{"cells":[{"cell_type":"markdown","metadata":{"id":"yfRGAymdTmBA"},"source":["#**Project: Predicting stock market index using LSTM**\n","\n","**Authors**: Abdul-Razak Bawa Yussif\n"]},{"cell_type":"markdown","metadata":{"id":"0m5kjGJtGNbQ"},"source":["# **Part I:  Data Creation and Exploration**\n","\n","In this module we use various financial resources to create input data for our machine learning models. This includes the following data.\n","- BIST 100 index data: Open price and Close price\n","- Vix index\n","- Interest Rate\n","- Unemployment Rate\n","- Consumer sentiment index\n","- Dollar Index\n","- MACD\n","- RSI\n","- ATR\n","\n","Our data will be the time series data with the following start and end date.\n","\n","- start date: 2016/01/04\n","- end date:  2025/12/31\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":531,"status":"ok","timestamp":1779651836194,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"e9GvUIRynTIp","outputId":"171f320d-7a85-4643-8370-32f1979e4bbe"},"outputs":[{"output_type":"stream","name":"stdout","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"]}],"source":["from google.colab import drive\n","drive.mount('/content/drive')"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"8-Ztn89P6M1A"},"outputs":[],"source":["output_dir_path = \"/content/drive/My Drive/BIST100_Prediction/Final_ Results/\"\n"]},{"cell_type":"markdown","metadata":{"id":"DqXKpN9QjM0H"},"source":["### **Importing basic libraries and APIs**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"WqxVeoCuX3mS","executionInfo":{"status":"ok","timestamp":1779651839442,"user_tz":300,"elapsed":30,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"65b12843-2eb6-4c60-d9b1-f18ba44ef344"},"outputs":[{"output_type":"stream","name":"stdout","text":["The tensorboard extension is already loaded. To reload it, use:\n","  %reload_ext tensorboard\n"]}],"source":["\n","import pandas as pd  ##import necessary libraries\n","import numpy as np\n","import math\n","import matplotlib.pyplot as plt\n","\n","import seaborn as sns\n","sns.set_theme(style=\"whitegrid\")\n","plt.style.use('ggplot')\n","\n","\n","import tensorflow as tf\n","%load_ext tensorboard\n","\n","import warnings\n","warnings.filterwarnings('ignore')\n","\n","import os\n","import datetime as dt"]},{"cell_type":"markdown","metadata":{"id":"o-0dsDrzEaeM"},"source":["## **Loading input data**\n","\n","---\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":238},"executionInfo":{"elapsed":539,"status":"ok","timestamp":1779651843882,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"w6Su9vxu28HO","outputId":"4c8b94b3-9203-4f89-87f2-303f50ebbd52"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["              Open   Close       MACD        RSI        ATR    VIX   USDX  \\\n","Date                                                                        \n","2016-01-04  706.78  705.18 -10.682082  55.204477  13.144286  20.70  98.87   \n","2016-01-05  710.56  706.88 -11.658441  41.781358  12.268571  19.34  99.40   \n","2016-01-06  703.80  711.98 -11.883697  40.686275  12.313571  20.59  99.18   \n","2016-01-07  706.00  714.96 -11.687032  36.471306  12.739286  24.99  98.22   \n","2016-01-08  718.68  706.13 -12.104152  38.192047  12.892857  27.01  98.54   \n","\n","            TR_INT  TR_UNEMP  TR_CCI  \n","Date                                  \n","2016-01-04  2.9632      11.5    91.9  \n","2016-01-05  2.9844      11.5    91.9  \n","2016-01-06  3.0014      11.5    91.9  \n","2016-01-07  2.9983      11.5    91.9  \n","2016-01-08  3.0204      11.5    91.9  "],"text/html":["\n","  <div id=\"df-21f6514d-6853-4a44-a222-19bb57cb1b4e\" 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>Open</th>\n","      <th>Close</th>\n","      <th>MACD</th>\n","      <th>RSI</th>\n","      <th>ATR</th>\n","      <th>VIX</th>\n","      <th>USDX</th>\n","      <th>TR_INT</th>\n","      <th>TR_UNEMP</th>\n","      <th>TR_CCI</th>\n","    </tr>\n","    <tr>\n","      <th>Date</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>2016-01-04</th>\n","      <td>706.78</td>\n","      <td>705.18</td>\n","      <td>-10.682082</td>\n","      <td>55.204477</td>\n","      <td>13.144286</td>\n","      <td>20.70</td>\n","      <td>98.87</td>\n","      <td>2.9632</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","    </tr>\n","    <tr>\n","      <th>2016-01-05</th>\n","      <td>710.56</td>\n","      <td>706.88</td>\n","      <td>-11.658441</td>\n","      <td>41.781358</td>\n","      <td>12.268571</td>\n","      <td>19.34</td>\n","      <td>99.40</td>\n","      <td>2.9844</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","    </tr>\n","    <tr>\n","      <th>2016-01-06</th>\n","      <td>703.80</td>\n","      <td>711.98</td>\n","      <td>-11.883697</td>\n","      <td>40.686275</td>\n","      <td>12.313571</td>\n","      <td>20.59</td>\n","      <td>99.18</td>\n","      <td>3.0014</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","    </tr>\n","    <tr>\n","      <th>2016-01-07</th>\n","      <td>706.00</td>\n","      <td>714.96</td>\n","      <td>-11.687032</td>\n","      <td>36.471306</td>\n","      <td>12.739286</td>\n","      <td>24.99</td>\n","      <td>98.22</td>\n","      <td>2.9983</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","    </tr>\n","    <tr>\n","      <th>2016-01-08</th>\n","      <td>718.68</td>\n","      <td>706.13</td>\n","      <td>-12.104152</td>\n","      <td>38.192047</td>\n","      <td>12.892857</td>\n","      <td>27.01</td>\n","      <td>98.54</td>\n","      <td>3.0204</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","    <div class=\"colab-df-buttons\">\n","\n","  <div class=\"colab-df-container\">\n","    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-21f6514d-6853-4a44-a222-19bb57cb1b4e')\"\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 -960 960 960\">\n","    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n","  </svg>\n","    </button>\n","\n","  <style>\n","    .colab-df-container {\n","      display:flex;\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","    .colab-df-buttons div {\n","      margin-bottom: 4px;\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-21f6514d-6853-4a44-a222-19bb57cb1b4e 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-21f6514d-6853-4a44-a222-19bb57cb1b4e');\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","\n","\n","    </div>\n","  </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"data","summary":"{\n  \"name\": \"data\",\n  \"rows\": 2507,\n  \"fields\": [\n    {\n      \"column\": \"Date\",\n      \"properties\": {\n        \"dtype\": \"date\",\n        \"min\": \"2016-01-04\",\n        \"max\": \"2025-12-31\",\n        \"num_unique_values\": 2507,\n        \"samples\": [\n          \"2024-06-21\",\n          \"2016-03-22\",\n          \"2025-11-24\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Open\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 3633.724714880431,\n        \"min\": 694.87,\n        \"max\": 11561.73,\n        \"num_unique_values\": 2488,\n        \"samples\": [\n          1156.26,\n          1409.85,\n          10403.23\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Close\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 3628.2877955965073,\n        \"min\": 685.68,\n        \"max\": 11529.81,\n        \"num_unique_values\": 2483,\n        \"samples\": [\n          2206.5,\n          1378.37,\n          1080.97\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"MACD\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 94.72808274606571,\n        \"min\": -271.2140287,\n        \"max\": 409.7139467,\n        \"num_unique_values\": 2507,\n        \"samples\": [\n          81.88586816,\n          22.2406607,\n          18.96875237\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"RSI\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 18.435014247103627,\n        \"min\": 6.928568916,\n        \"max\": 100.0,\n        \"num_unique_values\": 2507,\n        \"samples\": [\n          59.29614029,\n          73.90510949,\n          48.99786354\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ATR\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 79.80132139816995,\n        \"min\": 7.094285714,\n        \"max\": 378.6021429,\n        \"num_unique_values\": 2465,\n        \"samples\": [\n          13.04928571,\n          15.67571429,\n          22.68\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"VIX\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 7.345589526840591,\n        \"min\": 9.14,\n        \"max\": 82.69,\n        \"num_unique_values\": 1392,\n        \"samples\": [\n          29.83,\n          13.29,\n          15.09\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"USDX\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 5.030997011920962,\n        \"min\": 88.59,\n        \"max\": 114.11,\n        \"num_unique_values\": 1345,\n        \"samples\": [\n          96.8,\n          99.74,\n          101.36\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_INT\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 12.645052819249385,\n        \"min\": 2.7945,\n        \"max\": 42.9587,\n        \"num_unique_values\": 2464,\n        \"samples\": [\n          2.8903,\n          36.3935,\n          5.4558\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_UNEMP\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.8728669082325027,\n        \"min\": 7.6,\n        \"max\": 15.1,\n        \"num_unique_values\": 58,\n        \"samples\": [\n          11.5,\n          11.8,\n          13.6\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_CCI\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.925625708023312,\n        \"min\": 63.4,\n        \"max\": 95.2,\n        \"num_unique_values\": 91,\n        \"samples\": [\n          80.5,\n          92.3,\n          78.5\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"}},"metadata":{},"execution_count":61}],"source":["data = pd.read_csv('/content/drive/My Drive/BIST100_Prediction/input_data.csv')\n","data['Date']=pd.to_datetime(data['Date']).dt.date\n","data.set_index(\"Date\",inplace=True)\n","data.head()"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":238},"executionInfo":{"elapsed":60,"status":"ok","timestamp":1779651848644,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"KL2KiiGddcAL","outputId":"8a959dec-7e51-4f84-b5b5-a09a691da1ba"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["                Open     Close        MACD        RSI         ATR    VIX  \\\n","Date                                                                       \n","2025-12-25  11378.44  11336.18  114.014327  68.946125  121.179286  13.47   \n","2025-12-26  11340.42  11294.37  106.409596  57.208254  112.651429  13.60   \n","2025-12-29  11305.05  11150.90   87.793929  44.680308  118.197857  14.20   \n","2025-12-30  11172.63  11220.17   77.734315  51.552260  117.310714  14.33   \n","2025-12-31  11249.67  11261.52   72.265566  51.641914  117.155000  14.95   \n","\n","             USDX   TR_INT  TR_UNEMP  TR_CCI  \n","Date                                          \n","2025-12-25    NaN  42.8512       7.8    83.5  \n","2025-12-26  98.02  42.8375       7.8    83.5  \n","2025-12-29  98.04  42.9358       7.8    83.5  \n","2025-12-30  98.24  42.9393       7.8    83.5  \n","2025-12-31  98.32  42.9587       7.8    83.5  "],"text/html":["\n","  <div id=\"df-e4b95795-22bf-4ef1-a51d-36e953873af8\" 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>Open</th>\n","      <th>Close</th>\n","      <th>MACD</th>\n","      <th>RSI</th>\n","      <th>ATR</th>\n","      <th>VIX</th>\n","      <th>USDX</th>\n","      <th>TR_INT</th>\n","      <th>TR_UNEMP</th>\n","      <th>TR_CCI</th>\n","    </tr>\n","    <tr>\n","      <th>Date</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>2025-12-25</th>\n","      <td>11378.44</td>\n","      <td>11336.18</td>\n","      <td>114.014327</td>\n","      <td>68.946125</td>\n","      <td>121.179286</td>\n","      <td>13.47</td>\n","      <td>NaN</td>\n","      <td>42.8512</td>\n","      <td>7.8</td>\n","      <td>83.5</td>\n","    </tr>\n","    <tr>\n","      <th>2025-12-26</th>\n","      <td>11340.42</td>\n","      <td>11294.37</td>\n","      <td>106.409596</td>\n","      <td>57.208254</td>\n","      <td>112.651429</td>\n","      <td>13.60</td>\n","      <td>98.02</td>\n","      <td>42.8375</td>\n","      <td>7.8</td>\n","      <td>83.5</td>\n","    </tr>\n","    <tr>\n","      <th>2025-12-29</th>\n","      <td>11305.05</td>\n","      <td>11150.90</td>\n","      <td>87.793929</td>\n","      <td>44.680308</td>\n","      <td>118.197857</td>\n","      <td>14.20</td>\n","      <td>98.04</td>\n","      <td>42.9358</td>\n","      <td>7.8</td>\n","      <td>83.5</td>\n","    </tr>\n","    <tr>\n","      <th>2025-12-30</th>\n","      <td>11172.63</td>\n","      <td>11220.17</td>\n","      <td>77.734315</td>\n","      <td>51.552260</td>\n","      <td>117.310714</td>\n","      <td>14.33</td>\n","      <td>98.24</td>\n","      <td>42.9393</td>\n","      <td>7.8</td>\n","      <td>83.5</td>\n","    </tr>\n","    <tr>\n","      <th>2025-12-31</th>\n","      <td>11249.67</td>\n","      <td>11261.52</td>\n","      <td>72.265566</td>\n","      <td>51.641914</td>\n","      <td>117.155000</td>\n","      <td>14.95</td>\n","      <td>98.32</td>\n","      <td>42.9587</td>\n","      <td>7.8</td>\n","      <td>83.5</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","    <div class=\"colab-df-buttons\">\n","\n","  <div class=\"colab-df-container\">\n","    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e4b95795-22bf-4ef1-a51d-36e953873af8')\"\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 -960 960 960\">\n","    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n","  </svg>\n","    </button>\n","\n","  <style>\n","    .colab-df-container {\n","      display:flex;\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","    .colab-df-buttons div {\n","      margin-bottom: 4px;\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-e4b95795-22bf-4ef1-a51d-36e953873af8 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-e4b95795-22bf-4ef1-a51d-36e953873af8');\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","\n","\n","    </div>\n","  </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n  \"name\": \"data\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"Date\",\n      \"properties\": {\n        \"dtype\": \"date\",\n        \"min\": \"2025-12-25\",\n        \"max\": \"2025-12-31\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"2025-12-26\",\n          \"2025-12-31\",\n          \"2025-12-29\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Open\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 80.60655289739206,\n        \"min\": 11172.63,\n        \"max\": 11378.44,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          11340.42,\n          11249.67,\n          11305.05\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Close\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 71.07127879811962,\n        \"min\": 11150.9,\n        \"max\": 11336.18,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          11294.37,\n          11261.52,\n          11150.9\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"MACD\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 18.043561715927304,\n        \"min\": 72.26556579,\n        \"max\": 114.0143273,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          106.4095964,\n          72.26556579,\n          87.79392858\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"RSI\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 9.066998847783863,\n        \"min\": 44.68030753,\n        \"max\": 68.94612504,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          57.20825372,\n          51.64191419,\n          44.68030753\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ATR\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 3.061264209469772,\n        \"min\": 112.6514286,\n        \"max\": 121.1792857,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          112.6514286,\n          117.155,\n          118.1978571\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"VIX\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.5982892277151575,\n        \"min\": 13.47,\n        \"max\": 14.95,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          13.6,\n          14.95,\n          14.2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"USDX\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.1482115605027675,\n        \"min\": 98.02,\n        \"max\": 98.32,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          98.04,\n          98.32,\n          98.02\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_INT\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.055808287914969675,\n        \"min\": 42.8375,\n        \"max\": 42.9587,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          42.8375,\n          42.9587,\n          42.9358\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_UNEMP\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.0,\n        \"min\": 7.8,\n        \"max\": 7.8,\n        \"num_unique_values\": 1,\n        \"samples\": [\n          7.8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_CCI\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.0,\n        \"min\": 83.5,\n        \"max\": 83.5,\n        \"num_unique_values\": 1,\n        \"samples\": [\n          83.5\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"}},"metadata":{},"execution_count":62}],"source":["data.tail()"]},{"cell_type":"markdown","metadata":{"id":"e5EWNJs65Ecs"},"source":["### **Data Exploration**"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":477},"executionInfo":{"elapsed":280,"status":"ok","timestamp":1779651851978,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"EIVKYEZm5Di2","outputId":"e330a937-47a2-4832-d522-2f68c286f333"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x500 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["df = data.copy()\n","df['MA_50'] = df['Close'].rolling(50).mean()\n","df['MA_200'] = df['Close'].rolling(200).mean()\n","\n","fig = plt.figure(figsize=(10,5))\n","fig.set(facecolor = \"white\")\n","plt.plot(df['Close'], 'mediumblue',label=['BIST100 Close Price'], linewidth = 2.2)\n","plt.plot(df['MA_50'],'red', label=['50-day MA'], linewidth = 2.2)\n","plt.plot(df['MA_200'],'darkgreen', label=['200-day MA'], linewidth = 2.2)\n","plt.legend(['BIST100 Close Price', '50-day Moving Average', '200-day Moving Average'], loc='upper left')\n","plt.title('')\n","plt.xlabel('Time (years)')\n","plt.ylabel('Close price')\n","#fig.savefig(output_dir_path+ \"original_data_plus_moving_averages.png\",dpi=600)\n","plt.show()"]},{"cell_type":"markdown","metadata":{"id":"y-Vap-SkxQjy"},"source":["####  **Correlation heatmap**"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":523},"executionInfo":{"elapsed":442,"status":"ok","timestamp":1779651858874,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"1Fv9RQuWqbzk","outputId":"4e73ec5c-23d1-436f-89e7-b35fc44027e7"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x500 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["\n","fig = plt.figure(figsize= (10,5))\n","sns.heatmap(data.corr(), annot=True)\n","sns.set_style(\"whitegrid\")\n","#fig.savefig(output_dir_path+\"correlation_heatmap.png\",dpi=600)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":457},"executionInfo":{"elapsed":1633,"status":"ok","timestamp":1779652055130,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"qACPoxnPiNSC","outputId":"fd1b7c14-f017-4b7c-c303-f80d674eb34e"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x500 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["data = data.iloc[:, 1:]\n","fig = plt.figure(figsize= (10,5))\n","sns.heatmap(data.corr(), annot=True)\n","sns.set_style(\"whitegrid\")\n","fig.savefig(output_dir_path+\"correlation_heatmap2.png\",dpi=600)\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":875},"executionInfo":{"elapsed":2027,"status":"ok","timestamp":1779652747777,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"pcaukJvvNTPN","outputId":"273efd42-223b-4cef-c0b0-3e6a89ffaabe"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 2000x1200 with 0 Axes>"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["<Figure size 1200x1000 with 8 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["fig = plt.figure(figsize = (20, 12))\n","#fig.axes.get_yaxis().set_visible(False)\n","data.plot(subplots = True, figsize = (12,10),grid=False)\n","sns.set_style(\"whitegrid\")\n","#fig.savefig(output_dir_path+\"timeseries.png\",dpi=600)\n","plt.show()"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":300},"executionInfo":{"elapsed":33,"status":"ok","timestamp":1779652747827,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"lpHdLNaUvDja","outputId":"2886179b-376b-4d21-ee3c-63eb34535866"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["              MACD          RSI          ATR          VIX         USDX  \\\n","count  2507.000000  2507.000000  2507.000000  2448.000000  2500.000000   \n","mean     28.982194    56.322978    74.583387    18.487373    98.509804   \n","std      94.728083    18.435014    79.801321     7.345590     5.030997   \n","min    -271.214029     6.928569     7.094286     9.140000    88.590000   \n","25%      -8.054593    42.810082    15.678571    13.525000    94.720000   \n","50%       9.816481    55.654744    23.972857    16.670000    97.815000   \n","75%      35.979351    69.953212   144.483929    21.492500   102.400000   \n","max     409.713947   100.000000   378.602143    82.690000   114.110000   \n","\n","            TR_INT     TR_UNEMP       TR_CCI  \n","count  2507.000000  2507.000000  2507.000000  \n","mean     14.582208    10.867012    82.331990  \n","std      12.645053     1.872867     6.925626  \n","min       2.794500     7.600000    63.400000  \n","25%       4.677150     9.300000    78.400000  \n","50%       7.661900    10.600000    81.100000  \n","75%      25.999000    12.500000    87.700000  \n","max      42.958700    15.100000    95.200000  "],"text/html":["\n","  <div id=\"df-e503e688-55e6-4319-a9a8-398fffb1e4e4\" 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>MACD</th>\n","      <th>RSI</th>\n","      <th>ATR</th>\n","      <th>VIX</th>\n","      <th>USDX</th>\n","      <th>TR_INT</th>\n","      <th>TR_UNEMP</th>\n","      <th>TR_CCI</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>count</th>\n","      <td>2507.000000</td>\n","      <td>2507.000000</td>\n","      <td>2507.000000</td>\n","      <td>2448.000000</td>\n","      <td>2500.000000</td>\n","      <td>2507.000000</td>\n","      <td>2507.000000</td>\n","      <td>2507.000000</td>\n","    </tr>\n","    <tr>\n","      <th>mean</th>\n","      <td>28.982194</td>\n","      <td>56.322978</td>\n","      <td>74.583387</td>\n","      <td>18.487373</td>\n","      <td>98.509804</td>\n","      <td>14.582208</td>\n","      <td>10.867012</td>\n","      <td>82.331990</td>\n","    </tr>\n","    <tr>\n","      <th>std</th>\n","      <td>94.728083</td>\n","      <td>18.435014</td>\n","      <td>79.801321</td>\n","      <td>7.345590</td>\n","      <td>5.030997</td>\n","      <td>12.645053</td>\n","      <td>1.872867</td>\n","      <td>6.925626</td>\n","    </tr>\n","    <tr>\n","      <th>min</th>\n","      <td>-271.214029</td>\n","      <td>6.928569</td>\n","      <td>7.094286</td>\n","      <td>9.140000</td>\n","      <td>88.590000</td>\n","      <td>2.794500</td>\n","      <td>7.600000</td>\n","      <td>63.400000</td>\n","    </tr>\n","    <tr>\n","      <th>25%</th>\n","      <td>-8.054593</td>\n","      <td>42.810082</td>\n","      <td>15.678571</td>\n","      <td>13.525000</td>\n","      <td>94.720000</td>\n","      <td>4.677150</td>\n","      <td>9.300000</td>\n","      <td>78.400000</td>\n","    </tr>\n","    <tr>\n","      <th>50%</th>\n","      <td>9.816481</td>\n","      <td>55.654744</td>\n","      <td>23.972857</td>\n","      <td>16.670000</td>\n","      <td>97.815000</td>\n","      <td>7.661900</td>\n","      <td>10.600000</td>\n","      <td>81.100000</td>\n","    </tr>\n","    <tr>\n","      <th>75%</th>\n","      <td>35.979351</td>\n","      <td>69.953212</td>\n","      <td>144.483929</td>\n","      <td>21.492500</td>\n","      <td>102.400000</td>\n","      <td>25.999000</td>\n","      <td>12.500000</td>\n","      <td>87.700000</td>\n","    </tr>\n","    <tr>\n","      <th>max</th>\n","      <td>409.713947</td>\n","      <td>100.000000</td>\n","      <td>378.602143</td>\n","      <td>82.690000</td>\n","      <td>114.110000</td>\n","      <td>42.958700</td>\n","      <td>15.100000</td>\n","      <td>95.200000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","    <div class=\"colab-df-buttons\">\n","\n","  <div class=\"colab-df-container\">\n","    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e503e688-55e6-4319-a9a8-398fffb1e4e4')\"\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 -960 960 960\">\n","    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n","  </svg>\n","    </button>\n","\n","  <style>\n","    .colab-df-container {\n","      display:flex;\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","    .colab-df-buttons div {\n","      margin-bottom: 4px;\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-e503e688-55e6-4319-a9a8-398fffb1e4e4 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-e503e688-55e6-4319-a9a8-398fffb1e4e4');\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","\n","\n","    </div>\n","  </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n  \"name\": \"data\",\n  \"rows\": 8,\n  \"fields\": [\n    {\n      \"column\": \"MACD\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 890.6553490173422,\n        \"min\": -271.2140287,\n        \"max\": 2507.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          28.98219353692461,\n          9.816480569,\n          2507.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"RSI\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 869.1582054039663,\n        \"min\": 6.928568916,\n        \"max\": 2507.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          56.32297822100319,\n          55.65474368,\n          2507.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ATR\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 858.3182028360562,\n        \"min\": 7.094285714,\n        \"max\": 2507.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          74.58338680313362,\n          23.97285714,\n          2507.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"VIX\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 857.2904348452448,\n        \"min\": 7.345589526840591,\n        \"max\": 2448.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          18.48737336601307,\n          16.67,\n          2448.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"USDX\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 854.1882334074608,\n        \"min\": 5.030997011920962,\n        \"max\": 2500.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          98.509804,\n          97.815,\n          2500.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_INT\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 880.834210753245,\n        \"min\": 2.7945,\n        \"max\": 2507.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          14.582207818109294,\n          7.6619,\n          2507.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_UNEMP\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 882.9404137071109,\n        \"min\": 1.8728669082325027,\n        \"max\": 2507.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          10.867012365376944,\n          10.6,\n          2507.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_CCI\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 861.794592664048,\n        \"min\": 6.925625708023312,\n        \"max\": 2507.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          82.33199042680494,\n          81.1,\n          2507.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"}},"metadata":{},"execution_count":68}],"source":["data.describe()"]},{"cell_type":"markdown","metadata":{"id":"gMsCJU6A38Zk"},"source":["### **Denoising Closing Price**"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":238},"executionInfo":{"elapsed":8,"status":"ok","timestamp":1779652747867,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"},"user_tz":300},"id":"6TYqklCZ32TS","outputId":"a5b0516b-a5f8-42a2-9525-76f3e0368ded"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["                 MACD        RSI        ATR    VIX   USDX  TR_INT  TR_UNEMP  \\\n","Date                                                                          \n","2016-01-04 -10.682082  55.204477  13.144286  20.70  98.87  2.9632      11.5   \n","2016-01-05 -11.658441  41.781358  12.268571  19.34  99.40  2.9844      11.5   \n","2016-01-06 -11.883697  40.686275  12.313571  20.59  99.18  3.0014      11.5   \n","2016-01-07 -11.687032  36.471306  12.739286  24.99  98.22  2.9983      11.5   \n","2016-01-08 -12.104152  38.192047  12.892857  27.01  98.54  3.0204      11.5   \n","\n","            TR_CCI    Close  \n","Date                         \n","2016-01-04    91.9 -9.32775  \n","2016-01-05    91.9 -9.32775  \n","2016-01-06    91.9 -9.32775  \n","2016-01-07    91.9 -9.32775  \n","2016-01-08    91.9 -9.32775  "],"text/html":["\n","  <div id=\"df-ded5c4c4-aeb2-4dd3-86f5-5fbe25840f5f\" 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>MACD</th>\n","      <th>RSI</th>\n","      <th>ATR</th>\n","      <th>VIX</th>\n","      <th>USDX</th>\n","      <th>TR_INT</th>\n","      <th>TR_UNEMP</th>\n","      <th>TR_CCI</th>\n","      <th>Close</th>\n","    </tr>\n","    <tr>\n","      <th>Date</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>2016-01-04</th>\n","      <td>-10.682082</td>\n","      <td>55.204477</td>\n","      <td>13.144286</td>\n","      <td>20.70</td>\n","      <td>98.87</td>\n","      <td>2.9632</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","      <td>-9.32775</td>\n","    </tr>\n","    <tr>\n","      <th>2016-01-05</th>\n","      <td>-11.658441</td>\n","      <td>41.781358</td>\n","      <td>12.268571</td>\n","      <td>19.34</td>\n","      <td>99.40</td>\n","      <td>2.9844</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","      <td>-9.32775</td>\n","    </tr>\n","    <tr>\n","      <th>2016-01-06</th>\n","      <td>-11.883697</td>\n","      <td>40.686275</td>\n","      <td>12.313571</td>\n","      <td>20.59</td>\n","      <td>99.18</td>\n","      <td>3.0014</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","      <td>-9.32775</td>\n","    </tr>\n","    <tr>\n","      <th>2016-01-07</th>\n","      <td>-11.687032</td>\n","      <td>36.471306</td>\n","      <td>12.739286</td>\n","      <td>24.99</td>\n","      <td>98.22</td>\n","      <td>2.9983</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","      <td>-9.32775</td>\n","    </tr>\n","    <tr>\n","      <th>2016-01-08</th>\n","      <td>-12.104152</td>\n","      <td>38.192047</td>\n","      <td>12.892857</td>\n","      <td>27.01</td>\n","      <td>98.54</td>\n","      <td>3.0204</td>\n","      <td>11.5</td>\n","      <td>91.9</td>\n","      <td>-9.32775</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","    <div class=\"colab-df-buttons\">\n","\n","  <div class=\"colab-df-container\">\n","    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ded5c4c4-aeb2-4dd3-86f5-5fbe25840f5f')\"\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 -960 960 960\">\n","    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n","  </svg>\n","    </button>\n","\n","  <style>\n","    .colab-df-container {\n","      display:flex;\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","    .colab-df-buttons div {\n","      margin-bottom: 4px;\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-ded5c4c4-aeb2-4dd3-86f5-5fbe25840f5f 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-ded5c4c4-aeb2-4dd3-86f5-5fbe25840f5f');\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","\n","\n","    </div>\n","  </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"data","summary":"{\n  \"name\": \"data\",\n  \"rows\": 2507,\n  \"fields\": [\n    {\n      \"column\": \"Date\",\n      \"properties\": {\n        \"dtype\": \"date\",\n        \"min\": \"2016-01-04\",\n        \"max\": \"2025-12-31\",\n        \"num_unique_values\": 2507,\n        \"samples\": [\n          \"2024-06-21\",\n          \"2016-03-22\",\n          \"2025-11-24\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"MACD\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 94.72808274606571,\n        \"min\": -271.2140287,\n        \"max\": 409.7139467,\n        \"num_unique_values\": 2507,\n        \"samples\": [\n          81.88586816,\n          22.2406607,\n          18.96875237\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"RSI\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 18.435014247103627,\n        \"min\": 6.928568916,\n        \"max\": 100.0,\n        \"num_unique_values\": 2507,\n        \"samples\": [\n          59.29614029,\n          73.90510949,\n          48.99786354\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ATR\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 79.80132139816995,\n        \"min\": 7.094285714,\n        \"max\": 378.6021429,\n        \"num_unique_values\": 2465,\n        \"samples\": [\n          13.04928571,\n          15.67571429,\n          22.68\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"VIX\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 7.345589526840591,\n        \"min\": 9.14,\n        \"max\": 82.69,\n        \"num_unique_values\": 1392,\n        \"samples\": [\n          29.83,\n          13.29,\n          15.09\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"USDX\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 5.030997011920962,\n        \"min\": 88.59,\n        \"max\": 114.11,\n        \"num_unique_values\": 1345,\n        \"samples\": [\n          96.8,\n          99.74,\n          101.36\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_INT\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 12.645052819249385,\n        \"min\": 2.7945,\n        \"max\": 42.9587,\n        \"num_unique_values\": 2464,\n        \"samples\": [\n          2.8903,\n          36.3935,\n          5.4558\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_UNEMP\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.8728669082325027,\n        \"min\": 7.6,\n        \"max\": 15.1,\n        \"num_unique_values\": 58,\n        \"samples\": [\n          11.5,\n          11.8,\n          13.6\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TR_CCI\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.925625708023312,\n        \"min\": 63.4,\n        \"max\": 95.2,\n        \"num_unique_values\": 91,\n        \"samples\": [\n          80.5,\n          92.3,\n          78.5\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Close\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 93.00842202340321,\n        \"min\": -258.53003343117746,\n        \"max\": 399.0037676193992,\n        \"num_unique_values\": 901,\n        \"samples\": [\n          -21.822474484398906,\n          79.31378077487835,\n          214.70654859762328\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"}},"metadata":{},"execution_count":69}],"source":["from skimage.restoration import (denoise_wavelet, estimate_sigma)\n","\n","data['Close']= denoise_wavelet(data.iloc[:, 0], wavelet='haar',\n","                                      #method='BayesShrink',\n","                                      method='VisuShrink',\n","                                      mode='soft', rescale_sigma = True)\n","data.head()"]},{"cell_type":"markdown","metadata":{"id":"h7CN1Pi07Nzi"},"source":["### **Machine learning libraries**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Fi67Y5BNKB1k"},"outputs":[],"source":["import tensorflow\n","from tensorflow.keras.models import Sequential\n","from tensorflow.keras.layers import LSTM\n","from tensorflow.keras.layers import Dense\n","from tensorflow.keras.layers import Dropout\n","from tensorflow import keras\n","from tensorflow.keras import optimizers\n","from sklearn.preprocessing import MinMaxScaler\n","from sklearn.metrics import mean_squared_error\n","from sklearn.metrics import mean_absolute_error\n","from sklearn.metrics import r2_score\n","import time\n","import math\n"]},{"cell_type":"markdown","metadata":{"id":"ifiXXSC7KluW"},"source":["### **Supporting functions**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"ikplCNqpJ8CN"},"outputs":[],"source":["def mean_absolute_percentage_error(y_true, y_pred):\n","    return (np.mean(np.abs((y_true - y_pred)/(y_true))*100)) #some issues with zero denominator\n","\n","def calculate_scores(y_true, y_pred):\n","  rmse = math.sqrt(mean_squared_error(y_true, y_pred))\n","  #R2_score = r2_score(y_true, y_pred)\n","  R = np.corrcoef(y_true, y_pred)\n","  #mae = mean_absolute_error(y_true, y_pred)\n","  mape = mean_absolute_percentage_error(y_true, y_pred)\n","  #dic = {'rmse':rmse, 'R2_score': R2_score, 'R':R[0,1], 'mae': mae, 'mape': mape}\n","  dic = {'rmse':rmse, 'R': R[0,1], 'mape': mape}\n","  return (dic)\n","\n","\n","\n","def DatasetCreation(dataset, time_step = 1):  ##defining a function that gives a dataset and a time step, which then returns the input and output data\n","   DataX, DataY = [], []\n","   for i in range(len(dataset)- time_step -1):\n","         a = dataset[i:(i+ time_step), ]\n","         DataX.append(a)\n","         DataY.append(dataset[i + time_step, 0]) #ydata consists close price\n","   return np.array(DataX), np.array(DataY)\n","\n","\n","def data_split(data, split = 0.2):\n","  #======= creating training and test data===#\n","  l1   = int(len(data) * (1- split))\n","  l2    = len(data) - l1\n","  data1  = data.iloc[0:l1,:]\n","  data2   = data.iloc[l1:len(data),:]\n","  return data1, data2\n","\n","def min_max_transform(data, feature_range=(0, 1)):\n","   scaler = MinMaxScaler(feature_range)\n","   return scaler.fit_transform(data)\n","\n","def min_max_inverse_transform(data_scaled, min_original, max_original):\n","    return min_original +  data_scaled*(max_original - min_original)\n","\n","\n","\n","def write_dic_to_file(dic_name, file_name):\n","  file = open(file_name, 'w')\n","  file.write(str(dic_name))\n","  file.close()\n","\n","import ast\n","def read_dic_from_file(file_name):\n","  file = open(file_name, \"r\")\n","  contents = file.read()\n","  dictionary = ast.literal_eval(contents)\n","  file.close()\n","  return dictionary\n","\n"]},{"cell_type":"markdown","metadata":{"id":"ZYBcGcR7LSZW"},"source":["## **Supporting functions for results visualization**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"7mNqwC_BQ-o7"},"outputs":[],"source":["def test_scores_plot(model_output):\n","  neurons = model_output['avg_scores']['neurons']\n","  rmse = model_output['avg_scores']['rmse']\n","  #mae =  model_output['avg_scores']['mae']\n","  mape =  model_output['avg_scores']['mape']\n","  #R2 =   model_output['avg_scores']['R2']\n","  R =    model_output['avg_scores']['R']\n","  #time =  model_output['avg_scores']['elapsed_time']\n","\n","  fig = plt.figure(figsize = (18, 4))\n","  plt.subplot(131)\n","  plt.plot(neurons, rmse, '--o', linewidth = 2, color = 'indigo')\n","  plt.title(\"(a)\")\n","  plt.xlabel(\"Neurons\")\n","  plt.ylabel(\"Avg. RMSE\")\n","  sns.set_style(\"whitegrid\")\n","\n","\n","  plt.subplot(132)\n","  plt.plot(neurons, mape, '--o', linewidth = 2, color = 'darkgreen')\n","  plt.title(\"(b)\")\n","  plt.xlabel(\"Neurons\")\n","  plt.ylabel(\"Avg. MAPE\")\n","\n","\n","  plt.subplot(133)\n","  plt.plot(neurons, R, '--o', linewidth = 2, color = 'darkred')\n","  plt.title(\"(c)\")\n","  plt.xlabel(\"Neurons\")\n","  plt.ylabel(\"Avg. R \")\n","\n","  fig.savefig(output_dir_path+\"multiple_avg_scores_plots.png\",dpi=600)\n","  plt.show()\n","\n","\n","def true_pred_plot(model_output):\n","\n","  y_train = model_output['datasets']['y_train']\n","  y_test =  model_output['datasets']['y_test']\n","\n","  train_pred = model_output['best_model']['train_predictions']\n","  test_pred = model_output['best_model']['test_predictions']\n","\n","  ##====== Visualizing true vs predicted plots ========#\n","  fig = plt.figure(figsize= (14,5))\n","  plt.subplot(121)\n","  #sns.relplot(x = y_train_original, y = train_pred_original)\n","  plt.scatter(y_train, train_pred, marker= \"+\", color = 'mediumblue')\n","  identity_line = np.linspace(max(min(y_train), min(train_pred)), min(max(y_train), max(train_pred)))\n","  plt.plot(identity_line, identity_line, color=\"red\", linestyle=\"dashed\", linewidth= 2.5)\n","\n","  plt.xlabel(\"True\")\n","  plt.ylabel(\"Predicted\")\n","  plt.title(\"(a)\")\n","\n","  plt.subplot(122)\n","  #sns.relplot(x = y_test_original, y = test_pred_original)\n","  plt.scatter(y_test, test_pred, marker = \"+\", color = 'mediumblue')\n","  identity_line = np.linspace(max(min(y_test), min(test_pred)), min(max(y_test), max(test_pred)))\n","  plt.plot(identity_line, identity_line, color=\"red\", linestyle=\"dashed\", linewidth= 2.5)\n","  plt.xlabel(\"True\")\n","  plt.ylabel(\"Predicted\")\n","  plt.title(\"(b)\")\n","  fig.savefig(output_dir_path+\"True_vs_predicted_plot.png\", dpi=600)\n","  plt.show()\n","\n","\n","def prediction_plot(model_output):\n","  time_step =  model_output['hyper_parameters']['time_step']\n","  best_replicate = model_output['best_model']['replicate']\n","\n","  data = model_output['datasets']['data']\n","\n","  train_predict_plot_data = np.empty_like(data.values[:,0])# extracting closing price\n","  train_predict_plot_data[:] = np.nan\n","\n","  test_predict_plot_data = np.empty_like(data.values[:,0])\n","  test_predict_plot_data[:] = np.nan\n","\n","  fig1 = plt.figure(figsize = (18,12))\n","\n","  plt.subplot(231)\n","\n","  train_pred = model_output['train_predictions'][0][best_replicate]\n","  test_pred = model_output['test_predictions'][0][best_replicate]\n","\n","\n","  train_predict_plot_data[time_step:len(train_pred)+ time_step] =  train_pred\n","  test_predict_plot_data[len(train_pred)+(time_step*2)+1:len(data.values)-1] = test_pred\n","\n","  plt.plot(data.values[:,0],'k',linewidth = 1.5)\n","  plt.plot(train_predict_plot_data,'mediumblue',linewidth = 1.5)\n","  plt.plot(test_predict_plot_data,'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(a)\")\n","  plt.legend(['True value', 'Predicted value in train set', 'Predicted value in test set'], loc = 'upper left')\n","\n","  plt.subplot(232)\n","\n","  train_pred = model_output['train_predictions'][1][best_replicate]\n","  test_pred = model_output['test_predictions'][1][best_replicate]\n","\n","  train_predict_plot_data[time_step:len(train_pred)+ time_step] =  train_pred\n","  test_predict_plot_data[len(train_pred)+(time_step*2)+1:len(data.values)-1] =  test_pred\n","\n","  plt.plot(data.values[:,0],'k',linewidth = 1.5)\n","  plt.plot(train_predict_plot_data,'mediumblue',linewidth = 1.5)\n","  plt.plot(test_predict_plot_data,'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(b)\")\n","  plt.legend(['True value', 'Predicted value in train set', 'Predicted value in test set'], loc = 'upper left')\n","\n","  plt.subplot(233)\n","\n","  train_pred = model_output['train_predictions'][2][best_replicate]\n","  test_pred = model_output['test_predictions'][2][best_replicate]\n","\n","  train_predict_plot_data[time_step:len(train_pred)+ time_step] = train_pred\n","  test_predict_plot_data[len(train_pred)+(time_step*2)+1:len(data.values)-1] =  test_pred\n","\n","  plt.plot(data.values[:,0],'k',linewidth = 1.5)\n","  plt.plot(train_predict_plot_data,'mediumblue',linewidth = 1.5)\n","  plt.plot(test_predict_plot_data,'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(c)\")\n","  plt.legend(['True value', 'Predicted value in train set', 'Predicted value in test set'], loc = 'upper left')\n","\n","\n","  plt.subplot(234)\n","\n","  train_pred = model_output['train_predictions'][3][best_replicate]\n","  test_pred = model_output['test_predictions'][3][best_replicate]\n","\n","  train_predict_plot_data[time_step:len(train_pred)+ time_step] = train_pred\n","  test_predict_plot_data[len(train_pred)+(time_step*2)+1:len(data.values)-1] = test_pred\n","\n","  plt.plot(data.values[:,0],'k',linewidth = 1.5)\n","  plt.plot(train_predict_plot_data,'mediumblue',linewidth = 1.5)\n","  plt.plot(test_predict_plot_data,'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(d)\")\n","  plt.legend(['True value', 'Predicted value in train set', 'Predicted value in test set'], loc = 'upper left')\n","\n","\n","  plt.subplot(235)\n","\n","  train_pred = model_output['train_predictions'][4][best_replicate]\n","  test_pred = model_output['test_predictions'][4][best_replicate]\n","\n","  train_predict_plot_data[time_step:len(train_pred)+ time_step] = train_pred\n","  test_predict_plot_data[len(train_pred)+(time_step*2)+1:len(data.values)-1] = test_pred\n","\n","  plt.plot(data.values[:,0],'k',linewidth = 1.5)\n","  plt.plot(train_predict_plot_data,'mediumblue',linewidth = 1.5)\n","  plt.plot(test_predict_plot_data,'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(e)\")\n","  plt.legend(['True value', 'Predicted value in train set', 'Predicted value in test set'], loc = 'upper left')\n","\n","  plt.subplot(236)\n","\n","  train_pred = model_output['train_predictions'][5][best_replicate]\n","  test_pred = model_output['test_predictions'][5][best_replicate]\n","\n","  train_predict_plot_data[time_step:len(train_pred)+ time_step] = train_pred\n","  test_predict_plot_data[len(train_pred)+(time_step*2)+1:len(data.values)-1] = test_pred\n","\n","  plt.plot(data.values[:,0],'k',linewidth = 1.5)\n","  plt.plot(train_predict_plot_data,'mediumblue',linewidth = 1.5)\n","  plt.plot(test_predict_plot_data,'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(f)\")\n","  plt.legend(['True value', 'Predicted value in train set', 'Predicted value in test set'], loc = 'upper left')\n","\n","\n","  fig1.savefig(output_dir_path+\"predictions_plots_fullset.png\",dpi=600)\n","  plt.show()\n","\n","  fig2 = plt.figure(figsize = (18,12))\n","\n","  plt.subplot(231)\n","  plt.plot(data.values[len(train_pred)+(time_step*2)+1:-1, 0],'mediumblue',linewidth = 1.5)\n","  plt.plot(model_output['test_predictions'][0][best_replicate], 'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(a)\")\n","  plt.legend(['True value', 'Predicted value'], loc='upper left')\n","\n","\n","  plt.subplot(232)\n","\n","  plt.plot(data.values[len(train_pred)+(time_step*2)+1:-1, 0],'mediumblue',linewidth = 1.5)\n","  plt.plot(model_output['test_predictions'][1][best_replicate], 'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(b)\")\n","  plt.legend(['True value', 'Predicted value'], loc='upper left')\n","\n","\n","  plt.subplot(233)\n","\n","  plt.plot(data.values[len(train_pred)+(time_step*2)+1:-1, 0],'mediumblue',linewidth = 1.5)\n","  plt.plot(model_output['test_predictions'][2][best_replicate], 'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(c)\")\n","  plt.legend(['True value', 'Predicted value'], loc='upper left')\n","\n","\n","\n","  plt.subplot(234)\n","\n","  plt.plot(data.values[len(train_pred)+(time_step*2)+1:-1, 0],'mediumblue',linewidth = 1.5)\n","  plt.plot(model_output['test_predictions'][3][best_replicate], 'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(d)\")\n","  plt.legend(['True value', 'Predicted value'], loc='upper left')\n","\n","  plt.subplot(235)\n","\n","  plt.plot(data.values[len(train_pred)+(time_step*2)+1:-1, 0],'mediumblue',linewidth = 1.5)\n","  plt.plot(model_output['test_predictions'][4][best_replicate], 'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(e)\")\n","  plt.legend(['True value', 'Predicted value'], loc='upper left')\n","\n","\n","  plt.subplot(236)\n","\n","  plt.plot(data.values[len(train_pred)+(time_step*2)+1:-1, 0],'mediumblue',linewidth = 1.5)\n","  plt.plot(model_output['test_predictions'][5][best_replicate], 'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(f)\")\n","  plt.legend(['True value', 'Predicted value'], loc='upper left')\n","\n","\n","\n","  fig2.savefig(output_dir_path+\"predictions_plots_testset.png\",dpi=600)\n","\n","  plt.show()\n","\n","def best_model_prediction_plot(model_output):\n","\n","  time_step =  model_output['hyper_parameters']['time_step']\n","\n","  data = model_output['datasets']['data']\n","\n","  train_predict_plot_data = np.empty_like(data.values[:,0])# extracting closing price\n","  train_predict_plot_data[:] = np.nan\n","\n","  test_predict_plot_data = np.empty_like(data.values[:,0])\n","  test_predict_plot_data[:] = np.nan\n","\n","\n","  fig = plt.figure(figsize = (14,5))\n","\n","  plt.subplot(121)\n","\n","  train_pred = model_output['best_model']['train_predictions']\n","  test_pred = model_output['best_model']['test_predictions']\n","\n","  train_predict_plot_data[time_step:len(train_pred)+ time_step] =  train_pred\n","  test_predict_plot_data[len(train_pred)+(time_step*2)+1:len(data.values)-1] = test_pred\n","\n","  plt.plot(data.values[:,0],'k', linewidth = 1.5)\n","  plt.plot(train_predict_plot_data,'mediumblue',linewidth = 1.5)\n","  plt.plot(test_predict_plot_data,'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(a)\")\n","  plt.legend(['True value', 'Predicted value in train set', 'Predicted value in test set'], loc = 'upper left')\n","\n","\n","  plt.subplot(122)\n","  plt.plot(data.values[len(train_pred)+(time_step*2)+1:-1, 0],'k',linewidth = 1.5)\n","  plt.plot(test_pred,'darkgreen',linewidth = 1.5)\n","  plt.xlabel('')\n","  plt.ylabel('Close price')\n","  plt.title(\"(b)\")\n","  plt.legend(['True value', 'Predicted value'], loc='upper left')\n","\n","  fig.savefig(output_dir_path+\"best_model_predictions_plots.png\",dpi=600)\n","\n","\n","def rmse_boxplots(model_output):\n","  fig = plt.figure(figsize = (6,5))\n","  plt.boxplot(model_output['scores']['rmse'], patch_artist=True)\n","  plt.xticks([1,2,3,4,5,6], ['10', '30', '50', '100', '150', '200'])\n","  plt.xlabel('Number of neurons')\n","  plt.ylabel('RMSE')\n","  fig.savefig(output_dir_path+\"rmse_boxplots.png\",dpi=600)\n","  plt.show()\n","\n","\n","def rmse_violinplotplots(model_output):\n","  fig = plt.figure(figsize = (6,5))\n","  plt.violinplot(model_output['scores']['rmse'])\n","  plt.xticks([1,2,3,4,5,6], ['10', '30', '50', '100', '150', '200'])\n","  plt.xlabel('Number of neurons')\n","  plt.ylabel('RMSE')\n","  fig.savefig(output_dir_path+\"rmse_violinplots.png\",dpi=600)\n","  plt.show()\n","\n","def all_scores_boxplots(model_output):\n","\n","  fig = plt.figure(figsize = (18,5))\n","  plt.subplot(131)\n","  p1 = plt.boxplot(model_output['scores']['rmse'],patch_artist=True)\n","  for i, box in enumerate(p1['boxes']):\n","    # change outline color\n","    box.set(color= 'blue', linewidth = 1.2)\n","    # change fill color\n","    box.set(facecolor = 'mediumblue')\n","  plt.xticks([1,2,3,4,5,6], ['10', '30', '50', '100', '150', '200'])\n","  plt.title(\"(a)\")\n","  plt.xlabel('Number of neurons')\n","  plt.ylabel('RMSE')\n","\n","  plt.subplot(132)\n","  p2 = plt.boxplot(model_output['scores']['mape'],patch_artist=True)\n","  for i, box in enumerate(p2['boxes']):\n","    #change outline color\n","    box.set(color= 'blue', linewidth = 1.2)\n","    # change fill color\n","    box.set(facecolor = 'indigo')\n","\n","  plt.xticks([1,2,3,4,5,6], ['10', '30', '50', '100', '150', '200'])\n","  plt.title(\"(b)\")\n","  plt.xlabel('Number of neurons')\n","  plt.ylabel('MAPE')\n","\n","  plt.subplot(133)\n","  p3 = plt.boxplot(model_output['scores']['R'],patch_artist=True)\n","  for i, box in enumerate(p3['boxes']):\n","    #change outline color\n","    box.set(color= 'blue', linewidth = 1.2)\n","    # change fill color\n","    box.set(facecolor = 'darkgreen')\n","  plt.xticks([1,2,3,4,5,6], ['10', '30', '50', '100', '150', '200'])\n","  plt.title(\"(c)\")\n","  plt.xlabel('Number of neurons')\n","  plt.ylabel('R')\n","\n","  fig.savefig(output_dir_path+\"all_scores_boxplots.png\",dpi=600)\n","  plt.show()\n","\n","\n","def create_visualization(model_output):\n","  true_pred_plot(model_output)\n","  test_scores_plot(model_output)\n","  prediction_plot(model_output)\n","  best_model_prediction_plot(model_output)\n","  rmse_boxplots(model_output)\n","  all_scores_boxplots(model_output)\n","\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"z3f0bSL1I0SD"},"outputs":[],"source":[]},{"cell_type":"markdown","metadata":{"id":"ED4G64o5Hihx"},"source":[" **Function for Building Models**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"a8PHScCoHoUn"},"outputs":[],"source":["def build_model(layers, time_step, num_features, optimizer = 'Adam', learning_rate = 0.001, verbose = 1):\n","\n","  model = Sequential()\n","\n","  for i in range(len(layers)):\n","    if len(layers)==1:\n","      model.add(LSTM(int(layers[i]), input_shape = (time_step, num_features)))\n","    else:\n","      if i < len(layers)-1:\n","        if i == 0:\n","          model.add(LSTM(int(layers[i]), input_shape=(time_step, num_features), return_sequences= True))\n","          #model.add(Dropout(0.10))\n","        else:\n","          model.add(LSTM(int(layers[i]), return_sequences=True))\n","          #model.add(Dropout(0.10))\n","      else:\n","        model.add(LSTM(int(layers[i])))\n","        #model.add(Dropout(0.10))\n","  model.add(Dense(1, activation = 'linear'))\n","\n","  if optimizer == 'Adam':\n","    opt = optimizers.Adam(learning_rate = learning_rate)\n","  elif optimizer == 'Adagrad':\n","    opt = optimizers.Adagrad(learning_rate = learning_rate)\n","  elif optimizer == 'Nadam':\n","    opt = optimizers.Nadam(learning_rate = learning_rate)\n","  elif optimizer == 'Adadelta':\n","    opt = optimizers.Adadelta(learning_rate= learning_rate)\n","  elif optimizer == 'RMSprop':\n","    opt = optimizers.RMSprop(learning_rate= learning_rate)\n","  else:\n","    print(\"No optimizer found in the list(['Adam', 'Adagrad','Nadam', 'Adadelta', 'RMSprop'])! Please apply your optimizer manually...\")\n","\n","  model.compile(loss='mean_squared_error', optimizer= opt)\n","\n","  if verbose == 1:\n","    print(model.summary())\n","  return model"]},{"cell_type":"markdown","metadata":{"id":"mbqNfEZEfGKo"},"source":["### Testing"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":239},"id":"LHuOeDc1I3Iu","outputId":"4966896a-5812-4a7a-e346-cb64cca6430c","executionInfo":{"status":"ok","timestamp":1779652747992,"user_tz":300,"elapsed":56,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}}},"outputs":[{"output_type":"display_data","data":{"text/plain":["\u001b[1mModel: \"sequential_5\"\u001b[0m\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_5\"</span>\n","</pre>\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n","┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n","┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n","│ lstm_6 (\u001b[38;5;33mLSTM\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m250\u001b[0m)            │       \u001b[38;5;34m261,000\u001b[0m │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense_5 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │           \u001b[38;5;34m251\u001b[0m │\n","└─────────────────────────────────┴────────────────────────┴───────────────┘\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n","┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n","┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n","│ lstm_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">250</span>)            │       <span style=\"color: #00af00; text-decoration-color: #00af00\">261,000</span> │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │           <span style=\"color: #00af00; text-decoration-color: #00af00\">251</span> │\n","└─────────────────────────────────┴────────────────────────┴───────────────┘\n","</pre>\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Total params: \u001b[0m\u001b[38;5;34m261,251\u001b[0m (1020.51 KB)\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">261,251</span> (1020.51 KB)\n","</pre>\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m261,251\u001b[0m (1020.51 KB)\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">261,251</span> (1020.51 KB)\n","</pre>\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n","</pre>\n"]},"metadata":{}},{"output_type":"stream","name":"stdout","text":["None\n"]},{"output_type":"execute_result","data":{"text/plain":["<Sequential name=sequential_5, built=True>"]},"metadata":{},"execution_count":74}],"source":["optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","\n","build_model([250], 5, 10, optimizers_names[2], 0.001, 1)"]},{"cell_type":"markdown","metadata":{"id":"bdv4kf0CIcdR"},"source":["## **Function for Hyperparameter Tuning**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"R_uHVktrIj6K"},"outputs":[],"source":["def hyper_parameter_tuning(layers, data, time_step, split, optimizers_names, learning_rates, batch_sizes, epochs, num_replicates = 2):\n","  #======= creating training and test data===#\n","  train_data, val_data = data_split(data, split)\n","\n","  num_features = train_data.shape[1]\n","\n","  min_train, max_train  = train_data[\"Close\"].min(), train_data[\"Close\"].max()\n","  min_val, max_val   =    val_data[\"Close\"].min(), val_data[\"Close\"].max()\n","\n","  train_data_scaled  =  min_max_transform(train_data)\n","  val_data_scaled    = min_max_transform(val_data)\n","\n","  X_train, y_train =   DatasetCreation(train_data_scaled, time_step)\n","  X_val, y_val     =   DatasetCreation(val_data_scaled, time_step)\n","\n","  #========dealing with time series=========#\n","\n","  best_avg_rmse = 99999999999\n","\n","  collect_rmse = []\n","\n","  all_avg_rmse = np.zeros((len(optimizers_names), len(learning_rates), len(batch_sizes)))\n","\n","  best_hyper_parameters = {\"model\": layers, \"optimizer\": None, \"learning_rate\": None, \"batch_size\": None,\"best_avg_rmse\": None}\n","\n","  for opt in range(len(optimizers_names)):\n","\n","    for lr in range(len(learning_rates)):\n","\n","      for batch_size in range(len(batch_sizes)):\n","\n","        for i in range(num_replicates):\n","\n","          print(\"Running for \" + optimizers_names[opt] + \" optimizer \" + str(learning_rates[lr]) +  \" learning_rate \" +  str(batch_sizes[batch_size]) + \" batch_size and \" + str(i) +  \" replicate \" +  \"\\n\")\n","\n","          model = build_model(layers, time_step, num_features, optimizers_names[opt], learning_rate = learning_rates[lr], verbose = 0)\n","\n","          callback = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience= 5)\n","\n","          history = model.fit(X_train, y_train, batch_size = batch_sizes[batch_size], epochs= epochs, validation_data = (X_val, y_val), callbacks=[callback], verbose = 1)\n","\n","\n","\n","          #==============Making predictions in original scale ==========\n","\n","          val_pred    =  min_max_inverse_transform(model.predict(X_val).ravel(), min_val, max_val)\n","\n","          collect_rmse.append(math.sqrt(mean_squared_error(min_max_inverse_transform(y_val, min_val, max_val),val_pred)))\n","\n","        avg_rmse = np.mean(np.array(collect_rmse))\n","        all_avg_rmse[opt][lr][batch_size] = avg_rmse\n","\n","        if avg_rmse < best_avg_rmse:\n","          best_avg_rmse = avg_rmse\n","          best_hyper_parameters = {\"model\": layers,  \"optimizer\": optimizers_names[opt], \"learning_rate\": learning_rates[lr], \"batch_size\": batch_sizes[batch_size], \"best_avg_rmse\": best_avg_rmse}\n","\n","\n","  output_dictionary = {\n","      \"best_hyper_parameters\":  best_hyper_parameters,\n","      \"all_avg_rmse\": all_avg_rmse\n","       }\n","\n","  #writing output dictionary in the file\n","\n","  file_name = output_dir_path+ \"sl-lstm-\" + str(layers[0])+ \"-neurons-validation_results\"+ str(time.time())+ \".txt\"\n","  write_dic_to_file(output_dictionary, file_name)\n","\n","  print(\"Best_hyper_parameters: \\n\", output_dictionary['best_hyper_parameters'])\n","  print(\"all_avg_rmse: \\n\", output_dictionary['all_avg_rmse'])\n","\n","  return output_dictionary['best_hyper_parameters']\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"htx8uH9O2epV"},"outputs":[],"source":[]},{"cell_type":"markdown","metadata":{"id":"aQYNu38-1Xo4"},"source":["### **Case I: Tuning parameter of 10N-singlelayer-LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":70},"id":"wsNMrDUN1YaR","outputId":"5a68743c-4107-4cd3-fd1b-6892e5645829","executionInfo":{"status":"ok","timestamp":1779652748061,"user_tz":300,"elapsed":19,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\nlayers = [10]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\n#optimizers_names = ['Adam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs = 50\\nnum_replicates = 10\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN10_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN10_best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":76}],"source":["\"\"\"\n","layers = [10]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","#optimizers_names = ['Adam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs = 50\n","num_replicates = 10\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N10_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N10_best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"fArhgB-XNd1U"},"source":["### **Case II: Tuning parameter of 30N-singlelayer-LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":70},"id":"p6tMjJzrNb0O","outputId":"cafc0f95-1782-4dfb-a334-7d0b136bb95b","executionInfo":{"status":"ok","timestamp":1779652748066,"user_tz":300,"elapsed":18,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\nlayers = [30]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  50\\nnum_replicates = 10\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN30_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN30_best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":77}],"source":["\"\"\"\n","layers = [30]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  50\n","num_replicates = 10\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N30_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N30_best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"bAJ26vPr9JkW"},"source":["### **Case III: Tuning parameter of 50N-singlelayer-LSTM**\n","\n","---\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":70},"id":"wqPnDy0bIvc9","outputId":"71aef7a3-e52b-4283-fbf0-c51bba879129","executionInfo":{"status":"ok","timestamp":1779652748068,"user_tz":300,"elapsed":14,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\nlayers = [50]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs = 50\\nnum_replicates = 10\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN50_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN50_best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":78}],"source":["\"\"\"\n","layers = [50]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs = 50\n","num_replicates = 10\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N50_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N50_best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"DPt8YYZ4Oc6-"},"source":["### **Case IV: Tuning parameter of 100N-singlelayer-LSTM**\n","\n","---"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":70},"id":"W_gm5CWzOYh5","outputId":"54883b9b-8a27-4b3f-d639-3fe76005f16e","executionInfo":{"status":"ok","timestamp":1779652748176,"user_tz":300,"elapsed":112,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\nlayers = [100]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  50\\nnum_replicates = 10\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN100_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN100_best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":79}],"source":["\"\"\"\n","layers = [100]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  50\n","num_replicates = 10\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N100_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N100_best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"A99NZ7EeOjlP"},"source":["### **Case V: Tuning parameter of 150N-singlelayer-LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":70},"id":"6YaGcabQOqBs","outputId":"1e1335cb-861b-43ce-8546-a647e5f53c87","executionInfo":{"status":"ok","timestamp":1779652748179,"user_tz":300,"elapsed":15,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\nlayers = [150]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  50\\nnum_replicates = 10\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN150_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN150_best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":80}],"source":["\"\"\"\n","layers = [150]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  50\n","num_replicates = 10\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N150_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N150_best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"UQpYD_UFOvJu"},"source":["### **Case VI: Tuning parameter of 200N-singlelayer-LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":70},"id":"KYPP2AMRO0Fj","outputId":"3c5c00e8-7f19-46e4-acac-9beacd28c054","executionInfo":{"status":"ok","timestamp":1779652748180,"user_tz":300,"elapsed":10,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\nlayers = [200]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  50\\nnum_replicates = 10\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN200_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN200_best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":81}],"source":["\"\"\"\n","layers = [200]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  50\n","num_replicates = 10\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N200_best_hyper_parameters = hyper_parameter_tuning(layers, train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N200_best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"aA13pmJ6keN0"},"source":["## **Building and Running  Single-Layer Models in Full Scale**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"YTI2jGWgRQg8"},"outputs":[],"source":["#========== Model hyper parameters settting ==========================#\n","def LSTM_model(neurons, hyper_parameters, data, time_step = 5, test_split = 0.2, epochs = 20,  num_replicates = 2):\n","\n","  #====== data transformation==========#\n","  print(\"Progress: Performing data preparation steps.......\\n\")\n","\n","  # Fill NaN values before splitting and scaling\n","  data = data.ffill().bfill()\n","\n","  #======= creating training and test data===#\n","\n","  train_data, test_data = data_split(data, test_split)\n","\n","  num_features = train_data.shape[1]\n","\n","  min_train, max_train  = train_data[\"Close\"].min(), train_data[\"Close\"].max()\n","  min_test, max_test   =  test_data[\"Close\"].min(), test_data[\"Close\"].max()\n","\n","  train_data_scaled  =  min_max_transform(train_data)\n","  test_data_scaled   = min_max_transform(test_data)\n","\n","\n","  X_train, y_train  =   DatasetCreation(train_data_scaled, time_step)\n","  X_test, y_test    =   DatasetCreation(test_data_scaled, time_step)\n","\n","  y_train_original  =  min_max_inverse_transform(y_train, min_train, max_train) #in original scale\n","  y_test_original  =  min_max_inverse_transform(y_test, min_test, max_test) #in original scale\n","\n","\n","  print(\"Progress: Building and training models.......\\n\")\n","\n","  neurons = np.array(neurons)\n","  #============ arrays for collecting test scores ================#\n","  rmse_array = np.zeros((len(neurons), num_replicates))\n","  #mae_array  = np.zeros((len(neurons), num_replicates))\n","  mape_array = np.zeros((len(neurons), num_replicates))\n","  #R2_array   = np.zeros((len(neurons), num_replicates))\n","  R_array    = np.zeros((len(neurons), num_replicates))\n","  elapsed_time_array = np.zeros((len(neurons), num_replicates))\n","\n","\n","  #========== array for collecting history and predictions =======#\n","  models_history = []\n","  train_predictions = []\n","  test_predictions  = []\n","\n","  for i in range(len(neurons)):\n","\n","    print(\"Model hyperparameters used:\\n \", hyper_parameters[i])\n","    #========== saving history and predictions per replicate=====#\n","    model_history_per_replicate = []\n","    train_predictions_per_replicate = []\n","    test_predictions_per_replicate  = []\n","\n","    hidden_nodes = int(neurons[i])\n","\n","   # print(\"Program is running for %d neurons ----->\\n\" %np.int(neurons[i]))\n","\n","\n","    for k in range(num_replicates):\n","\n","      print(\"Program is running for %d neurons and %d replicate ----->\\n\" %(hidden_nodes, k))\n","\n","\n","\n","      layers = [hidden_nodes]\n","\n","\n","      model = build_model(layers, time_step, num_features, optimizer = hyper_parameters[i][0], learning_rate = hyper_parameters[i][1], verbose = 0)\n","      callback = tf.keras.callbacks.EarlyStopping(monitor='loss', patience= 5)\n","      # This callback will stop the training when there is no improvement in\n","      # the loss for three consecutive epochs\n","      start = time.time()\n","      history = model.fit(X_train, y_train, batch_size = hyper_parameters[i][2], epochs= epochs, callbacks=[callback], verbose = 1)\n","      end = time.time()\n","      elapsed_time = end - start\n","\n","      model_history_per_replicate.append(history)\n","\n","\n","      #==============Making train and test prediction in original scales ==========\n","      train_pred   =  min_max_inverse_transform(model.predict(X_train).ravel(), min_train, max_train) #in original scale\n","      test_pred    =  min_max_inverse_transform(model.predict(X_test).ravel(), min_test, max_test)\n","\n","      train_predictions_per_replicate.append(train_pred)\n","      test_predictions_per_replicate.append(test_pred)\n","\n","      #============== Calculating performance scores=========\n","      scores =   calculate_scores(min_max_inverse_transform(y_test, min_test, max_test),test_pred)\n","      rmse_array[i][k] = scores['rmse']\n","      mape_array[i][k] =  scores['mape']\n","      R_array[i][k] = scores['R']\n","      elapsed_time_array[i][k] = elapsed_time\n","\n","    models_history.append(model_history_per_replicate)\n","    train_predictions.append(train_predictions_per_replicate)\n","    test_predictions.append(test_predictions_per_replicate)\n","\n","  print(\"Progress: Collecting outputs.......\\n\")\n","\n","  neurons_df = pd.DataFrame(neurons)\n","  rmse_df = pd.DataFrame(rmse_array)\n","  #mae_df  = pd.DataFrame(mae_array)\n","  mape_df  = pd.DataFrame(mape_array)\n","  #R2_df   = pd.DataFrame(R2_array)\n","  R_df    = pd.DataFrame(R_array)\n","  elapsed_time_df =  pd.DataFrame(elapsed_time_array)\n","\n","  train_predictions  = np.array(train_predictions)\n","  test_predictions   = np.array(test_predictions)\n","\n","  #==== Idendifying  the best model results based on rmse ===============#\n","  min_index = pd.DataFrame(rmse_df.min(axis = 1)).idxmin()[0]\n","  min_col =   pd.DataFrame(rmse_df.min(axis = 0)).idxmin()[0]\n","\n","  num_neurons_with_best_rmse = neurons_df.loc[min_index,0]\n","\n","  best_rmse = rmse_df.loc[min_index, min_col]\n","  #mae_with_best_rmse = mae_df.loc[min_index, min_col]\n","  mape_with_best_rmse = mape_df.loc[min_index, min_col]\n","  #R2_with_best_rmse = R2_df.loc[min_index, min_col]\n","  R_with_best_rmse =  R_df.loc[min_index, min_col]\n","  elapsed_time_with_best_rmse = elapsed_time_df.loc[min_index, min_col]\n","\n","  train_predictions_with_best_rmse = train_predictions[min_index][min_col]\n","  test_predictions_with_best_rmse = test_predictions[min_index][min_col]\n","\n","  loss_with_best_rmse = models_history[min_index][min_col].history['loss']\n","  #val_loss_with_best_rmse = models_history[min_index][min_col].history['val_loss']\n","\n","  #======= Collecting hyperparameters=============#\n","  hyper_parameters = { 'neurons': neurons,\n","                       'model_specific_hyper_parameters': hyper_parameters,#additional best_hyper_parmeters for each models\n","                       'epochs': epochs,\n","                       'time_step':time_step,\n","                       'num_replicates': num_replicates,\n","                       'test_split':test_split\n","                       #'validataion_split':validation_split\n","                        }\n","\n","  #======= Collecting test scores =============#\n","  scores = {'neurons': neurons_df, 'rmse': rmse_df, 'mape': mape_df, 'R': R_df, 'elapsed_time': elapsed_time_df}\n","\n","  #======= Collecting average test scores =============#\n","  avg_scores = pd.DataFrame({'neurons': neurons,\n","                            'rmse': rmse_df.mean(axis = 1),\n","                            'mape': mape_df.mean(axis = 1),\n","                            'R': R_df.mean(axis = 1),\n","                            'elapsed_time': elapsed_time_df.mean(axis = 1)})\n","\n"," #======= Collecting average test scores =============#\n","  all_stds = pd.DataFrame({'neurons': neurons,\n","                            'rmse': rmse_df.std(axis = 1),\n","                            'mape': mape_df.std(axis = 1),\n","                            'R': R_df.std(axis = 1),\n","                            'elapsed_time': elapsed_time_df.std(axis = 1)})\n","\n","\n"," #======= Collecting average test scores =============#\n","  all_minimums = pd.DataFrame({'neurons': neurons,\n","                            'rmse': rmse_df.min(axis = 1),\n","                            'mape': mape_df.min(axis = 1),\n","                            'R': R_df.min(axis = 1),\n","                            'elapsed_time': elapsed_time_df.min(axis = 1)})\n","\n","  #======= Collecting average test scores =============#\n","  all_maximums = pd.DataFrame({'neurons': neurons,\n","                            'rmse': rmse_df.max(axis = 1),\n","                            'mape': mape_df.max(axis = 1),\n","                            'R': R_df.max(axis = 1),\n","                            'elapsed_time': elapsed_time_df.max(axis = 1)})\n","\n","\n","\n","  #======= Collecting the best model results =============#\n","  model_with_best_rmse = {  'neurons': num_neurons_with_best_rmse,\n","                            'replicate': min_col,\n","                            'rmse': best_rmse,\n","                            'mape': mape_with_best_rmse,\n","                            'R':  R_with_best_rmse,\n","                            'elapsed_time': elapsed_time_with_best_rmse,\n","                            'train_predictions':train_predictions_with_best_rmse,\n","                            'test_predictions': test_predictions_with_best_rmse,\n","                            'loss':loss_with_best_rmse,\n","\n","                         }\n","\n","  datasets  =    {'data': data,\n","                  'X_train': X_train,\n","                  'X_test': X_test,\n","                  'y_train': y_train_original,\n","                  'y_test': y_test_original\n","                  }\n","\n","  #======= Collecting all the outputs together =============#\n","  output_dictionary = { 'hyper_parameters': hyper_parameters,\n","                        'best_model': model_with_best_rmse,\n","                        'scores': scores,\n","                        'avg_scores': avg_scores,\n","                        'all_stds': all_stds,\n","                        'all_minimums': all_minimums,\n","                        'all_maximums': all_maximums,\n","                        'train_predictions': train_predictions,\n","                        'test_predictions':  test_predictions,\n","                        'models_history': models_history,\n","                        'datasets': datasets\n","                       }\n","\n","  print(\"\\nBest model (neurons, replicate, rmse): \", num_neurons_with_best_rmse, min_col, best_rmse)\n","  print('\\nAverage scores:\\n', avg_scores)\n","  print('\\nStandard_deviations:\\n', all_stds)\n","  print('\\nMinimums:\\n', all_minimums)\n","  print('\\nMaximums:\\n', all_maximums)\n","  print(\"\\nProgress: All works are done successfully, congratulations!!\\n\")\n","\n","\n","\n","  #Save all rmses in a file for statistical study\n","  scores['rmse'].to_csv(output_dir_path+'sl-lstm-all-rmse.csv')\n","\n","  #writing output dictionary in the file\n","  file_name = output_dir_path + \"sl-lstm-results.txt\"\n","  write_dic_to_file(output_dictionary, file_name)\n","\n","  return output_dictionary"]},{"cell_type":"markdown","metadata":{"id":"esscOyB2nCG8"},"source":["# **Building Multi-Layer LSTM Models**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"hF61ZHIRdvyC"},"outputs":[],"source":["def hyper_parameter_tuning_multilayer(layers, data, time_step, split, optimizers_names, learning_rates, batch_sizes, epochs, num_replicates = 2):\n","  #======= creating training and test data===#\n","  train_data, val_data = data_split(data, split)\n","\n","  num_features = train_data.shape[1]\n","\n","  min_train, max_train  = train_data[\"Close\"].min(), train_data[\"Close\"].max()\n","  min_val, max_val   =    val_data[\"Close\"].min(), val_data[\"Close\"].max()\n","\n","  train_data_scaled  =  min_max_transform(train_data)\n","  val_data_scaled    = min_max_transform(val_data)\n","\n","  X_train, y_train =   DatasetCreation(train_data_scaled, time_step)\n","  X_val, y_val     =   DatasetCreation(val_data_scaled, time_step)\n","\n","  #========dealing with time series=========#\n","\n","  best_avg_rmse = 99999999999\n","\n","  collect_rmse = []\n","\n","  all_avg_rmse = np.zeros((len(optimizers_names), len(learning_rates), len(batch_sizes)))\n","\n","  best_hyper_parameters = {\"model\": layers, \"optimizer\": None, \"learning_rate\": None, \"batch_size\": None,\"best_avg_rmse\": None}\n","\n","\n","  for opt in range(len(optimizers_names)):\n","\n","    for lr in range(len(learning_rates)):\n","\n","      for batch_size in range(len(batch_sizes)):\n","\n","        for i in range(num_replicates):\n","\n","          print(\"Running for \" + optimizers_names[opt] + \" optimizer \" + str(learning_rates[lr]) +  \" learning_rate \" +  str(batch_sizes[batch_size]) + \" batch_size and \" + str(i) +  \" replicate \" +  \"\\n\")\n","\n","          model = build_model(layers, time_step, num_features, optimizers_names[opt], learning_rate = learning_rates[lr], verbose = 0)\n","\n","          callback = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience= 5)\n","\n","          history = model.fit(X_train, y_train, batch_size = batch_sizes[batch_size], epochs= epochs, validation_data = (X_val, y_val), callbacks=[callback], verbose = 1)\n","\n","\n","          #==============Making predictions in original scale ==========\n","          #train_pred  =  min_max_inverse_transform(model.predict(X_train).ravel(), min_train, max_train)\n","          val_pred    =  min_max_inverse_transform(model.predict(X_val).ravel(), min_val, max_val)\n","\n","          #train_scores =  calculate_scores(min_max_inverse_transform(y_train, min_train, max_train),train_pred)\n","          #scores =   calculate_scores(min_max_inverse_transform(y_val, min_val, max_val),val_pred)\n","\n","          collect_rmse.append(math.sqrt(mean_squared_error(min_max_inverse_transform(y_val, min_val, max_val),val_pred)))\n","\n","        avg_rmse = np.mean(np.array(collect_rmse))\n","        all_avg_rmse[opt][lr][batch_size] = avg_rmse\n","\n","        if avg_rmse < best_avg_rmse:\n","          best_avg_rmse = avg_rmse\n","          best_hyper_parameters = {\"model\": layers,  \"optimizer\": optimizers_names[opt], \"learning_rate\": learning_rates[lr], \"batch_size\": batch_sizes[batch_size], \"best_avg_rmse\": best_avg_rmse}\n","\n","\n","  output_dictionary = {\n","      \"best_hyper_parameters\":  best_hyper_parameters,\n","      \"all_avg_rmse\": all_avg_rmse\n","       }\n","\n","  #writing output dictionary in the file\n","\n","  file_name = output_dir_path+ \"ml-lstm-\" + str(layers)+ \"-neurons-validation_results.txt\"\n","  write_dic_to_file(output_dictionary, file_name)\n","\n","  print(\"Best_hyper_parameters: \\n\", output_dictionary['best_hyper_parameters'])\n","  print(\"all_avg_rmse: \\n\", output_dictionary['all_avg_rmse'])\n","\n","  return output_dictionary['best_hyper_parameters']\n"]},{"cell_type":"markdown","metadata":{"id":"KhO55nHFgPU6"},"source":["### Testing"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"723GijDtjUcd","colab":{"base_uri":"https://localhost:8080/","height":273},"executionInfo":{"status":"ok","timestamp":1779652748251,"user_tz":300,"elapsed":12,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}},"outputId":"9af52e75-7ea1-44e0-80d1-7df78a27686a"},"outputs":[{"output_type":"display_data","data":{"text/plain":["\u001b[1mModel: \"sequential_6\"\u001b[0m\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_6\"</span>\n","</pre>\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n","┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n","┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n","│ lstm_7 (\u001b[38;5;33mLSTM\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m10\u001b[0m)          │           \u001b[38;5;34m840\u001b[0m │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ lstm_8 (\u001b[38;5;33mLSTM\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m)              │           \u001b[38;5;34m320\u001b[0m │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense_6 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │             \u001b[38;5;34m6\u001b[0m │\n","└─────────────────────────────────┴────────────────────────┴───────────────┘\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n","┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n","┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n","│ lstm_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)          │           <span style=\"color: #00af00; text-decoration-color: #00af00\">840</span> │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ lstm_8 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>)              │           <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span> │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │             <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span> │\n","└─────────────────────────────────┴────────────────────────┴───────────────┘\n","</pre>\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Total params: \u001b[0m\u001b[38;5;34m1,166\u001b[0m (4.55 KB)\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,166</span> (4.55 KB)\n","</pre>\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m1,166\u001b[0m (4.55 KB)\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,166</span> (4.55 KB)\n","</pre>\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"],"text/html":["<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n","</pre>\n"]},"metadata":{}},{"output_type":"stream","name":"stdout","text":["None\n"]},{"output_type":"execute_result","data":{"text/plain":["<Sequential name=sequential_6, built=True>"]},"metadata":{},"execution_count":84}],"source":["layers = [[10, 5]]\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","build_model(layers[0], 5, 10, optimizers_names[2], 0.001, 1)"]},{"cell_type":"markdown","metadata":{"id":"DA34iwsmiKEl"},"source":["## **Case I: Tuning Hyperparameters of 10-5N LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"eX3YK-hOeFDZ","colab":{"base_uri":"https://localhost:8080/","height":70},"executionInfo":{"status":"ok","timestamp":1779652748255,"user_tz":300,"elapsed":6,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}},"outputId":"8791e18c-b84e-4347-bc2d-760757a8a278"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\n#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\\nlayers = [[10, 5]]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  2\\nnum_replicates = 2\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN10_5__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN10_5__best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":85}],"source":["\"\"\"\n","#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\n","layers = [[10, 5]]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  2\n","num_replicates = 2\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N10_5__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N10_5__best_hyper_parameters\n","\"\"\"\n"]},{"cell_type":"markdown","metadata":{"id":"9PUeeIF7nQq1"},"source":["## **Case II: Tuning Hyperparameters of 20-10N LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"CE5fU4HVnKH9","colab":{"base_uri":"https://localhost:8080/","height":70},"executionInfo":{"status":"ok","timestamp":1779652748261,"user_tz":300,"elapsed":5,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}},"outputId":"27ecbb3a-8efb-4ac2-dd78-65f2f6cb1649"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\n#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\\nlayers = [[20, 10]]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  2\\nnum_replicates = 2\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN20_10__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN20_10__best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":86}],"source":["\"\"\"\n","#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\n","layers = [[20, 10]]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  2\n","num_replicates = 2\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N20_10__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N20_10__best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"7rkakLo_nmOK"},"source":["## **Case III: Tuning Hyperparameters of 50-20N LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"KS3h5fzSnlJD","colab":{"base_uri":"https://localhost:8080/","height":70},"executionInfo":{"status":"ok","timestamp":1779652748270,"user_tz":300,"elapsed":6,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}},"outputId":"03cc977f-eef5-4dd1-b7e5-8ca881f8c9d7"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\n#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\\nlayers = [[50, 20]]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  2\\nnum_replicates = 2\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN50_20__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN50_20__best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":87}],"source":["\"\"\"\n","#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\n","layers = [[50, 20]]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  2\n","num_replicates = 2\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N50_20__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N50_20__best_hyper_parameters\n","\"\"\"\n"]},{"cell_type":"markdown","metadata":{"id":"ueI4qZkrn6DR"},"source":["## **Case IV: Tuning Hyperparameters of 100-50N LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"P8OQluGrn-BR","colab":{"base_uri":"https://localhost:8080/","height":70},"executionInfo":{"status":"ok","timestamp":1779652748288,"user_tz":300,"elapsed":15,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}},"outputId":"50f951bd-ae8d-497e-dd06-aaf3b1e48642"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\n#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\\nlayers = [[100, 50]]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  2\\nnum_replicates = 2\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN100_50__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN100_50__best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":88}],"source":["\"\"\"\n","#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\n","layers = [[100, 50]]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  2\n","num_replicates = 2\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N100_50__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N100_50__best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"JEAwGO5zoZKN"},"source":["## **Case V: Tuning Hyperparameters of 150-100N LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"dPD3Z1Y9oYxc","colab":{"base_uri":"https://localhost:8080/","height":70},"executionInfo":{"status":"ok","timestamp":1779652748365,"user_tz":300,"elapsed":67,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}},"outputId":"267c4a34-0c95-4903-977e-b2b4a2a12e87"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\n#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\\nlayers = [[150, 100]]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  2\\nnum_replicates = 2\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN150_100__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN150_100__best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":89}],"source":["\"\"\"\n","#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\n","layers = [[150, 100]]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  2\n","num_replicates = 2\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N150_100__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N150_100__best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"q8klpfxhot5y"},"source":["## **Case VI: Tuning Hyperparameters of 100-50-20N LSTM**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"jClFlbJUouiV","colab":{"base_uri":"https://localhost:8080/","height":70},"executionInfo":{"status":"ok","timestamp":1779652748373,"user_tz":300,"elapsed":6,"user":{"displayName":"Abdul-Razak Bawa Yussif","userId":"03524206631392465860"}},"outputId":"1070d97b-eeef-4133-e3b5-ebbcd42b9815"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\n#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\\nlayers = [[100, 50, 20]]\\ntime_step = 5\\noptimizers_names = ['Adam', 'Adagrad', 'Nadam']\\nlearning_rates =  [0.1, 0.01, 0.001]\\nbatch_sizes =  [4, 8, 16]\\nepochs =  2\\nnum_replicates = 2\\ntest_split = 0.2\\nval_split = 0.2\\n\\ntrain_data, test_data = data_split(data, test_split)\\n\\nN100_50_20__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\\nN100_50_20__best_hyper_parameters\\n\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":90}],"source":["\"\"\"\n","#hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\n","layers = [[100, 50, 20]]\n","time_step = 5\n","optimizers_names = ['Adam', 'Adagrad', 'Nadam']\n","learning_rates =  [0.1, 0.01, 0.001]\n","batch_sizes =  [4, 8, 16]\n","epochs =  2\n","num_replicates = 2\n","test_split = 0.2\n","val_split = 0.2\n","\n","train_data, test_data = data_split(data, test_split)\n","\n","N100_50_20__best_hyper_parameters = hyper_parameter_tuning_multilayer(layers[0], train_data, time_step, val_split, optimizers_names, learning_rates, batch_sizes, epochs = epochs, num_replicates = num_replicates)\n","N100_50_20__best_hyper_parameters\n","\"\"\""]},{"cell_type":"markdown","metadata":{"id":"b40CrlyMcrlv"},"source":["## **Building and Running Multi-Layers LSTM Models in Full Scale**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"RbAtH3KenO9D"},"outputs":[],"source":["def multi_layer_LSTM_Model(layers, hyper_parameters, data, time_step = 5, test_split = 0.2, epochs = 5,  num_replicates = 2):\n","    #====== data transformation==========#\n","    print(\"Progress: Performing data preparation steps.......\\n\")\n","\n","    #======= creating training and test data===#\n","\n","    train_data, test_data = data_split(data, test_split)\n","\n","    num_features = train_data.shape[1]\n","\n","    min_train, max_train  = train_data[\"Close\"].min(), train_data[\"Close\"].max()\n","    min_test, max_test   =  test_data[\"Close\"].min(), test_data[\"Close\"].max()\n","\n","    train_data_scaled  =  min_max_transform(train_data)\n","    test_data_scaled   = min_max_transform(test_data)\n","\n","\n","    X_train, y_train  =   DatasetCreation(train_data_scaled, time_step)\n","    X_test, y_test    =   DatasetCreation(test_data_scaled, time_step)\n","\n","    y_train_original  =  min_max_inverse_transform(y_train, min_train, max_train) #in original scale\n","    y_test_original  =  min_max_inverse_transform(y_test, min_test, max_test) #in original scale\n","\n","\n","    #============ arrays for collecting test scores ================#\n","    rmse_array = np.zeros(num_replicates)\n","    mape_array = np.zeros(num_replicates)\n","    R_array    = np.zeros(num_replicates)\n","    elapsed_time_array = np.zeros(num_replicates)\n","\n","    models_history = []\n","    train_predictions = []\n","    test_predictions = []\n","\n","    for i in range(num_replicates):\n","\n","      print(\"Program is running for %d replicate ----->\\n\" %i)\n","\n","      model = build_model(layers, time_step, num_features, optimizer = hyper_parameters[0], learning_rate = hyper_parameters[1], verbose = 0)\n","      callback = tf.keras.callbacks.EarlyStopping(monitor='loss', patience= 5)\n","      # This callback will stop the training when there is no improvement in\n","      # the loss for three consecutive epochs\n","      start = time.time()\n","      history = model.fit(X_train, y_train, batch_size = hyper_parameters[2], epochs= epochs, callbacks=[callback], verbose = 1)\n","      end = time.time()\n","      elapsed_time = end - start\n","\n","      models_history.append(history)\n","\n","\n","      #==============Making train and test prediction in original scales ==========\n","      train_pred   =  min_max_inverse_transform(model.predict(X_train).ravel(), min_train, max_train) #in original scale\n","      test_pred    =  min_max_inverse_transform(model.predict(X_test).ravel(), min_test, max_test)\n","\n","      train_predictions.append(train_pred)\n","      test_predictions.append(test_pred)\n","\n","      #============== Calculating performance scores==========\n","      scores =   calculate_scores(min_max_inverse_transform(y_test, min_test, max_test),test_pred)\n","\n","      rmse_array[i] =  scores['rmse']\n","      mape_array[i] =  scores['mape']\n","      R_array[i] = scores['R']\n","      elapsed_time_array[i] = elapsed_time\n","\n","    min_index = rmse_array.argmin()\n","    best_rmse = rmse_array[min_index]\n","    mape_with_best_rmse = mape_array[min_index]\n","    R_with_best_rmse =  R_array[min_index]\n","    elapsed_time_with_best_rmse = elapsed_time_array[min_index]\n","\n","    train_predictions_with_best_rmse = train_predictions[min_index]\n","    test_predictions_with_best_rmse = test_predictions[min_index]\n","\n","    loss_with_best_rmse = models_history[min_index].history['loss']\n","\n","\n","    #======= Collecting average test scores =============#\n","    all_scores = {'rmse': rmse_array, 'mape': mape_array, 'R': R_array, 'elapsed_time': elapsed_time_array}\n","\n","    avg_scores = {'rmse': np.mean(rmse_array),\n","                  'mape': np.mean(mape_array),\n","                  'R': np.mean(R_array),\n","                  'elapsed_time': np.mean(elapsed_time_array)}\n","\n","    #======= Collecting standard deviations of scores =============#\n","    stds = {'rmse': np.std(rmse_array),\n","              'mape': np.std(mape_array),\n","                  'R': np.std(R_array),\n","                  'elapsed_time': np.std(elapsed_time_array)}\n","\n","    #======= Collecting minimum values of test scores =============#\n","    minimums = {'rmse': np.min(rmse_array),\n","                'mape': np.min(mape_array),\n","                'R': np.min(R_array),\n","                'elapsed_time': np.min(elapsed_time_array)}\n","\n","\n","\n","    #======= Collecting maximum values of test scores =============#\n","    maximums = {'rmse': np.max(rmse_array),\n","                'mape': np.max(mape_array),\n","                'R': np.max(R_array),\n","                'elapsed_time': np.max(elapsed_time_array)}\n","\n","\n","    model_with_best_rmse = {\n","                            'replicate': min_index,\n","                            'rmse': best_rmse,\n","                            'mape': mape_with_best_rmse,\n","                            'R':  R_with_best_rmse,\n","                            'elapsed_time': elapsed_time_with_best_rmse,\n","                            'train_predictions':train_predictions_with_best_rmse,\n","                            'test_predictions': test_predictions_with_best_rmse,\n","                            'loss':loss_with_best_rmse,\n","                         }\n","     #======= Collecting hyperparameters=============#\n","    hyper_parameters = {'layers': layers,\n","                        'model_specific_hyper_parameters': hyper_parameters,#additional best_hyper_parmeters for each models\n","                       'epochs': epochs,\n","                       'time_step':time_step,\n","                       'num_replicates': num_replicates,\n","                       'test_split':test_split\n","                        }\n","\n","\n","\n","    datasets  =   {'data': data,\n","                  'X_train': X_train,\n","                  'X_test': X_test,\n","                  'y_train': y_train_original,\n","                  'y_test': y_test_original\n","                  }\n","\n","    #======= Collecting all the outputs together =============#\n","    output_dictionary = {'hyper_parameters': hyper_parameters,\n","                        'best_model': model_with_best_rmse,\n","                        'all_scores': all_scores,\n","                        'avg_scores': avg_scores,\n","                        'standard deviations': stds,\n","                        'minimums': minimums,\n","                        'maximums': maximums,\n","                        'train_predictions': train_predictions,\n","                        'test_predictions':  test_predictions,\n","                        'datasets': datasets\n","                       }\n","\n","    return output_dictionary\n"]},{"cell_type":"markdown","metadata":{"id":"sb2MaY_kynwh"},"source":["## **Training Multiple Models Together**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"1GugZupYxeSc"},"outputs":[],"source":["def run_multi_layer_LSTM_Model(hidden_layers, hyper_parameters, data, time_step = 5, test_split = 0.2, epochs = 5,  num_replicates = 2):\n","\n","  num_models = len(hidden_layers)\n","\n","  #== to collect all scores===#\n","  rmse = []\n","  mape = []\n","  R = []\n","  elapsed_time = []\n","\n","  #===to collect all avg scores===#\n","  avg_rmse = []\n","  avg_mape = []\n","  avg_R = []\n","  avg_elapsed_time = []\n","\n","  #=== to iteratively update the best rmse and the corresponding model\n","  best_avg_rmse = 99999999999\n","  best_rmse = 99999999999\n","  best_model_hidden_layers = None\n","  best_model_output = None\n","\n","  for i in range(num_models):\n","    print(\"Running model with hidden neurons: \", hidden_layers[i])\n","\n","    print(\"\\n\")\n","\n","    print(\"Best Hyper_parameters used: \", hyper_parameters[i])\n","\n","    print(\"\\n\")\n","\n","    output = multi_layer_LSTM_Model(hidden_layers[i], hyper_parameters[i], data, time_step, test_split, epochs, num_replicates)\n","\n","    rmse.append(output['all_scores']['rmse'])\n","    mape.append(output['all_scores']['mape'])\n","    R.append(output['all_scores']['R'])\n","    elapsed_time.append(output['all_scores']['elapsed_time'])\n","\n","    avg_rmse.append(output['avg_scores']['rmse'])\n","    avg_mape.append(output['avg_scores']['mape'])\n","    avg_R.append(output['avg_scores']['R'])\n","    avg_elapsed_time.append(output['avg_scores']['elapsed_time'])\n","\n","    if avg_rmse[i] < best_avg_rmse:\n","      best_avg_rmse = avg_rmse[i]\n","      best_rmse = output['best_model']['rmse']\n","      best_model_hidden_layers = hidden_layers[i]\n","      best_model_output = output\n","\n","\n","  rmse = np.array(rmse)\n","  mape = np.array(mape)\n","  R =  np.array(R)\n","\n","  # ===== Collecting all  scores================================#\n","  scores = {'layers': hidden_layers, 'rmse': rmse, 'mape': mape, 'R':R, 'elapsed_time': elapsed_time}\n","\n","  # ======= Collecting avg scores ===============================#\n","  avg_scores = pd.DataFrame({'layers': hidden_layers, 'rmse': np.array(avg_rmse), 'mape': np.array(avg_mape), 'R':np.array(avg_R), 'elapsed_time':np.array(avg_elapsed_time)})\n","\n","  #======= Collecting standard deviations of scores =============#\n","  stds = pd.DataFrame({'layers': hidden_layers, 'rmse': np.std(rmse, axis = 1), 'mape': np.std(mape, axis = 1), 'R':  np.std(R, axis = 1 ),  'elapsed_time': np.std(elapsed_time, axis = 1 )})\n","\n","  #======= Collecting minimum values of test scores =============#\n","  minimums = pd.DataFrame({'layers': hidden_layers, 'rmse': np.min(rmse, axis =1 ), 'mape': np.min(mape, axis= 1), 'R': np.min(R, axis =1), 'elapsed_time': np.min(elapsed_time, axis =1)})\n","\n","  #======= Collecting maximum values of test scores =============#\n","  maximums = pd.DataFrame({'layers': hidden_layers, 'rmse': np.max(rmse, axis =1), 'mape': np.max(mape, axis =1), 'R': np.max(R, axis =1),  'elapsed_time': np.max(elapsed_time,axis =1)})\n","\n","  output_dictionary = {\n","                     'hyper_parameters': hyper_parameters[i],\n","                     'scores': scores,\n","                     'avg_scores': avg_scores,\n","                     'stds':stds,\n","                     'minimums': minimums,\n","                     'maximums': maximums,\n","                      'best_avg_rmse': best_avg_rmse,\n","                      'best_rmse': best_rmse,\n","                      'best_model_hidden_layers': best_model_hidden_layers,\n","                      'best_model_output': best_model_output\n","                      }\n","\n","\n","  print(\"\\nBest model and its avg rmse and minimum rmse):\\n\", best_model_hidden_layers, best_avg_rmse, best_rmse)\n","  print('\\nAverage scores:\\n', avg_scores)\n","  print('\\nStandard_deviations:\\n', stds)\n","  print('\\nMinimums:\\n', minimums)\n","  print('\\nMaximums:\\n', maximums)\n","\n","\n","  #Save all rmse scores in the file for future analysis\n","  pd.DataFrame(scores['rmse']).to_csv(output_dir_path+'ml-lstm-all-rmse.csv')\n","\n","  #writing output dictionary in the file\n","  file_name = output_dir_path + \"ml-lstm-results.txt\"\n","  write_dic_to_file(output_dictionary, file_name)\n","\n","  print(\"Progress: All works are done successfully, congratulations!!\\n\")\n","\n","  return output_dictionary\n","\n"]},{"cell_type":"markdown","metadata":{"id":"fnSVFxltgwGi"},"source":["# **Functions for Visualizing Multilayer Scores Boxplots**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"JdkLSYHeoNa4"},"outputs":[],"source":["def multi_layers_all_scores_boxplots(model_output):\n","\n","  fig = plt.figure(figsize = (18,5))\n","  plt.subplot(131)\n","  p1 = plt.boxplot(pd.DataFrame(model_output['scores']['rmse']), patch_artist= True )\n","  for i, box in enumerate(p1['boxes']):\n","    #change outline color\n","    box.set(color= 'blue', linewidth = 1.2)\n","    # change fill color\n","    box.set(facecolor = 'mediumblue')\n","  plt.xticks([1,2,3,4,5,6], ['(10, 5)', '(20, 10)', '(50, 20)', '(100, 50)', '(150, 100)', '(100, 50, 20)'], rotation = 30)\n","  plt.title(\"(a)\")\n","  plt.xlabel('Number of neurons')\n","  plt.ylabel('RMSE')\n","\n","  plt.subplot(132)\n","  p2 = plt.boxplot(pd.DataFrame(model_output['scores']['mape']), patch_artist= True )\n","  for i, box in enumerate(p2['boxes']):\n","    #change outline color\n","    box.set(color= 'blue', linewidth = 1.2)\n","    # change fill color\n","    box.set(facecolor = 'indigo')\n","  plt.xticks([1,2,3,4,5,6], ['(10, 5)', '(20, 10)', '(50, 20)', '(100, 50)', '(150, 100)', '(100, 50, 20)'],  rotation = 30)\n","  plt.title(\"(b)\")\n","  plt.xlabel('Number of neurons')\n","  plt.ylabel('MAPE')\n","\n","  plt.subplot(133)\n","  p3 = plt.boxplot(pd.DataFrame(model_output['scores']['R']), patch_artist= True )\n","  for i, box in enumerate(p3['boxes']):\n","    #change outline color\n","    box.set(color= 'blue', linewidth = 1.2)\n","    # change fill color\n","    box.set(facecolor = 'darkgreen')\n","\n","  plt.xticks([1,2,3,4,5,6], ['(10, 5)', '(20, 10)', '(50, 20)', '(100, 50)', '(150, 100)', '(100, 50, 20)'],  rotation = 30)\n","  plt.title(\"(c)\")\n","  plt.xlabel('Number of neurons')\n","  plt.ylabel('R')\n","\n","  fig.savefig(output_dir_path+ \"multi_layers_all_scores_boxplots.png\",dpi=600)\n","  plt.show()\n"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"XU9whQmJ-FdN"},"outputs":[],"source":["import scipy\n","\n","def read_df_from_file(file_name):\n","   return pd.read_csv(file_name)\n","\n","def comparative_qq_plots(data1, data2):\n","  fig = plt.figure(figsize = (14,5))\n","  ax1= fig.add_subplot(121)\n","  scipy.stats.probplot(data1, dist=scipy.stats.norm, sparams=(0,1), plot=ax1)\n","  ax1.get_lines()[0].set_marker('o')\n","  ax1.get_lines()[0].set_markerfacecolor('mediumblue')\n","  ax1.get_lines()[0].set_markersize(8.0)\n","  ax1.get_lines()[1].set_linewidth(3.0)\n","  plt.title(\"(a)\")\n","\n","  ax2= fig.add_subplot(122)\n","  scipy.stats.probplot(data2, dist=scipy.stats.norm, sparams=(0,1), plot=ax2)\n","  ax2.get_lines()[0].set_marker('o')\n","  ax2.get_lines()[0].set_markerfacecolor('mediumblue')\n","  ax2.get_lines()[0].set_markersize(8.0)\n","  ax2.get_lines()[1].set_linewidth(3.0)\n","  plt.title(\"(b)\")\n","\n","  fig.savefig(output_dir_path+ \"Errors_QQ_Plots.png\",dpi=600)\n","  plt.show()\n","\n","\n","def comparative_boxplots(data1, data2):\n","  data = pd.DataFrame()\n","  data['Single-layer-LSTM'] = data1\n","  data['Multi-layer-LSTM'] = data2\n","\n","  fig = plt.figure(figsize = (6,4))\n","  p = plt.boxplot(data, patch_artist= True) # Changed data.T to data\n","  colors = ['mediumblue', 'darkred']\n","  for i, box in enumerate(p['boxes']):\n","    # change outline color\n","    box.set(color= 'blue', linewidth = 1.2)\n","    # change fill color\n","    box.set(facecolor = colors[i])\n","\n","  plt.xticks([1,2], ['Single-layer LSTM','Multi-layer LSTM'])\n","  plt.ylabel('RMSE')\n","  fig.savefig(output_dir_path+\"comparative_boxplots.png\",dpi=600)\n","  plt.show()\n","\n","\n","def perform_normality_test(data1, data2):\n","\n","  print(\"Performaing Normality Tests\\n\")\n","  print(\"Data1: \")\n","  print(scipy.stats.normaltest(data1))\n","  print(\"Data 2:\")\n","  print(scipy.stats.normaltest(data2))\n","\n","#=========F test for equality of variances =====================================\n","def ftest(data1, data2):\n","    x = np.array(data1)\n","    y = np.array(data2)\n","    f = np.var(x, ddof=1)/np.var(y, ddof=1) #calculate F test statistic\n","    dfn = x.size-1 #define degrees of freedom numerator\n","    dfd = y.size-1 #define degrees of freedom denominator\n","    p = 1-scipy.stats.f.cdf(f, dfn, dfd) #find p-value of F test statistic\n","    return f, p\n","\n","\n","def perform_ttest(data1, data2):\n","  #student_ttest, pvalue_student=scipy.stats.ttest_ind(data1, data2, equal_var = True) # It requires variences to be equal.\n","  print(\"\\nTwo-sample ttest\")\n","  print(scipy.stats.ttest_ind(data1, data2, equal_var = False))  # It does not require variences to be equal.\n","\n","\n","def perform_statistical_analysis(data1, data2):\n","  comparative_qq_plots(data1, data2)\n","  comparative_boxplots(data1, data2)\n","  perform_normality_test(data1, data2)\n","  perform_ttest(data1, data2)\n"]},{"cell_type":"markdown","metadata":{"id":"fuzAIoWqzDGX"},"source":["# **Final Step: Models Executions and Results Visualizatoin**"]},{"cell_type":"markdown","metadata":{"id":"gU_3lOc2l6oN"},"source":["### **Case I: Executing Single-Layer Models**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"FsxPTsekmClV","colab":{"base_uri":"https://localhost:8080/"},"outputId":"30376fdf-66cc-4848-801f-5181442821f8"},"outputs":[{"output_type":"stream","name":"stdout","text":["Progress: Performing data preparation steps.......\n","\n","Progress: Building and training models.......\n","\n","Model hyperparameters used:\n","  ['Adam', 0.01, 16]\n","Program is running for 10 neurons and 0 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 4ms/step - loss: 0.0054\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 4.6696e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 3.2575e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 2.8138e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 2.3378e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.6147e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5058e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.6450e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.4106e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.2208e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3774e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.5801e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.2981e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0735e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3440e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0475e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0757e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0441e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.2784e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0271e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 9.3514e-05\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1063e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0279e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1333e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.2888e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 9.2621e-05\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.2360e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3866e-04\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3516e-04\n","Epoch 30/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0365e-04\n","Epoch 31/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.2057e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 1 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0086\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.3104e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 4.2967e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.5324e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.6356e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.6827e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.0192e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.0514e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.6922e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.6527e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.2390e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3834e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1799e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.4561e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2477e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1376e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3547e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1691e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2368e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0080e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3205e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1376e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.5364e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1385e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0033e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3136e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3280e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.3734e-04\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.0784e-04\n","Epoch 30/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.4382e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 2 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0098\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 4.5738e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 3.6893e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 2.7655e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.5507e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8940e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.1245e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.5855e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.4492e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.4744e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3994e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.2095e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1478e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3942e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.3101e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.2460e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0319e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0248e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1756e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1750e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3961e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1256e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0119e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1815e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1636e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0986e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0811e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0082e-04\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 8.9315e-05\n","Epoch 30/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 9.2338e-05\n","Epoch 31/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1752e-04\n","Epoch 32/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0345e-04\n","Epoch 33/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.0978e-05\n","Epoch 34/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3494e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step \n","Program is running for 10 neurons and 3 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 0.0120\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 5.7991e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.3324e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.3802e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6789e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 2.2020e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.8274e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.7476e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.6579e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 2.1973e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3663e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1256e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3014e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4419e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2409e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.3733e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0822e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0693e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1989e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0946e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1612e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.2003e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0003e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 9.9276e-05\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1417e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2188e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.1181e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6097e-04\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0235e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 4 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0047\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.3469e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.2601e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.8628e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3000e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7274e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7219e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5030e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.3162e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1515e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.4702e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.4984e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1828e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0504e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0795e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1398e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0131e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2462e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0971e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3016e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0260e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1389e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 5 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0032\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.8704e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.6639e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3762e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.3148e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.5682e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.4844e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.6219e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.4791e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2262e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2905e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4438e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2012e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3580e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1027e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1359e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3899e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4433e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2671e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1681e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 6 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0062\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.2490e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 5.1076e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 3.7092e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 3.6186e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.3852e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9958e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.0151e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9050e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2877e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2092e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2879e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1652e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0684e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1695e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2740e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1624e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0087e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0830e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1041e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3060e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0529e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.2658e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step \n","Program is running for 10 neurons and 7 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0038\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.8184e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.3064e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6255e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3483e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.2188e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6333e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3278e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2731e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4626e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3748e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4575e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1516e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1340e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5125e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2587e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.0061e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0081e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0676e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3301e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0940e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0861e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 8 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0087\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.6430e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.0829e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.1832e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3558e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.1969e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6543e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8759e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8253e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3595e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2349e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.1869e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1073e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0601e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1328e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.0465e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.4397e-05\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1277e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0664e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2301e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 8.9119e-05\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0062e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.5947e-05\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0722e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0394e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.1773e-05\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 9 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0094\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 4.8946e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 4.2916e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 3.1384e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 2.6323e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 2.3825e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.5185e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.0740e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7940e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5601e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4178e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4515e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4051e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4578e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4853e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3811e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3458e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2013e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3126e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0927e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.2604e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1307e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.6353e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.5840e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.1620e-05\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0460e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.7888e-05\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0929e-04\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2027e-04\n","Epoch 30/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1769e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 10 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0032\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.6061e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.2064e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.5120e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.4913e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7782e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4731e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 1.4513e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.4120e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3577e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.2106e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2190e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2869e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0851e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2262e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2415e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 1.0510e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.9062e-05\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.6113e-05\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0292e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0570e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.8000e-05\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0160e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.3241e-05\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2214e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1118e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 9.0626e-05\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 9.5935e-05\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 9.4250e-05\n","Epoch 30/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1642e-04\n","Epoch 31/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.1379e-04\n","Epoch 32/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.7944e-05\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 11 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0111\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 7.1975e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8126e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.8945e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6984e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3685e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8734e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7049e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3887e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5586e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 1.3327e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2669e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1880e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.4354e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2014e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3777e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4813e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0721e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2880e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3462e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3316e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1409e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0405e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1282e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2267e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1398e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2083e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1988e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step \n","Program is running for 10 neurons and 12 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 0.0087\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 6.2133e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 4.7502e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.9710e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.4548e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.9356e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6417e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.1696e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9774e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7847e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5752e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5618e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3885e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.1723e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2158e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3570e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2053e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3931e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.3760e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3786e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1515e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.4693e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2780e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1170e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1449e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2322e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0799e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.7346e-05\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1633e-04\n","Epoch 30/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0989e-04\n","Epoch 31/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0078e-04\n","Epoch 32/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.6655e-05\n","Epoch 33/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0922e-04\n","Epoch 34/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2572e-04\n","Epoch 35/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3911e-04\n","Epoch 36/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1041e-04\n","Epoch 37/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0628e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step\n","Program is running for 10 neurons and 13 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0164\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 6.2138e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8966e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.5027e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.0844e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6174e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.5048e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8422e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9435e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9380e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6332e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4307e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5143e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6209e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1936e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3317e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 1.3536e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3119e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2195e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.0401e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2686e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3223e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2272e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2197e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2587e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 14 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0036\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.7531e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.9788e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6527e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8225e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9470e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5453e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.3887e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.4181e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2192e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.2077e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.4814e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.2724e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1116e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4645e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4223e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1805e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1502e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1679e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 15 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0136\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.7943e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.1337e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.7682e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.4635e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 2.2047e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.9195e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.8510e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.6756e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7225e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3467e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4855e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1624e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2170e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4517e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1779e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1592e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1409e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1925e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3139e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2257e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.8109e-05\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.2982e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0719e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.0938e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1170e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 9.5907e-05\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0192e-04\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3287e-04\n","Epoch 30/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.0062e-05\n","Epoch 31/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3966e-04\n","Epoch 32/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2031e-04\n","Epoch 33/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.3983e-05\n","Epoch 34/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2787e-04\n","Epoch 35/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1228e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 16 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0030\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.1970e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.4167e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.3842e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 3.1874e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 2.6740e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.2129e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.7598e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.7045e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5694e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5883e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1502e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3763e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4994e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2083e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0843e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0741e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1331e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4516e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1628e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1286e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2752e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 17 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0041\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 4.1244e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 3.6056e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 2.2457e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.6217e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.6266e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9884e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6604e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3124e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0831e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1502e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1541e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0947e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1926e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.4524e-05\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0444e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0152e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0038e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.3126e-05\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.6580e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2074e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0969e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2118e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0737e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 18 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0033\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.3342e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.9783e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.8374e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.1300e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.6983e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7060e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3353e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3147e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3261e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4503e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3855e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.1593e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3766e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2208e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3048e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 9.8981e-05\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.4329e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2329e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0579e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3310e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4852e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 19 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0066\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.3938e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8097e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.8429e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.0024e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6269e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 2.3125e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 2.0666e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.7770e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.7184e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.5321e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.5108e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3970e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4526e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2946e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2921e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1198e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3151e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4514e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1518e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4202e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2060e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 20 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0075\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 6.6320e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 4.5563e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 3.2977e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 2.4926e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 2.1579e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.0461e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6609e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8337e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5240e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2713e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3272e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3208e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1038e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4381e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3000e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8141e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1134e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2681e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 21 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0070\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 5.4909e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 4.4593e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 3.9012e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.1877e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6036e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9685e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.0827e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6812e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8465e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5524e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4608e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8965e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3363e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3062e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2812e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1998e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8109e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1521e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2396e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.5083e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.3024e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1179e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.1429e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2798e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1963e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2034e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4481e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 22 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0089\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 7.0591e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8485e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.4778e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.8374e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.2694e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.0722e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.6921e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.9410e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3207e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1750e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1647e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3611e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1799e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.3164e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3398e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1524e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4318e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0442e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1620e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0826e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3056e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.7408e-05\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4104e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0561e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0615e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.2109e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.0132e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step \n","Program is running for 10 neurons and 23 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 0.0052\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.3116e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.9128e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.6560e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.7204e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3955e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.2630e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.2109e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.6165e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8309e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4153e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2080e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5984e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4991e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.2779e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.5165e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1872e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3467e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3027e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1826e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0309e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3113e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4355e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1342e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1329e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5290e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 24 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0032\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.4190e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.3097e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.7754e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3655e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 2.1778e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.8850e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.6903e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.5260e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3942e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.6830e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3287e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.9459e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2929e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2372e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2225e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2385e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2928e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.2037e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1654e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1875e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1352e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0822e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5122e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.0831e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1219e-04\n","Epoch 27/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.0065e-04\n","Epoch 28/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2174e-04\n","Epoch 29/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.0617e-04\n","Epoch 30/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2080e-04\n","Epoch 31/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0605e-04\n","Epoch 32/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1635e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 25 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0052\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.2336e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.0218e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.3593e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.8260e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3199e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.1835e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.0053e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7239e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.5154e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.4905e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2917e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.3879e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.4062e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.4142e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3195e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1023e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.6121e-05\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1259e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.7972e-05\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.7593e-05\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0256e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0473e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 26 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0131\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.9076e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.1828e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 3.7990e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 2.7977e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 2.8005e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 2.1737e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.8086e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.6982e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.7395e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6791e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5589e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6423e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3527e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.6037e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2589e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.4730e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0870e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0477e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3939e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.7003e-05\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0847e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1040e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2565e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2262e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.0333e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step\n","Program is running for 10 neurons and 27 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0024\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.2489e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.2010e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.5733e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3279e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.0324e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5102e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5738e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1967e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.3061e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3784e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3237e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.2918e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1012e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.1566e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.5212e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 9ms/step - loss: 1.0255e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2485e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0635e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.1248e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3370e-04\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1596e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 10 neurons and 28 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0129\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 7.1911e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.2248e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.3023e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.3318e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3652e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.9897e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.7808e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.5218e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.5552e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.6251e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.5565e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3965e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 1.3243e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3052e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1537e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3699e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3264e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2696e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2835e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 9.7684e-05\n","Epoch 22/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1847e-04\n","Epoch 23/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3756e-04\n","Epoch 24/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0794e-04\n","Epoch 25/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2867e-04\n","Epoch 26/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 1.1984e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step \n","Program is running for 10 neurons and 29 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0039\n","Epoch 2/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 4.7306e-04\n","Epoch 3/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 3.5435e-04\n","Epoch 4/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.8003e-04\n","Epoch 5/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 2.3643e-04\n","Epoch 6/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.8665e-04\n","Epoch 7/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5661e-04\n","Epoch 8/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.5853e-04\n","Epoch 9/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4460e-04\n","Epoch 10/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4090e-04\n","Epoch 11/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.2243e-04\n","Epoch 12/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.3307e-04\n","Epoch 13/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4866e-04\n","Epoch 14/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.4942e-04\n","Epoch 15/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.1631e-04\n","Epoch 16/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 1.0250e-04\n","Epoch 17/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1167e-04\n","Epoch 18/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2031e-04\n","Epoch 19/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 1.1857e-04\n","Epoch 20/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.0615e-04\n","Epoch 21/100\n","\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - loss: 1.2611e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step \n","Model hyperparameters used:\n","  ['Adagrad', 0.01, 8]\n","Program is running for 30 neurons and 0 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0094\n","Epoch 2/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0049\n","Epoch 3/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 0.0033\n","Epoch 4/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 0.0022\n","Epoch 5/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 0.0016\n","Epoch 6/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 0.0012\n","Epoch 7/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 9.7303e-04\n","Epoch 8/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 8.3306e-04\n","Epoch 9/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 7.5773e-04\n","Epoch 10/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 7.0972e-04\n","Epoch 11/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.7505e-04\n","Epoch 12/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.5783e-04\n","Epoch 13/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.4582e-04\n","Epoch 14/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.3471e-04\n","Epoch 15/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.2683e-04\n","Epoch 16/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.1941e-04\n","Epoch 17/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.1477e-04\n","Epoch 18/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 6.0938e-04\n","Epoch 19/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 6.0259e-04\n","Epoch 20/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.9675e-04\n","Epoch 21/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 5.9227e-04\n","Epoch 22/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.9174e-04\n","Epoch 23/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.8837e-04\n","Epoch 24/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.8350e-04\n","Epoch 25/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.7979e-04\n","Epoch 26/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.7807e-04\n","Epoch 27/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.7309e-04\n","Epoch 28/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.6748e-04\n","Epoch 29/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 5.6505e-04\n","Epoch 30/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 5.5951e-04\n","Epoch 31/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.5903e-04\n","Epoch 32/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.5466e-04\n","Epoch 33/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.5163e-04\n","Epoch 34/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.4993e-04\n","Epoch 35/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.4796e-04\n","Epoch 36/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.4305e-04\n","Epoch 37/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 5.4023e-04\n","Epoch 38/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - loss: 5.3834e-04\n","Epoch 39/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 5.3382e-04\n","Epoch 40/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.3337e-04\n","Epoch 41/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.3072e-04\n","Epoch 42/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.2611e-04\n","Epoch 43/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.2366e-04\n","Epoch 44/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.2054e-04\n","Epoch 45/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.1721e-04\n","Epoch 46/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.1555e-04\n","Epoch 47/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.1382e-04\n","Epoch 48/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.1316e-04\n","Epoch 49/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.0956e-04\n","Epoch 50/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.0594e-04\n","Epoch 51/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.0628e-04\n","Epoch 52/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.0281e-04\n","Epoch 53/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 4.9784e-04\n","Epoch 54/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 4.9482e-04\n","Epoch 55/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.9557e-04\n","Epoch 56/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8996e-04\n","Epoch 57/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.9207e-04\n","Epoch 58/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8729e-04\n","Epoch 59/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8695e-04\n","Epoch 60/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8275e-04\n","Epoch 61/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8078e-04\n","Epoch 62/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8078e-04\n","Epoch 63/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 4.7844e-04\n","Epoch 64/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 4.7369e-04\n","Epoch 65/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.7435e-04\n","Epoch 66/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.7085e-04\n","Epoch 67/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.7088e-04\n","Epoch 68/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.6793e-04\n","Epoch 69/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.6835e-04\n","Epoch 70/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.6348e-04\n","Epoch 71/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.6058e-04\n","Epoch 72/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 4.6084e-04\n","Epoch 73/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 4.5854e-04\n","Epoch 74/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 4.5766e-04\n","Epoch 75/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 4.5542e-04\n","Epoch 76/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.5234e-04\n","Epoch 77/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.5306e-04\n","Epoch 78/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.4987e-04\n","Epoch 79/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.4855e-04\n","Epoch 80/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.4707e-04\n","Epoch 81/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.4495e-04\n","Epoch 82/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.4219e-04\n","Epoch 83/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 4.4222e-04\n","Epoch 84/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 4.4011e-04\n","Epoch 85/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 4.3771e-04\n","Epoch 86/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.3645e-04\n","Epoch 87/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.3131e-04\n","Epoch 88/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.3210e-04\n","Epoch 89/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.3283e-04\n","Epoch 90/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.2869e-04\n","Epoch 91/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.2786e-04\n","Epoch 92/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.2756e-04\n","Epoch 93/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.2472e-04\n","Epoch 94/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 4.2499e-04\n","Epoch 95/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 4.2106e-04\n","Epoch 96/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 4.2155e-04\n","Epoch 97/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.1919e-04\n","Epoch 98/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.1915e-04\n","Epoch 99/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.1504e-04\n","Epoch 100/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.1357e-04\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step\n","\u001b[1m16/16\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step \n","Program is running for 30 neurons and 1 replicate ----->\n","\n","Epoch 1/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0116\n","Epoch 2/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 0.0035\n","Epoch 3/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 0.0019\n","Epoch 4/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 0.0012\n","Epoch 5/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 9.2373e-04\n","Epoch 6/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 7.8274e-04\n","Epoch 7/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 7.0791e-04\n","Epoch 8/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.6192e-04\n","Epoch 9/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.3294e-04\n","Epoch 10/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 6.0816e-04\n","Epoch 11/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.9943e-04\n","Epoch 12/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.8588e-04\n","Epoch 13/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.8062e-04\n","Epoch 14/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.6752e-04\n","Epoch 15/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.6456e-04\n","Epoch 16/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.5750e-04\n","Epoch 17/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.5642e-04\n","Epoch 18/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 5.5045e-04\n","Epoch 19/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 5.4703e-04\n","Epoch 20/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.4240e-04\n","Epoch 21/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.3851e-04\n","Epoch 22/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.3677e-04\n","Epoch 23/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.3284e-04\n","Epoch 24/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.2853e-04\n","Epoch 25/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 5.2890e-04\n","Epoch 26/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 5.2636e-04\n","Epoch 27/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 5.2022e-04\n","Epoch 28/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.2029e-04\n","Epoch 29/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.1761e-04\n","Epoch 30/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.1370e-04\n","Epoch 31/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.1192e-04\n","Epoch 32/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.1006e-04\n","Epoch 33/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.0563e-04\n","Epoch 34/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.0687e-04\n","Epoch 35/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 5.0142e-04\n","Epoch 36/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - loss: 4.9694e-04\n","Epoch 37/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 4.9776e-04\n","Epoch 38/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 4.9620e-04\n","Epoch 39/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.9019e-04\n","Epoch 40/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8720e-04\n","Epoch 41/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8869e-04\n","Epoch 42/100\n","\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - loss: 4.8785e-04\n","Epoch 43/100\n","\u001b[1m 45/250\u001b[0m \u001b[32m━━━\u001b[0m\u001b[37m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 4.1228e-04"]}],"source":["import numpy as np\n","\n","neurons = np.array([10, 30, 50, 100, 150,200])\n","\n","best_hyper_parameters = [['Adam', 0.01, 16],# 10N model\n","                      ['Adagrad', 0.01, 8], # 30N model\n","                      ['Adagrad', 0.01, 4], # 50N model\n","                      ['Adagrad', 0.01, 16], # 1000N model\n","                      ['Adagrad', 0.01, 16], # 150N model\n","                      ['Adagrad', 0.001, 4] # 200N model\n","                   ]\n","\n","\n","sl_model_output = LSTM_model(neurons, best_hyper_parameters, data, time_step = 5, test_split = 0.2,\n","                          epochs = 100, num_replicates = 30)\n","\n","create_visualization(sl_model_output)"]},{"cell_type":"markdown","metadata":{"id":"193ec81a"},"source":["Now, let's create a comparative summary of the best performing single-layer and multi-layer models based on their average RMSE, MAPE, and R values."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"8f1e6c82"},"outputs":[],"source":["##  neurons = np.array([10, 30, 50, 100, 150,200])\n","\n","## best_hyper_parameters = [['Adam', 0.01, 16],# 10N model\n","##                      ['Adagrad', 0.01, 8], # 30N model\n","##                      ['Adagrad', 0.01, 4], # 50N model\n","##                      ['Adagrad', 0.01, 16], # 1000N model\n","##                      ['Adagrad', 0.01, 16], # 150N model\n","##                      ['Adagrad', 0.001, 4] # 200N model\n","##                   ]\n","\n","\n","## sl_model_output = LSTM_model(neurons, best_hyper_parameters, data, time_step = 5, test_split = 0.2,\n","                          epochs = 100, num_replicates = 30)\n","\n","## create_visualization(sl_model_output)"]},{"cell_type":"markdown","metadata":{"id":"7j-WjM1UmJuQ"},"source":["### **Case II: Executing Multi-Layer Models**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"pblzmIsbmN1v"},"outputs":[],"source":["hidden_layers = [[10, 5], [20, 10], [50, 20], [100, 50], [150, 100], [100, 50, 20]]\n","\n","best_hyper_parameters_multilayers  = [['Adagrad', 0.1, 4],#10-5N model\n","                                     ['Adagrad', 0.01, 16],#20-10N model\n","                                    ['Adagrad', 0.01, 16], #50-20N model\n","                                   ['Adagrad', 0.01, 16], #100-50N model\n","                                   ['Adagrad', 0.01, 16], #150-100N model\n","                                   ['Adagrad', 0.001, 8] #100-50-20N model\n","                                      ]\n","\n","# Fill NaN values in the data before passing it to the model\n","data = data.ffill().bfill()\n","\n","ml_model_output = run_multi_layer_LSTM_Model(hidden_layers, best_hyper_parameters_multilayers, data, time_step = 5, test_split = 0.2, epochs = 100,  num_replicates = 30)\n","\n","multi_layers_all_scores_boxplots(ml_model_output)\n"]},{"cell_type":"markdown","metadata":{"id":"PjSnbNidmiXy"},"source":["\n","\n","### **Case III: Performing Statistical Analysis**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"d264b34b"},"outputs":[],"source":["print(\"Average scores for Single-Layer LSTM Models:\")\n","display(sl_model_output['avg_scores'])\n","\n","print(\"Average scores for Multi-Layer LSTM Models:\")\n","display(ml_model_output['avg_scores'])"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"nrkDXcn8mtI6"},"outputs":[],"source":["sl_index = 4 # index for the best single layers model(0: 10N,1:30N,2:50N,3:100N,4:150N,5:200N)\n","\n","ml_index = 4  # index for the best multilayer models\n","\n","sl_all_rmse = read_df_from_file(output_dir_path+'sl-lstm-all-rmse.csv')\n","ml_all_rmse = read_df_from_file(output_dir_path+'ml-lstm-all-rmse.csv')\n","\n","data1 = sl_all_rmse.iloc[sl_index,1:]\n","data2 = ml_all_rmse.iloc[ml_index,1:]\n","\n","perform_statistical_analysis(data1, data2)\n"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"5d985774"},"outputs":[],"source":["best_sl_model_avg = sl_model_output['avg_scores'].loc[sl_model_output['avg_scores']['rmse'].idxmin()]\n","best_ml_model_avg = ml_model_output['avg_scores'].loc[ml_model_output['avg_scores']['rmse'].idxmin()]\n","\n","comparison_df = pd.DataFrame({\n","    'Metric': ['RMSE', 'MAPE', 'R'],\n","    'Best Single-Layer Model (Avg)': [best_sl_model_avg['rmse'], best_sl_model_avg['mape'], best_sl_model_avg['R']],\n","    'Best Multi-Layer Model (Avg)': [best_ml_model_avg['rmse'], best_ml_model_avg['mape'], best_ml_model_avg['R']]\n","})\n","\n","print(\"Comparison of Best Performing Models (Average Scores):\")\n","display(comparison_df)"]}],"metadata":{"accelerator":"GPU","colab":{"gpuType":"T4","machine_shape":"hm","provenance":[]},"kernelspec":{"display_name":"Python 3","name":"python3"}},"nbformat":4,"nbformat_minor":0}