2.4833 - categorical_accuracy: 0.3840 - val_loss: 2.6119 - val_categorical_accuracy: 0.3729 - 362ms/epoch - 2ms/step\n","Epoch 37/500\n","197/197 - 0s - loss: 2.3235 - categorical_accuracy: 0.3848 - val_loss: 2.5661 - val_categorical_accuracy: 0.3743 - 324ms/epoch - 2ms/step\n","Epoch 38/500\n","197/197 - 0s - loss: 2.1580 - categorical_accuracy: 0.3841 - val_loss: 2.1716 - val_categorical_accuracy: 0.3743 - 331ms/epoch - 2ms/step\n","Epoch 39/500\n","197/197 - 0s - loss: 3.2161 - categorical_accuracy: 0.3846 - val_loss: 13.8949 - val_categorical_accuracy: 0.3743 - 355ms/epoch - 2ms/step\n","Epoch 40/500\n","197/197 - 0s - loss: 2.5310 - categorical_accuracy: 0.3849 - val_loss: 2.2759 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 41/500\n","197/197 - 0s - loss: 1.9821 - categorical_accuracy: 0.3843 - val_loss: 2.3185 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 42/500\n","197/197 - 0s - loss: 1.9555 - categorical_accuracy: 0.3841 - val_loss: 2.0684 - val_categorical_accuracy: 0.3743 - 363ms/epoch - 2ms/step\n","Epoch 43/500\n","197/197 - 0s - loss: 2.7710 - categorical_accuracy: 0.3846 - val_loss: 2.0869 - val_categorical_accuracy: 0.3743 - 358ms/epoch - 2ms/step\n","Epoch 44/500\n","197/197 - 0s - loss: 3.8560 - categorical_accuracy: 0.3844 - val_loss: 15.9838 - val_categorical_accuracy: 0.3743 - 353ms/epoch - 2ms/step\n","Epoch 45/500\n","197/197 - 0s - loss: 3.0227 - categorical_accuracy: 0.3841 - val_loss: 2.2470 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 46/500\n","197/197 - 0s - loss: 2.9636 - categorical_accuracy: 0.3848 - val_loss: 2.1022 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 47/500\n","197/197 - 0s - loss: 1.9102 - categorical_accuracy: 0.3844 - val_loss: 2.0461 - val_categorical_accuracy: 0.3743 - 371ms/epoch - 2ms/step\n","Epoch 48/500\n","197/197 - 0s - loss: 1.9060 - categorical_accuracy: 0.3843 - val_loss: 2.0488 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 49/500\n","197/197 - 0s - loss: 2.1386 - categorical_accuracy: 0.3844 - val_loss: 2.2517 - val_categorical_accuracy: 0.3743 - 358ms/epoch - 2ms/step\n","Epoch 50/500\n","197/197 - 0s - loss: 3.3571 - categorical_accuracy: 0.3846 - val_loss: 1.9943 - val_categorical_accuracy: 0.3743 - 355ms/epoch - 2ms/step\n","Epoch 51/500\n","197/197 - 0s - loss: 2.4841 - categorical_accuracy: 0.3844 - val_loss: 2.5097 - val_categorical_accuracy: 0.3743 - 383ms/epoch - 2ms/step\n","Epoch 52/500\n","197/197 - 0s - loss: 1.9535 - categorical_accuracy: 0.3841 - val_loss: 2.1529 - val_categorical_accuracy: 0.3743 - 357ms/epoch - 2ms/step\n","Epoch 53/500\n","197/197 - 0s - loss: 1.9590 - categorical_accuracy: 0.3844 - val_loss: 2.3734 - val_categorical_accuracy: 0.3743 - 337ms/epoch - 2ms/step\n","Epoch 54/500\n","197/197 - 0s - loss: 1.8824 - categorical_accuracy: 0.3846 - val_loss: 2.0235 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 55/500\n","197/197 - 0s - loss: 1.8270 - categorical_accuracy: 0.3848 - val_loss: 2.0378 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 56/500\n","197/197 - 0s - loss: 1.8253 - categorical_accuracy: 0.3843 - val_loss: 1.9931 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 57/500\n","197/197 - 0s - loss: 1.8164 - categorical_accuracy: 0.3844 - val_loss: 1.8715 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 58/500\n","197/197 - 0s - loss: 1.9028 - categorical_accuracy: 0.3849 - val_loss: 2.3119 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 59/500\n","197/197 - 0s - loss: 5.6480 - categorical_accuracy: 0.3843 - val_loss: 2.2571 - val_categorical_accuracy: 0.3743 - 321ms/epoch - 2ms/step\n","Epoch 60/500\n","197/197 - 0s - loss: 1.9147 - categorical_accuracy: 0.3840 - val_loss: 2.1267 - val_categorical_accuracy: 0.3729 - 365ms/epoch - 2ms/step\n","Epoch 61/500\n","197/197 - 0s - loss: 1.8159 - categorical_accuracy: 0.3843 - val_loss: 2.0618 - val_categorical_accuracy: 0.3729 - 385ms/epoch - 2ms/step\n","Epoch 62/500\n","197/197 - 0s - loss: 1.8144 - categorical_accuracy: 0.3846 - val_loss: 2.0399 - val_categorical_accuracy: 0.3729 - 363ms/epoch - 2ms/step\n","Epoch 63/500\n","197/197 - 0s - loss: 1.8046 - categorical_accuracy: 0.3848 - val_loss: 1.9315 - val_categorical_accuracy: 0.3729 - 352ms/epoch - 2ms/step\n","Epoch 64/500\n","197/197 - 0s - loss: 1.8032 - categorical_accuracy: 0.3849 - val_loss: 1.9377 - val_categorical_accuracy: 0.3743 - 348ms/epoch - 2ms/step\n","Epoch 65/500\n","197/197 - 0s - loss: 1.7992 - categorical_accuracy: 0.3846 - val_loss: 1.8275 - val_categorical_accuracy: 0.3729 - 378ms/epoch - 2ms/step\n","Epoch 66/500\n","197/197 - 0s - loss: 1.7965 - categorical_accuracy: 0.3848 - val_loss: 1.9154 - val_categorical_accuracy: 0.3743 - 317ms/epoch - 2ms/step\n","Epoch 67/500\n","197/197 - 0s - loss: 1.7973 - categorical_accuracy: 0.3846 - val_loss: 1.8195 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 68/500\n","197/197 - 0s - loss: 1.7950 - categorical_accuracy: 0.3849 - val_loss: 1.8900 - val_categorical_accuracy: 0.3729 - 332ms/epoch - 2ms/step\n","Epoch 69/500\n","197/197 - 0s - loss: 1.7948 - categorical_accuracy: 0.3844 - val_loss: 1.8956 - val_categorical_accuracy: 0.3729 - 349ms/epoch - 2ms/step\n","Epoch 70/500\n","197/197 - 0s - loss: 1.7921 - categorical_accuracy: 0.3848 - val_loss: 1.8192 - val_categorical_accuracy: 0.3729 - 367ms/epoch - 2ms/step\n","Epoch 71/500\n","197/197 - 0s - loss: 1.7920 - categorical_accuracy: 0.3849 - val_loss: 1.8781 - val_categorical_accuracy: 0.3729 - 331ms/epoch - 2ms/step\n","Epoch 72/500\n","197/197 - 0s - loss: 1.7891 - categorical_accuracy: 0.3851 - val_loss: 1.8746 - val_categorical_accuracy: 0.3729 - 358ms/epoch - 2ms/step\n","Epoch 73/500\n","197/197 - 0s - loss: 1.7893 - categorical_accuracy: 0.3849 - val_loss: 1.8934 - val_categorical_accuracy: 0.3729 - 336ms/epoch - 2ms/step\n","Epoch 74/500\n","197/197 - 0s - loss: 1.7900 - categorical_accuracy: 0.3849 - val_loss: 1.8965 - val_categorical_accuracy: 0.3729 - 358ms/epoch - 2ms/step\n","Epoch 75/500\n","197/197 - 0s - loss: 1.7896 - categorical_accuracy: 0.3849 - val_loss: 1.8987 - val_categorical_accuracy: 0.3729 - 357ms/epoch - 2ms/step\n","Epoch 76/500\n","197/197 - 0s - loss: 1.7982 - categorical_accuracy: 0.3851 - val_loss: 1.8831 - val_categorical_accuracy: 0.3729 - 362ms/epoch - 2ms/step\n","Epoch 77/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8186 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 78/500\n","197/197 - 0s - loss: 1.8233 - categorical_accuracy: 0.3848 - val_loss: 1.9392 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 79/500\n","197/197 - 0s - loss: 3.8830 - categorical_accuracy: 0.3838 - val_loss: 1.8630 - val_categorical_accuracy: 0.3714 - 368ms/epoch - 2ms/step\n","Epoch 80/500\n","197/197 - 0s - loss: 2.6926 - categorical_accuracy: 0.3840 - val_loss: 2.1967 - val_categorical_accuracy: 0.3729 - 366ms/epoch - 2ms/step\n","Epoch 81/500\n","197/197 - 0s - loss: 1.8713 - categorical_accuracy: 0.3846 - val_loss: 1.8135 - val_categorical_accuracy: 0.3743 - 381ms/epoch - 2ms/step\n","Epoch 82/500\n","197/197 - 0s - loss: 1.8354 - categorical_accuracy: 0.3849 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 367ms/epoch - 2ms/step\n","Epoch 83/500\n","197/197 - 0s - loss: 1.8132 - categorical_accuracy: 0.3851 - val_loss: 1.8575 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 84/500\n","197/197 - 0s - loss: 1.7999 - categorical_accuracy: 0.3851 - val_loss: 2.1319 - val_categorical_accuracy: 0.3743 - 375ms/epoch - 2ms/step\n","Epoch 85/500\n","197/197 - 0s - loss: 1.7926 - categorical_accuracy: 0.3851 - val_loss: 2.0917 - val_categorical_accuracy: 0.3743 - 359ms/epoch - 2ms/step\n","Epoch 86/500\n","197/197 - 0s - loss: 1.7916 - categorical_accuracy: 0.3852 - val_loss: 2.0624 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 87/500\n","197/197 - 0s - loss: 1.7910 - categorical_accuracy: 0.3851 - val_loss: 2.0216 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 88/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 2.0152 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 89/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 2.0091 - val_categorical_accuracy: 0.3743 - 356ms/epoch - 2ms/step\n","Epoch 90/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 2.0092 - val_categorical_accuracy: 0.3743 - 339ms/epoch - 2ms/step\n","Epoch 91/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 2.0081 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 92/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 2.0078 - val_categorical_accuracy: 0.3743 - 351ms/epoch - 2ms/step\n","Epoch 93/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 2.0077 - val_categorical_accuracy: 0.3743 - 371ms/epoch - 2ms/step\n","Epoch 94/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 1.9879 - val_categorical_accuracy: 0.3743 - 315ms/epoch - 2ms/step\n","Epoch 95/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 1.9716 - val_categorical_accuracy: 0.3743 - 338ms/epoch - 2ms/step\n","Epoch 96/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9745 - val_categorical_accuracy: 0.3743 - 358ms/epoch - 2ms/step\n","Epoch 97/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9767 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 98/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9797 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 99/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9819 - val_categorical_accuracy: 0.3743 - 358ms/epoch - 2ms/step\n","Epoch 100/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9839 - val_categorical_accuracy: 0.3743 - 355ms/epoch - 2ms/step\n","Epoch 101/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9839 - val_categorical_accuracy: 0.3743 - 358ms/epoch - 2ms/step\n","Epoch 102/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9860 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 103/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9888 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 104/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9886 - val_categorical_accuracy: 0.3743 - 374ms/epoch - 2ms/step\n","Epoch 105/500\n","197/197 - 0s - loss: 1.7901 - categorical_accuracy: 0.3851 - val_loss: 1.9952 - val_categorical_accuracy: 0.3743 - 359ms/epoch - 2ms/step\n","Epoch 106/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 1.9764 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 107/500\n","197/197 - 0s - loss: 1.7903 - categorical_accuracy: 0.3851 - val_loss: 1.9873 - val_categorical_accuracy: 0.3743 - 349ms/epoch - 2ms/step\n","Epoch 108/500\n","197/197 - 0s - loss: 1.7903 - categorical_accuracy: 0.3851 - val_loss: 1.9930 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 109/500\n","197/197 - 0s - loss: 1.7903 - categorical_accuracy: 0.3851 - val_loss: 1.9976 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 110/500\n","197/197 - 0s - loss: 1.7903 - categorical_accuracy: 0.3851 - val_loss: 2.0040 - val_categorical_accuracy: 0.3743 - 344ms/epoch - 2ms/step\n","Epoch 111/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 2.0109 - val_categorical_accuracy: 0.3743 - 319ms/epoch - 2ms/step\n","Epoch 112/500\n","197/197 - 0s - loss: 1.7903 - categorical_accuracy: 0.3851 - val_loss: 2.0191 - val_categorical_accuracy: 0.3743 - 350ms/epoch - 2ms/step\n","Epoch 113/500\n","197/197 - 0s - loss: 1.7903 - categorical_accuracy: 0.3851 - val_loss: 2.0254 - val_categorical_accuracy: 0.3743 - 349ms/epoch - 2ms/step\n","Epoch 114/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 2.0304 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 115/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 2.0383 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 116/500\n","197/197 - 0s - loss: 1.7902 - categorical_accuracy: 0.3851 - val_loss: 2.0482 - val_categorical_accuracy: 0.3743 - 376ms/epoch - 2ms/step\n","Epoch 117/500\n","197/197 - 0s - loss: 1.8018 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 118/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 337ms/epoch - 2ms/step\n","Epoch 119/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 320ms/epoch - 2ms/step\n","Epoch 120/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 365ms/epoch - 2ms/step\n","Epoch 121/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 122/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 372ms/epoch - 2ms/step\n","Epoch 123/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 124/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 344ms/epoch - 2ms/step\n","Epoch 125/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 126/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 127/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 382ms/epoch - 2ms/step\n","Epoch 128/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 356ms/epoch - 2ms/step\n","Epoch 129/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 338ms/epoch - 2ms/step\n","Epoch 130/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 356ms/epoch - 2ms/step\n","Epoch 131/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8133 - val_categorical_accuracy: 0.3743 - 357ms/epoch - 2ms/step\n","Epoch 132/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 133/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 134/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 135/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 340ms/epoch - 2ms/step\n","Epoch 136/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 137/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 357ms/epoch - 2ms/step\n","Epoch 138/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 139/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 355ms/epoch - 2ms/step\n","Epoch 140/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 141/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 142/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 359ms/epoch - 2ms/step\n","Epoch 143/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 379ms/epoch - 2ms/step\n","Epoch 144/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 381ms/epoch - 2ms/step\n","Epoch 145/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 376ms/epoch - 2ms/step\n","Epoch 146/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 147/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 344ms/epoch - 2ms/step\n","Epoch 148/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 149/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 345ms/epoch - 2ms/step\n","Epoch 150/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 346ms/epoch - 2ms/step\n","Epoch 151/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 152/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 153/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 389ms/epoch - 2ms/step\n","Epoch 154/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 375ms/epoch - 2ms/step\n","Epoch 155/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 365ms/epoch - 2ms/step\n","Epoch 156/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 157/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 383ms/epoch - 2ms/step\n","Epoch 158/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 159/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 392ms/epoch - 2ms/step\n","Epoch 160/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 161/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 338ms/epoch - 2ms/step\n","Epoch 162/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 163/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 337ms/epoch - 2ms/step\n","Epoch 164/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 376ms/epoch - 2ms/step\n","Epoch 165/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 166/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 167/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 374ms/epoch - 2ms/step\n","Epoch 168/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 341ms/epoch - 2ms/step\n","Epoch 169/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 170/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 171/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 376ms/epoch - 2ms/step\n","Epoch 172/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 173/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 174/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 341ms/epoch - 2ms/step\n","Epoch 175/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 176/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 177/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 376ms/epoch - 2ms/step\n","Epoch 178/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 338ms/epoch - 2ms/step\n","Epoch 179/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 339ms/epoch - 2ms/step\n","Epoch 180/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 181/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 340ms/epoch - 2ms/step\n","Epoch 182/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 183/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 184/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 185/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 367ms/epoch - 2ms/step\n","Epoch 186/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 187/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 341ms/epoch - 2ms/step\n","Epoch 188/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 351ms/epoch - 2ms/step\n","Epoch 189/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 190/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 191/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 337ms/epoch - 2ms/step\n","Epoch 192/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 193/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 377ms/epoch - 2ms/step\n","Epoch 194/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 195/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 196/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 197/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 331ms/epoch - 2ms/step\n","Epoch 198/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 331ms/epoch - 2ms/step\n","Epoch 199/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 200/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 376ms/epoch - 2ms/step\n","Epoch 201/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 202/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 203/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 204/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 382ms/epoch - 2ms/step\n","Epoch 205/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 374ms/epoch - 2ms/step\n","Epoch 206/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 207/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 338ms/epoch - 2ms/step\n","Epoch 208/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 209/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 210/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 211/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 212/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 213/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 214/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 215/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 331ms/epoch - 2ms/step\n","Epoch 216/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 378ms/epoch - 2ms/step\n","Epoch 217/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 390ms/epoch - 2ms/step\n","Epoch 218/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 343ms/epoch - 2ms/step\n","Epoch 219/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 367ms/epoch - 2ms/step\n","Epoch 220/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 320ms/epoch - 2ms/step\n","Epoch 221/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 331ms/epoch - 2ms/step\n","Epoch 222/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 369ms/epoch - 2ms/step\n","Epoch 223/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 355ms/epoch - 2ms/step\n","Epoch 224/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 376ms/epoch - 2ms/step\n","Epoch 225/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 226/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 227/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 387ms/epoch - 2ms/step\n","Epoch 228/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 319ms/epoch - 2ms/step\n","Epoch 229/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 230/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 385ms/epoch - 2ms/step\n","Epoch 231/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 332ms/epoch - 2ms/step\n","Epoch 232/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 374ms/epoch - 2ms/step\n","Epoch 233/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 345ms/epoch - 2ms/step\n","Epoch 234/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 235/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 236/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 372ms/epoch - 2ms/step\n","Epoch 237/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 238/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 239/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 332ms/epoch - 2ms/step\n","Epoch 240/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 322ms/epoch - 2ms/step\n","Epoch 241/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 359ms/epoch - 2ms/step\n","Epoch 242/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 366ms/epoch - 2ms/step\n","Epoch 243/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 244/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 338ms/epoch - 2ms/step\n","Epoch 245/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 246/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 356ms/epoch - 2ms/step\n","Epoch 247/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 248/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 249/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 250/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 377ms/epoch - 2ms/step\n","Epoch 251/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 322ms/epoch - 2ms/step\n","Epoch 252/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 315ms/epoch - 2ms/step\n","Epoch 253/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 332ms/epoch - 2ms/step\n","Epoch 254/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 255/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 256/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 374ms/epoch - 2ms/step\n","Epoch 257/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 258/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 259/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 260/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 261/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 322ms/epoch - 2ms/step\n","Epoch 262/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 342ms/epoch - 2ms/step\n","Epoch 263/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 318ms/epoch - 2ms/step\n","Epoch 264/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 265/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 336ms/epoch - 2ms/step\n","Epoch 266/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 352ms/epoch - 2ms/step\n","Epoch 267/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 353ms/epoch - 2ms/step\n","Epoch 268/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 369ms/epoch - 2ms/step\n","Epoch 269/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 356ms/epoch - 2ms/step\n","Epoch 270/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 271/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 358ms/epoch - 2ms/step\n","Epoch 272/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 353ms/epoch - 2ms/step\n","Epoch 273/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 348ms/epoch - 2ms/step\n","Epoch 274/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 357ms/epoch - 2ms/step\n","Epoch 275/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 371ms/epoch - 2ms/step\n","Epoch 276/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 341ms/epoch - 2ms/step\n","Epoch 277/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 278/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 359ms/epoch - 2ms/step\n","Epoch 279/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 280/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 322ms/epoch - 2ms/step\n","Epoch 281/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 366ms/epoch - 2ms/step\n","Epoch 282/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 283/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 332ms/epoch - 2ms/step\n","Epoch 284/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 285/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8133 - val_categorical_accuracy: 0.3743 - 344ms/epoch - 2ms/step\n","Epoch 286/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 287/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 320ms/epoch - 2ms/step\n","Epoch 288/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8133 - val_categorical_accuracy: 0.3743 - 341ms/epoch - 2ms/step\n","Epoch 289/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 342ms/epoch - 2ms/step\n","Epoch 290/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 291/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 367ms/epoch - 2ms/step\n","Epoch 292/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 365ms/epoch - 2ms/step\n","Epoch 293/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 337ms/epoch - 2ms/step\n","Epoch 294/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 295/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8133 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 296/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 377ms/epoch - 2ms/step\n","Epoch 297/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 355ms/epoch - 2ms/step\n","Epoch 298/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 380ms/epoch - 2ms/step\n","Epoch 299/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 300/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 301/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 302/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 303/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 304/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 396ms/epoch - 2ms/step\n","Epoch 305/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 378ms/epoch - 2ms/step\n","Epoch 306/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 307/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 308/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 338ms/epoch - 2ms/step\n","Epoch 309/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 371ms/epoch - 2ms/step\n","Epoch 310/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 369ms/epoch - 2ms/step\n","Epoch 311/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 312/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 313/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 354ms/epoch - 2ms/step\n","Epoch 314/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 365ms/epoch - 2ms/step\n","Epoch 315/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 316/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 357ms/epoch - 2ms/step\n","Epoch 317/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 318/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 376ms/epoch - 2ms/step\n","Epoch 319/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 332ms/epoch - 2ms/step\n","Epoch 320/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 321/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 367ms/epoch - 2ms/step\n","Epoch 322/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 323/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 324/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 325/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 380ms/epoch - 2ms/step\n","Epoch 326/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 369ms/epoch - 2ms/step\n","Epoch 327/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 328/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 348ms/epoch - 2ms/step\n","Epoch 329/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 330/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 356ms/epoch - 2ms/step\n","Epoch 331/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 332/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 333/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 337ms/epoch - 2ms/step\n","Epoch 334/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 375ms/epoch - 2ms/step\n","Epoch 335/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 379ms/epoch - 2ms/step\n","Epoch 336/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 337/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 338/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 339/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 341ms/epoch - 2ms/step\n","Epoch 340/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 341/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 342/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 343/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 344/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 319ms/epoch - 2ms/step\n","Epoch 345/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 369ms/epoch - 2ms/step\n","Epoch 346/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 363ms/epoch - 2ms/step\n","Epoch 347/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 332ms/epoch - 2ms/step\n","Epoch 348/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 349/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 359ms/epoch - 2ms/step\n","Epoch 350/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 351/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 371ms/epoch - 2ms/step\n","Epoch 352/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 365ms/epoch - 2ms/step\n","Epoch 353/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 354/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 349ms/epoch - 2ms/step\n","Epoch 355/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 356/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 357/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 358/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 359/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 360/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 352ms/epoch - 2ms/step\n","Epoch 361/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 362/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 383ms/epoch - 2ms/step\n","Epoch 363/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 370ms/epoch - 2ms/step\n","Epoch 364/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 320ms/epoch - 2ms/step\n","Epoch 365/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 366/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 367/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 368/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 369ms/epoch - 2ms/step\n","Epoch 369/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 330ms/epoch - 2ms/step\n","Epoch 370/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 371/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 385ms/epoch - 2ms/step\n","Epoch 372/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 365ms/epoch - 2ms/step\n","Epoch 373/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 374/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 375/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 353ms/epoch - 2ms/step\n","Epoch 376/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 357ms/epoch - 2ms/step\n","Epoch 377/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 343ms/epoch - 2ms/step\n","Epoch 378/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8133 - val_categorical_accuracy: 0.3743 - 352ms/epoch - 2ms/step\n","Epoch 379/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 371ms/epoch - 2ms/step\n","Epoch 380/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 339ms/epoch - 2ms/step\n","Epoch 381/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 382/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 383/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 382ms/epoch - 2ms/step\n","Epoch 384/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 385/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 352ms/epoch - 2ms/step\n","Epoch 386/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 387/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 388/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 380ms/epoch - 2ms/step\n","Epoch 389/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 390/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 391/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 392/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 355ms/epoch - 2ms/step\n","Epoch 393/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 374ms/epoch - 2ms/step\n","Epoch 394/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 355ms/epoch - 2ms/step\n","Epoch 395/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 327ms/epoch - 2ms/step\n","Epoch 396/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 397/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 375ms/epoch - 2ms/step\n","Epoch 398/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 371ms/epoch - 2ms/step\n","Epoch 399/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 402ms/epoch - 2ms/step\n","Epoch 400/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 401/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 393ms/epoch - 2ms/step\n","Epoch 402/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 381ms/epoch - 2ms/step\n","Epoch 403/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 374ms/epoch - 2ms/step\n","Epoch 404/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 365ms/epoch - 2ms/step\n","Epoch 405/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 406/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 321ms/epoch - 2ms/step\n","Epoch 407/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 408/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 366ms/epoch - 2ms/step\n","Epoch 409/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 410/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 411/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 372ms/epoch - 2ms/step\n","Epoch 412/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 413/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 414/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 351ms/epoch - 2ms/step\n","Epoch 415/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 322ms/epoch - 2ms/step\n","Epoch 416/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 380ms/epoch - 2ms/step\n","Epoch 417/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 418/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 367ms/epoch - 2ms/step\n","Epoch 419/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 375ms/epoch - 2ms/step\n","Epoch 420/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 421/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8133 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 422/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 377ms/epoch - 2ms/step\n","Epoch 423/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 374ms/epoch - 2ms/step\n","Epoch 424/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 425/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 342ms/epoch - 2ms/step\n","Epoch 426/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 427/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 363ms/epoch - 2ms/step\n","Epoch 428/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 429/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 430/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 341ms/epoch - 2ms/step\n","Epoch 431/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 380ms/epoch - 2ms/step\n","Epoch 432/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 433/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 323ms/epoch - 2ms/step\n","Epoch 434/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 331ms/epoch - 2ms/step\n","Epoch 435/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 436/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 369ms/epoch - 2ms/step\n","Epoch 437/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 438/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 439/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 334ms/epoch - 2ms/step\n","Epoch 440/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 441/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 371ms/epoch - 2ms/step\n","Epoch 442/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 340ms/epoch - 2ms/step\n","Epoch 443/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 359ms/epoch - 2ms/step\n","Epoch 444/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 375ms/epoch - 2ms/step\n","Epoch 445/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 446/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 447/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 448/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 339ms/epoch - 2ms/step\n","Epoch 449/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 359ms/epoch - 2ms/step\n","Epoch 450/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 341ms/epoch - 2ms/step\n","Epoch 451/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 338ms/epoch - 2ms/step\n","Epoch 452/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 453/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 369ms/epoch - 2ms/step\n","Epoch 454/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 339ms/epoch - 2ms/step\n","Epoch 455/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 456/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 364ms/epoch - 2ms/step\n","Epoch 457/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 378ms/epoch - 2ms/step\n","Epoch 458/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 324ms/epoch - 2ms/step\n","Epoch 459/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 326ms/epoch - 2ms/step\n","Epoch 460/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 343ms/epoch - 2ms/step\n","Epoch 461/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 462/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 463/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 362ms/epoch - 2ms/step\n","Epoch 464/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 373ms/epoch - 2ms/step\n","Epoch 465/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 339ms/epoch - 2ms/step\n","Epoch 466/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 363ms/epoch - 2ms/step\n","Epoch 467/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 468/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 343ms/epoch - 2ms/step\n","Epoch 469/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 470/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 357ms/epoch - 2ms/step\n","Epoch 471/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 383ms/epoch - 2ms/step\n","Epoch 472/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 473/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 366ms/epoch - 2ms/step\n","Epoch 474/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8128 - val_categorical_accuracy: 0.3743 - 372ms/epoch - 2ms/step\n","Epoch 475/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 324ms/epoch - 2ms/step\n","Epoch 476/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 363ms/epoch - 2ms/step\n","Epoch 477/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 345ms/epoch - 2ms/step\n","Epoch 478/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 328ms/epoch - 2ms/step\n","Epoch 479/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 325ms/epoch - 2ms/step\n","Epoch 480/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 349ms/epoch - 2ms/step\n","Epoch 481/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 331ms/epoch - 2ms/step\n","Epoch 482/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 483/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 343ms/epoch - 2ms/step\n","Epoch 484/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 354ms/epoch - 2ms/step\n","Epoch 485/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8129 - val_categorical_accuracy: 0.3743 - 348ms/epoch - 2ms/step\n","Epoch 486/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 487/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 361ms/epoch - 2ms/step\n","Epoch 488/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 337ms/epoch - 2ms/step\n","Epoch 489/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 379ms/epoch - 2ms/step\n","Epoch 490/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 363ms/epoch - 2ms/step\n","Epoch 491/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 343ms/epoch - 2ms/step\n","Epoch 492/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 368ms/epoch - 2ms/step\n","Epoch 493/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 494/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 333ms/epoch - 2ms/step\n","Epoch 495/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 335ms/epoch - 2ms/step\n","Epoch 496/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 360ms/epoch - 2ms/step\n","Epoch 497/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 340ms/epoch - 2ms/step\n","Epoch 498/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8132 - val_categorical_accuracy: 0.3743 - 363ms/epoch - 2ms/step\n","Epoch 499/500\n","197/197 - 0s - loss: 1.7906 - categorical_accuracy: 0.3851 - val_loss: 1.8131 - val_categorical_accuracy: 0.3743 - 329ms/epoch - 2ms/step\n","Epoch 500/500\n","197/197 - 0s - loss: 1.7905 - categorical_accuracy: 0.3851 - val_loss: 1.8130 - val_categorical_accuracy: 0.3743 - 372ms/epoch - 2ms/step\n"]}]},{"cell_type":"code","source":["x_teste.iloc[0]"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"YtzVQ1z44y2e","executionInfo":{"status":"ok","timestamp":1653644441318,"user_tz":180,"elapsed":243,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"4dd38a30-17cb-477a-c91f-4973f6595b34"},"execution_count":84,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Flow.Duration                  1.997361e+07\n","Flow.Bytes.s                   9.011891e-01\n","Flow.Packets.s                 1.501982e-01\n","Flow.IAT.Mean                  9.986806e+06\n","Flow.IAT.Std                   1.412322e+07\n","Flow.IAT.Max                   1.997343e+07\n","Flow.IAT.Min                   1.800000e+02\n","Total.Fwd.Packets              1.000000e+00\n","Total.Backward.Packets         2.000000e+00\n","Total.Length.of.Fwd.Packets    6.000000e+00\n","Total.Length.of.Bwd.Packets    1.200000e+01\n","Name: 9447, dtype: float64"]},"metadata":{},"execution_count":84}]},{"cell_type":"code","source":["inicio_predicao_mlp = time.time()\n","prev_mlp =  np.argmax(model.predict(x_teste), axis=-1)\n","tempo_predicao_mlp = time.time() - inicio_predicao_mlp"],"metadata":{"id":"jTaiZPy9KTwx","executionInfo":{"status":"ok","timestamp":1653644474256,"user_tz":180,"elapsed":671,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":86,"outputs":[]},{"cell_type":"markdown","source":["#### Performance"],"metadata":{"id":"qlzvtFkVKZKE"}},{"cell_type":"code","source":["prev_mlp"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"VHLj9UkfKs9J","executionInfo":{"status":"ok","timestamp":1653644477750,"user_tz":180,"elapsed":267,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"b609ca5c-aded-4e5a-ab34-6ab8acbefd06"},"execution_count":87,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([3, 3, 3, ..., 3, 3, 3])"]},"metadata":{},"execution_count":87}]},{"cell_type":"code","source":["acc_res = metrics.accuracy_score(y_teste,prev_mlp)"],"metadata":{"id":"zN4tDwJmKyGK","executionInfo":{"status":"ok","timestamp":1653644479218,"user_tz":180,"elapsed":377,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"execution_count":88,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653644480574,"user_tz":180,"elapsed":3,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"37327a6d-ba6f-4828-efe3-e75fccbdb293","id":"gk3I0bxeKKfN"},"source":["prec = metrics.precision_score(y_teste,prev_mlp,average='weighted')\n","rec = metrics.recall_score(y_teste,prev_mlp,average='weighted')\n","f1s = metrics.f1_score(y_teste,prev_mlp,average='weighted')"],"execution_count":89,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n","  _warn_prf(average, modifier, msg_start, len(result))\n"]}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653644482882,"user_tz":180,"elapsed":241,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"9b1162b0-cec8-4cfa-e6f4-1de975c0853d","id":"mXsPYtY_KKfN"},"source":["print(f'Precision: {prec:.2f}')\n","print(f'Recall:    {rec:.2f}')\n","print(f'F-1 Score: {f1s:.2f}')\n","print(f'Accuracy:  {acc_res:.2f}')\n","print(\" \")\n","print(f'Tempo Grid:  {tempo_treinamento_mlp:.2f}')\n","print(f'Tempo Execução:  {tempo_predicao_mlp:.2f}')"],"execution_count":90,"outputs":[{"output_type":"stream","name":"stdout","text":["Precision: 0.15\n","Recall:    0.39\n","F-1 Score: 0.22\n","Accuracy:  0.39\n"," \n","Tempo Grid:  131.32\n","Tempo Execução:  0.42\n"]}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":315},"executionInfo":{"status":"ok","timestamp":1653644487006,"user_tz":180,"elapsed":1153,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"32db6f06-f275-4cbf-ffe2-9c68bfb1ebd2","id":"wYO279ydKKfN"},"source":["ax= plt.subplot()\n","array = confusion_matrix(y_teste, prev_mlp)      \n","df_cm = pd.DataFrame(array, range(10),range(10))\n","#plt.figure(figsize = (10,7))\n","sn.set(font_scale=1.4)#for label size\n","sn.heatmap(df_cm, annot=True,annot_kws={\"size\": 14}, ax = ax,fmt=\"d\")# font size\n","\n","ax.set_xlabel('Predicted labels');ax.set_ylabel('True labels'); \n","ax.set_title('Confusion Matrix for Rede Neural');"],"execution_count":91,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","metadata":{"id":"FQ8mFTjENTE4"},"source":["# Persistência\n"]},{"cell_type":"code","metadata":{"id":"HJ-eC0uk5KdR","executionInfo":{"status":"ok","timestamp":1653644491537,"user_tz":180,"elapsed":246,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"source":["import joblib"],"execution_count":92,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653644573639,"user_tz":180,"elapsed":705,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}},"outputId":"84835046-9ab5-4d13-cef1-b573498942fa","id":"ZZ53Obtd5KdR"},"source":["joblib.dump(grid,'rf_model')"],"execution_count":98,"outputs":[{"output_type":"execute_result","data":{"text/plain":["['rf_model']"]},"metadata":{},"execution_count":98}]},{"cell_type":"code","execution_count":99,"metadata":{"id":"_MPKDjgINTE5","executionInfo":{"status":"ok","timestamp":1653644576645,"user_tz":180,"elapsed":683,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[],"source":["model = joblib.load('rf_model')\n","resultados = model.predict(x_teste)"]},{"cell_type":"code","execution_count":100,"metadata":{"id":"qY7RGteKNTE5","outputId":"aedd0c54-407e-469d-b78e-ef2e435abe46","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1653644579689,"user_tz":180,"elapsed":230,"user":{"displayName":"Sidney Loyola de Sá","userId":"06007322353479792623"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["[ 7  7  7 ... 15  7  7]\n"]}],"source":["print(resultados)"]},{"cell_type":"markdown","metadata":{"id":"MC4jEPwjNTE5"},"source":["# Conclusão"]},{"cell_type":"markdown","metadata":{"id":"oqXUxky8NTE5"},"source":["<div align=\"justify\">Foi realizada a classificação dos tráfegos de rede utilizando modelos de AM para exemplificação do minicurso apresentado no SRBC 2022. </div>"]},{"cell_type":"markdown","metadata":{"id":"916QUCACNTE5"},"source":["\n","# References"]},{"cell_type":"markdown","metadata":{"id":"Kyya8SNrNTE5"},"source":["<div>[1] https://www.kaggle.com/datasets/jsrojas/ip-network-traffic-flows-labeled-with-87-apps</div>"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"AmczOTYlNTE5"},"outputs":[],"source":[""]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.8.4rc1"},"colab":{"name":"Classificação de Tráfego.ipynb","provenance":[],"collapsed_sections":[]}},"nbformat":4,"nbformat_minor":0}