├── Calculation_&_Visualization_SOH.ipynb
├── Calculation_and_Visualization_of_refined_SoH.ipynb
├── README.md
├── SOH_estimation_with_LSTM.ipynb
├── batterysoc.mlx
├── neuralnetwork.ipynb
├── ridgemsoc.mlx
└── smdatasoc.mlx


/README.md:
--------------------------------------------------------------------------------
1 | # Battery-management-system-using-machine-learning
2 | Machine learning can enhance a BMS by improving SOC and SOH estimation, detecting faults, and optimizing control policies. By leveraging data from the battery and other sensors, ML can lead to more accurate and efficient battery management, enhancing safety, performance, and sustainability.
3 | 


--------------------------------------------------------------------------------
/SOH_estimation_with_LSTM.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |   "nbformat": 4,
   3 |   "nbformat_minor": 0,
   4 |   "metadata": {
   5 |     "colab": {
   6 |       "provenance": []
   7 |     },
   8 |     "kernelspec": {
   9 |       "name": "python3",
  10 |       "display_name": "Python 3"
  11 |     },
  12 |     "language_info": {
  13 |       "name": "python"
  14 |     }
  15 |   },
  16 |   "cells": [
  17 |     {
  18 |       "cell_type": "code",
  19 |       "execution_count": null,
  20 |       "metadata": {
  21 |         "id": "rNjQvbX-puem"
  22 |       },
  23 |       "outputs": [],
  24 |       "source": [
  25 |         "import numpy as np\n",
  26 |         "import pandas as pd\n",
  27 |         "import matplotlib.pyplot as plt\n",
  28 |         "import tensorflow as tf\n",
  29 |         "import keras\n",
  30 |         "import os\n",
  31 |         "import glob\n",
  32 |         "import seaborn as sns\n",
  33 |         "from mpl_toolkits.mplot3d import Axes3D\n",
  34 |         "import math\n",
  35 |         "\n",
  36 |         "import tensorflow as tf\n",
  37 |         "from sklearn.preprocessing import MinMaxScaler\n",
  38 |         "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
  39 |         "from keras.preprocessing.sequence import TimeseriesGenerator\n",
  40 |         "from keras.models import Sequential\n",
  41 |         "from keras.layers import Dense, LSTM, SimpleRNN, Dropout\n",
  42 |         "from keras.optimizers import RMSprop\n",
  43 |         "from keras.callbacks import ModelCheckpoint, EarlyStopping\n",
  44 |         "from keras.models import model_from_json\n",
  45 |         "from keras import optimizers"
  46 |       ]
  47 |     },
  48 |     {
  49 |       "cell_type": "code",
  50 |       "source": [
  51 |         "dir = 'refined_dataset'\n",
  52 |         "listdir = os.listdir(dir)\n",
  53 |         "\n",
  54 |         "print(listdir)\n",
  55 |         "print(\"The number of dataset :\", len(listdir))"
  56 |       ],
  57 |       "metadata": {
  58 |         "colab": {
  59 |           "base_uri": "https://localhost:8080/"
  60 |         },
  61 |         "id": "yhogkYGjp5av",
  62 |         "outputId": "1792c9a9-3245-44f4-c588-bae700c8fddc"
  63 |       },
  64 |       "execution_count": null,
  65 |       "outputs": [
  66 |         {
  67 |           "output_type": "stream",
  68 |           "name": "stdout",
  69 |           "text": [
  70 |             "['B05_discharge_soh.csv', 'B18_discharge_soh.csv', 'B33_discharge_soh.csv', 'B48_discharge_soh.csv', 'B46_discharge_soh.csv', 'B34_discharge_soh.csv', 'B47_discharge_soh.csv', 'B07_discharge_soh.csv']\n",
  71 |             "The number of dataset : 8\n"
  72 |           ]
  73 |         }
  74 |       ]
  75 |     },
  76 |     {
  77 |       "cell_type": "code",
  78 |       "source": [
  79 |         "num = ['B05', 'B07', 'B18', 'B33', 'B34', 'B46', 'B47', 'B48']\n",
  80 |         "for i in range(len(listdir)):\n",
  81 |         "    vector = np.zeros((1,3))\n",
  82 |         "    path = os.path.join(os.getcwd(), 'refined_dataset/', num[i] + '_discharge_soh.csv')\n",
  83 |         "    csv = pd.read_csv(path)\n",
  84 |         "    df = pd.DataFrame(csv)\n",
  85 |         "    \n",
  86 |         "    vec = df[['cycle', 'capacity', 'SOH']]\n",
  87 |         "    \n",
  88 |         "    globals()['data_{}'.format(num[i])] = vec"
  89 |       ],
  90 |       "metadata": {
  91 |         "id": "fkihXMEoqk6I"
  92 |       },
  93 |       "execution_count": null,
  94 |       "outputs": []
  95 |     },
  96 |     {
  97 |       "cell_type": "code",
  98 |       "source": [
  99 |         "data = pd.read_csv('refined_dataset/B05_discharge_soh.csv')\n",
 100 |         "df = pd.DataFrame(data)\n",
 101 |         "df"
 102 |       ],
 103 |       "metadata": {
 104 |         "colab": {
 105 |           "base_uri": "https://localhost:8080/",
 106 |           "height": 423
 107 |         },
 108 |         "id": "7yuOZtfjqv10",
 109 |         "outputId": "1c03a0c5-6805-495f-9183-af0835c99f75"
 110 |       },
 111 |       "execution_count": null,
 112 |       "outputs": [
 113 |         {
 114 |           "output_type": "execute_result",
 115 |           "data": {
 116 |             "text/plain": [
 117 |               "       terminal_voltage  terminal_current  temperature  charge_current  \\\n",
 118 |               "0              4.191492         -0.004902    24.330034         -0.0006   \n",
 119 |               "1              4.190749         -0.001478    24.325993         -0.0006   \n",
 120 |               "2              3.974871         -2.012528    24.389085         -1.9982   \n",
 121 |               "3              3.951717         -2.013979    24.544752         -1.9982   \n",
 122 |               "4              3.934352         -2.011144    24.731385         -1.9982   \n",
 123 |               "...                 ...               ...          ...             ...   \n",
 124 |               "50280          3.579262         -0.001569    34.864823          0.0006   \n",
 125 |               "50281          3.581964         -0.003067    34.814770          0.0006   \n",
 126 |               "50282          3.584484         -0.003079    34.676258          0.0006   \n",
 127 |               "50283          3.587336          0.001219    34.565580          0.0006   \n",
 128 |               "50284          3.589937         -0.000583    34.405920          0.0006   \n",
 129 |               "\n",
 130 |               "       charge_voltage      time  capacity  cycle       SOH  \n",
 131 |               "0               0.000     0.000  1.856487      1  0.928244  \n",
 132 |               "1               4.206    16.781  1.856487      1  0.928244  \n",
 133 |               "2               3.062    35.703  1.856487      1  0.928244  \n",
 134 |               "3               3.030    53.781  1.856487      1  0.928244  \n",
 135 |               "4               3.011    71.922  1.856487      1  0.928244  \n",
 136 |               "...               ...       ...       ...    ...       ...  \n",
 137 |               "50280           0.000  2781.312  1.325079    168  0.662540  \n",
 138 |               "50281           0.000  2791.062  1.325079    168  0.662540  \n",
 139 |               "50282           0.000  2800.828  1.325079    168  0.662540  \n",
 140 |               "50283           0.000  2810.640  1.325079    168  0.662540  \n",
 141 |               "50284           0.000  2820.390  1.325079    168  0.662540  \n",
 142 |               "\n",
 143 |               "[50285 rows x 9 columns]"
 144 |             ],
 145 |             "text/html": [
 146 |               "\n",
 147 |               "  <div id=\"df-4c5e174e-6dde-4763-9b69-ce49adbaacc0\">\n",
 148 |               "    <div class=\"colab-df-container\">\n",
 149 |               "      <div>\n",
 150 |               "<style scoped>\n",
 151 |               "    .dataframe tbody tr th:only-of-type {\n",
 152 |               "        vertical-align: middle;\n",
 153 |               "    }\n",
 154 |               "\n",
 155 |               "    .dataframe tbody tr th {\n",
 156 |               "        vertical-align: top;\n",
 157 |               "    }\n",
 158 |               "\n",
 159 |               "    .dataframe thead th {\n",
 160 |               "        text-align: right;\n",
 161 |               "    }\n",
 162 |               "</style>\n",
 163 |               "<table border=\"1\" class=\"dataframe\">\n",
 164 |               "  <thead>\n",
 165 |               "    <tr style=\"text-align: right;\">\n",
 166 |               "      <th></th>\n",
 167 |               "      <th>terminal_voltage</th>\n",
 168 |               "      <th>terminal_current</th>\n",
 169 |               "      <th>temperature</th>\n",
 170 |               "      <th>charge_current</th>\n",
 171 |               "      <th>charge_voltage</th>\n",
 172 |               "      <th>time</th>\n",
 173 |               "      <th>capacity</th>\n",
 174 |               "      <th>cycle</th>\n",
 175 |               "      <th>SOH</th>\n",
 176 |               "    </tr>\n",
 177 |               "  </thead>\n",
 178 |               "  <tbody>\n",
 179 |               "    <tr>\n",
 180 |               "      <th>0</th>\n",
 181 |               "      <td>4.191492</td>\n",
 182 |               "      <td>-0.004902</td>\n",
 183 |               "      <td>24.330034</td>\n",
 184 |               "      <td>-0.0006</td>\n",
 185 |               "      <td>0.000</td>\n",
 186 |               "      <td>0.000</td>\n",
 187 |               "      <td>1.856487</td>\n",
 188 |               "      <td>1</td>\n",
 189 |               "      <td>0.928244</td>\n",
 190 |               "    </tr>\n",
 191 |               "    <tr>\n",
 192 |               "      <th>1</th>\n",
 193 |               "      <td>4.190749</td>\n",
 194 |               "      <td>-0.001478</td>\n",
 195 |               "      <td>24.325993</td>\n",
 196 |               "      <td>-0.0006</td>\n",
 197 |               "      <td>4.206</td>\n",
 198 |               "      <td>16.781</td>\n",
 199 |               "      <td>1.856487</td>\n",
 200 |               "      <td>1</td>\n",
 201 |               "      <td>0.928244</td>\n",
 202 |               "    </tr>\n",
 203 |               "    <tr>\n",
 204 |               "      <th>2</th>\n",
 205 |               "      <td>3.974871</td>\n",
 206 |               "      <td>-2.012528</td>\n",
 207 |               "      <td>24.389085</td>\n",
 208 |               "      <td>-1.9982</td>\n",
 209 |               "      <td>3.062</td>\n",
 210 |               "      <td>35.703</td>\n",
 211 |               "      <td>1.856487</td>\n",
 212 |               "      <td>1</td>\n",
 213 |               "      <td>0.928244</td>\n",
 214 |               "    </tr>\n",
 215 |               "    <tr>\n",
 216 |               "      <th>3</th>\n",
 217 |               "      <td>3.951717</td>\n",
 218 |               "      <td>-2.013979</td>\n",
 219 |               "      <td>24.544752</td>\n",
 220 |               "      <td>-1.9982</td>\n",
 221 |               "      <td>3.030</td>\n",
 222 |               "      <td>53.781</td>\n",
 223 |               "      <td>1.856487</td>\n",
 224 |               "      <td>1</td>\n",
 225 |               "      <td>0.928244</td>\n",
 226 |               "    </tr>\n",
 227 |               "    <tr>\n",
 228 |               "      <th>4</th>\n",
 229 |               "      <td>3.934352</td>\n",
 230 |               "      <td>-2.011144</td>\n",
 231 |               "      <td>24.731385</td>\n",
 232 |               "      <td>-1.9982</td>\n",
 233 |               "      <td>3.011</td>\n",
 234 |               "      <td>71.922</td>\n",
 235 |               "      <td>1.856487</td>\n",
 236 |               "      <td>1</td>\n",
 237 |               "      <td>0.928244</td>\n",
 238 |               "    </tr>\n",
 239 |               "    <tr>\n",
 240 |               "      <th>...</th>\n",
 241 |               "      <td>...</td>\n",
 242 |               "      <td>...</td>\n",
 243 |               "      <td>...</td>\n",
 244 |               "      <td>...</td>\n",
 245 |               "      <td>...</td>\n",
 246 |               "      <td>...</td>\n",
 247 |               "      <td>...</td>\n",
 248 |               "      <td>...</td>\n",
 249 |               "      <td>...</td>\n",
 250 |               "    </tr>\n",
 251 |               "    <tr>\n",
 252 |               "      <th>50280</th>\n",
 253 |               "      <td>3.579262</td>\n",
 254 |               "      <td>-0.001569</td>\n",
 255 |               "      <td>34.864823</td>\n",
 256 |               "      <td>0.0006</td>\n",
 257 |               "      <td>0.000</td>\n",
 258 |               "      <td>2781.312</td>\n",
 259 |               "      <td>1.325079</td>\n",
 260 |               "      <td>168</td>\n",
 261 |               "      <td>0.662540</td>\n",
 262 |               "    </tr>\n",
 263 |               "    <tr>\n",
 264 |               "      <th>50281</th>\n",
 265 |               "      <td>3.581964</td>\n",
 266 |               "      <td>-0.003067</td>\n",
 267 |               "      <td>34.814770</td>\n",
 268 |               "      <td>0.0006</td>\n",
 269 |               "      <td>0.000</td>\n",
 270 |               "      <td>2791.062</td>\n",
 271 |               "      <td>1.325079</td>\n",
 272 |               "      <td>168</td>\n",
 273 |               "      <td>0.662540</td>\n",
 274 |               "    </tr>\n",
 275 |               "    <tr>\n",
 276 |               "      <th>50282</th>\n",
 277 |               "      <td>3.584484</td>\n",
 278 |               "      <td>-0.003079</td>\n",
 279 |               "      <td>34.676258</td>\n",
 280 |               "      <td>0.0006</td>\n",
 281 |               "      <td>0.000</td>\n",
 282 |               "      <td>2800.828</td>\n",
 283 |               "      <td>1.325079</td>\n",
 284 |               "      <td>168</td>\n",
 285 |               "      <td>0.662540</td>\n",
 286 |               "    </tr>\n",
 287 |               "    <tr>\n",
 288 |               "      <th>50283</th>\n",
 289 |               "      <td>3.587336</td>\n",
 290 |               "      <td>0.001219</td>\n",
 291 |               "      <td>34.565580</td>\n",
 292 |               "      <td>0.0006</td>\n",
 293 |               "      <td>0.000</td>\n",
 294 |               "      <td>2810.640</td>\n",
 295 |               "      <td>1.325079</td>\n",
 296 |               "      <td>168</td>\n",
 297 |               "      <td>0.662540</td>\n",
 298 |               "    </tr>\n",
 299 |               "    <tr>\n",
 300 |               "      <th>50284</th>\n",
 301 |               "      <td>3.589937</td>\n",
 302 |               "      <td>-0.000583</td>\n",
 303 |               "      <td>34.405920</td>\n",
 304 |               "      <td>0.0006</td>\n",
 305 |               "      <td>0.000</td>\n",
 306 |               "      <td>2820.390</td>\n",
 307 |               "      <td>1.325079</td>\n",
 308 |               "      <td>168</td>\n",
 309 |               "      <td>0.662540</td>\n",
 310 |               "    </tr>\n",
 311 |               "  </tbody>\n",
 312 |               "</table>\n",
 313 |               "<p>50285 rows × 9 columns</p>\n",
 314 |               "</div>\n",
 315 |               "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4c5e174e-6dde-4763-9b69-ce49adbaacc0')\"\n",
 316 |               "              title=\"Convert this dataframe to an interactive table.\"\n",
 317 |               "              style=\"display:none;\">\n",
 318 |               "        \n",
 319 |               "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
 320 |               "       width=\"24px\">\n",
 321 |               "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
 322 |               "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
 323 |               "  </svg>\n",
 324 |               "      </button>\n",
 325 |               "      \n",
 326 |               "  <style>\n",
 327 |               "    .colab-df-container {\n",
 328 |               "      display:flex;\n",
 329 |               "      flex-wrap:wrap;\n",
 330 |               "      gap: 12px;\n",
 331 |               "    }\n",
 332 |               "\n",
 333 |               "    .colab-df-convert {\n",
 334 |               "      background-color: #E8F0FE;\n",
 335 |               "      border: none;\n",
 336 |               "      border-radius: 50%;\n",
 337 |               "      cursor: pointer;\n",
 338 |               "      display: none;\n",
 339 |               "      fill: #1967D2;\n",
 340 |               "      height: 32px;\n",
 341 |               "      padding: 0 0 0 0;\n",
 342 |               "      width: 32px;\n",
 343 |               "    }\n",
 344 |               "\n",
 345 |               "    .colab-df-convert:hover {\n",
 346 |               "      background-color: #E2EBFA;\n",
 347 |               "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
 348 |               "      fill: #174EA6;\n",
 349 |               "    }\n",
 350 |               "\n",
 351 |               "    [theme=dark] .colab-df-convert {\n",
 352 |               "      background-color: #3B4455;\n",
 353 |               "      fill: #D2E3FC;\n",
 354 |               "    }\n",
 355 |               "\n",
 356 |               "    [theme=dark] .colab-df-convert:hover {\n",
 357 |               "      background-color: #434B5C;\n",
 358 |               "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
 359 |               "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
 360 |               "      fill: #FFFFFF;\n",
 361 |               "    }\n",
 362 |               "  </style>\n",
 363 |               "\n",
 364 |               "      <script>\n",
 365 |               "        const buttonEl =\n",
 366 |               "          document.querySelector('#df-4c5e174e-6dde-4763-9b69-ce49adbaacc0 button.colab-df-convert');\n",
 367 |               "        buttonEl.style.display =\n",
 368 |               "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
 369 |               "\n",
 370 |               "        async function convertToInteractive(key) {\n",
 371 |               "          const element = document.querySelector('#df-4c5e174e-6dde-4763-9b69-ce49adbaacc0');\n",
 372 |               "          const dataTable =\n",
 373 |               "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
 374 |               "                                                     [key], {});\n",
 375 |               "          if (!dataTable) return;\n",
 376 |               "\n",
 377 |               "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
 378 |               "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
 379 |               "            + ' to learn more about interactive tables.';\n",
 380 |               "          element.innerHTML = '';\n",
 381 |               "          dataTable['output_type'] = 'display_data';\n",
 382 |               "          await google.colab.output.renderOutput(dataTable, element);\n",
 383 |               "          const docLink = document.createElement('div');\n",
 384 |               "          docLink.innerHTML = docLinkHtml;\n",
 385 |               "          element.appendChild(docLink);\n",
 386 |               "        }\n",
 387 |               "      </script>\n",
 388 |               "    </div>\n",
 389 |               "  </div>\n",
 390 |               "  "
 391 |             ]
 392 |           },
 393 |           "metadata": {},
 394 |           "execution_count": 5
 395 |         }
 396 |       ]
 397 |     },
 398 |     {
 399 |       "cell_type": "code",
 400 |       "source": [
 401 |         "data_B05 "
 402 |       ],
 403 |       "metadata": {
 404 |         "colab": {
 405 |           "base_uri": "https://localhost:8080/",
 406 |           "height": 423
 407 |         },
 408 |         "id": "qE7uMG93qzmJ",
 409 |         "outputId": "0b5929ea-ee5d-49ef-8ccf-aef2882d1b28"
 410 |       },
 411 |       "execution_count": null,
 412 |       "outputs": [
 413 |         {
 414 |           "output_type": "execute_result",
 415 |           "data": {
 416 |             "text/plain": [
 417 |               "       cycle  capacity       SOH\n",
 418 |               "0          1  1.856487  0.928244\n",
 419 |               "1          1  1.856487  0.928244\n",
 420 |               "2          1  1.856487  0.928244\n",
 421 |               "3          1  1.856487  0.928244\n",
 422 |               "4          1  1.856487  0.928244\n",
 423 |               "...      ...       ...       ...\n",
 424 |               "50280    168  1.325079  0.662540\n",
 425 |               "50281    168  1.325079  0.662540\n",
 426 |               "50282    168  1.325079  0.662540\n",
 427 |               "50283    168  1.325079  0.662540\n",
 428 |               "50284    168  1.325079  0.662540\n",
 429 |               "\n",
 430 |               "[50285 rows x 3 columns]"
 431 |             ],
 432 |             "text/html": [
 433 |               "\n",
 434 |               "  <div id=\"df-fd1250c9-1da5-43e6-b58a-00e006e5f65a\">\n",
 435 |               "    <div class=\"colab-df-container\">\n",
 436 |               "      <div>\n",
 437 |               "<style scoped>\n",
 438 |               "    .dataframe tbody tr th:only-of-type {\n",
 439 |               "        vertical-align: middle;\n",
 440 |               "    }\n",
 441 |               "\n",
 442 |               "    .dataframe tbody tr th {\n",
 443 |               "        vertical-align: top;\n",
 444 |               "    }\n",
 445 |               "\n",
 446 |               "    .dataframe thead th {\n",
 447 |               "        text-align: right;\n",
 448 |               "    }\n",
 449 |               "</style>\n",
 450 |               "<table border=\"1\" class=\"dataframe\">\n",
 451 |               "  <thead>\n",
 452 |               "    <tr style=\"text-align: right;\">\n",
 453 |               "      <th></th>\n",
 454 |               "      <th>cycle</th>\n",
 455 |               "      <th>capacity</th>\n",
 456 |               "      <th>SOH</th>\n",
 457 |               "    </tr>\n",
 458 |               "  </thead>\n",
 459 |               "  <tbody>\n",
 460 |               "    <tr>\n",
 461 |               "      <th>0</th>\n",
 462 |               "      <td>1</td>\n",
 463 |               "      <td>1.856487</td>\n",
 464 |               "      <td>0.928244</td>\n",
 465 |               "    </tr>\n",
 466 |               "    <tr>\n",
 467 |               "      <th>1</th>\n",
 468 |               "      <td>1</td>\n",
 469 |               "      <td>1.856487</td>\n",
 470 |               "      <td>0.928244</td>\n",
 471 |               "    </tr>\n",
 472 |               "    <tr>\n",
 473 |               "      <th>2</th>\n",
 474 |               "      <td>1</td>\n",
 475 |               "      <td>1.856487</td>\n",
 476 |               "      <td>0.928244</td>\n",
 477 |               "    </tr>\n",
 478 |               "    <tr>\n",
 479 |               "      <th>3</th>\n",
 480 |               "      <td>1</td>\n",
 481 |               "      <td>1.856487</td>\n",
 482 |               "      <td>0.928244</td>\n",
 483 |               "    </tr>\n",
 484 |               "    <tr>\n",
 485 |               "      <th>4</th>\n",
 486 |               "      <td>1</td>\n",
 487 |               "      <td>1.856487</td>\n",
 488 |               "      <td>0.928244</td>\n",
 489 |               "    </tr>\n",
 490 |               "    <tr>\n",
 491 |               "      <th>...</th>\n",
 492 |               "      <td>...</td>\n",
 493 |               "      <td>...</td>\n",
 494 |               "      <td>...</td>\n",
 495 |               "    </tr>\n",
 496 |               "    <tr>\n",
 497 |               "      <th>50280</th>\n",
 498 |               "      <td>168</td>\n",
 499 |               "      <td>1.325079</td>\n",
 500 |               "      <td>0.662540</td>\n",
 501 |               "    </tr>\n",
 502 |               "    <tr>\n",
 503 |               "      <th>50281</th>\n",
 504 |               "      <td>168</td>\n",
 505 |               "      <td>1.325079</td>\n",
 506 |               "      <td>0.662540</td>\n",
 507 |               "    </tr>\n",
 508 |               "    <tr>\n",
 509 |               "      <th>50282</th>\n",
 510 |               "      <td>168</td>\n",
 511 |               "      <td>1.325079</td>\n",
 512 |               "      <td>0.662540</td>\n",
 513 |               "    </tr>\n",
 514 |               "    <tr>\n",
 515 |               "      <th>50283</th>\n",
 516 |               "      <td>168</td>\n",
 517 |               "      <td>1.325079</td>\n",
 518 |               "      <td>0.662540</td>\n",
 519 |               "    </tr>\n",
 520 |               "    <tr>\n",
 521 |               "      <th>50284</th>\n",
 522 |               "      <td>168</td>\n",
 523 |               "      <td>1.325079</td>\n",
 524 |               "      <td>0.662540</td>\n",
 525 |               "    </tr>\n",
 526 |               "  </tbody>\n",
 527 |               "</table>\n",
 528 |               "<p>50285 rows × 3 columns</p>\n",
 529 |               "</div>\n",
 530 |               "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fd1250c9-1da5-43e6-b58a-00e006e5f65a')\"\n",
 531 |               "              title=\"Convert this dataframe to an interactive table.\"\n",
 532 |               "              style=\"display:none;\">\n",
 533 |               "        \n",
 534 |               "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
 535 |               "       width=\"24px\">\n",
 536 |               "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
 537 |               "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
 538 |               "  </svg>\n",
 539 |               "      </button>\n",
 540 |               "      \n",
 541 |               "  <style>\n",
 542 |               "    .colab-df-container {\n",
 543 |               "      display:flex;\n",
 544 |               "      flex-wrap:wrap;\n",
 545 |               "      gap: 12px;\n",
 546 |               "    }\n",
 547 |               "\n",
 548 |               "    .colab-df-convert {\n",
 549 |               "      background-color: #E8F0FE;\n",
 550 |               "      border: none;\n",
 551 |               "      border-radius: 50%;\n",
 552 |               "      cursor: pointer;\n",
 553 |               "      display: none;\n",
 554 |               "      fill: #1967D2;\n",
 555 |               "      height: 32px;\n",
 556 |               "      padding: 0 0 0 0;\n",
 557 |               "      width: 32px;\n",
 558 |               "    }\n",
 559 |               "\n",
 560 |               "    .colab-df-convert:hover {\n",
 561 |               "      background-color: #E2EBFA;\n",
 562 |               "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
 563 |               "      fill: #174EA6;\n",
 564 |               "    }\n",
 565 |               "\n",
 566 |               "    [theme=dark] .colab-df-convert {\n",
 567 |               "      background-color: #3B4455;\n",
 568 |               "      fill: #D2E3FC;\n",
 569 |               "    }\n",
 570 |               "\n",
 571 |               "    [theme=dark] .colab-df-convert:hover {\n",
 572 |               "      background-color: #434B5C;\n",
 573 |               "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
 574 |               "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
 575 |               "      fill: #FFFFFF;\n",
 576 |               "    }\n",
 577 |               "  </style>\n",
 578 |               "\n",
 579 |               "      <script>\n",
 580 |               "        const buttonEl =\n",
 581 |               "          document.querySelector('#df-fd1250c9-1da5-43e6-b58a-00e006e5f65a button.colab-df-convert');\n",
 582 |               "        buttonEl.style.display =\n",
 583 |               "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
 584 |               "\n",
 585 |               "        async function convertToInteractive(key) {\n",
 586 |               "          const element = document.querySelector('#df-fd1250c9-1da5-43e6-b58a-00e006e5f65a');\n",
 587 |               "          const dataTable =\n",
 588 |               "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
 589 |               "                                                     [key], {});\n",
 590 |               "          if (!dataTable) return;\n",
 591 |               "\n",
 592 |               "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
 593 |               "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
 594 |               "            + ' to learn more about interactive tables.';\n",
 595 |               "          element.innerHTML = '';\n",
 596 |               "          dataTable['output_type'] = 'display_data';\n",
 597 |               "          await google.colab.output.renderOutput(dataTable, element);\n",
 598 |               "          const docLink = document.createElement('div');\n",
 599 |               "          docLink.innerHTML = docLinkHtml;\n",
 600 |               "          element.appendChild(docLink);\n",
 601 |               "        }\n",
 602 |               "      </script>\n",
 603 |               "    </div>\n",
 604 |               "  </div>\n",
 605 |               "  "
 606 |             ]
 607 |           },
 608 |           "metadata": {},
 609 |           "execution_count": 6
 610 |         }
 611 |       ]
 612 |     },
 613 |     {
 614 |       "cell_type": "code",
 615 |       "source": [
 616 |         "for i in range(len(listdir)):\n",
 617 |         "    print(\"Shape of data :\", np.shape(globals()['data_{}'.format(num[i])]))"
 618 |       ],
 619 |       "metadata": {
 620 |         "colab": {
 621 |           "base_uri": "https://localhost:8080/"
 622 |         },
 623 |         "id": "W7JQGYDqq2Bi",
 624 |         "outputId": "d7456e0f-f872-46cc-89d7-d616181f5f92"
 625 |       },
 626 |       "execution_count": null,
 627 |       "outputs": [
 628 |         {
 629 |           "output_type": "stream",
 630 |           "name": "stdout",
 631 |           "text": [
 632 |             "Shape of data : (50285, 3)\n",
 633 |             "Shape of data : (50285, 3)\n",
 634 |             "Shape of data : (34866, 3)\n",
 635 |             "Shape of data : (42200, 3)\n",
 636 |             "Shape of data : (45909, 3)\n",
 637 |             "Shape of data : (23478, 3)\n",
 638 |             "Shape of data : (23478, 3)\n",
 639 |             "Shape of data : (23478, 3)\n"
 640 |           ]
 641 |         }
 642 |       ]
 643 |     },
 644 |     {
 645 |       "cell_type": "code",
 646 |       "source": [
 647 |         "print(listdir)\n",
 648 |         "dataset = data_B48[\"SOH\"]\n",
 649 |         "cycle = data_B48['cycle']\n",
 650 |         "\n",
 651 |         "dataset = np.array(dataset)\n",
 652 |         "dataset = dataset.reshape((len(dataset),1))\n",
 653 |         "dataset.shape"
 654 |       ],
 655 |       "metadata": {
 656 |         "colab": {
 657 |           "base_uri": "https://localhost:8080/"
 658 |         },
 659 |         "id": "moJemyXfq4pj",
 660 |         "outputId": "d98c22df-73dd-4bc3-eaae-289d22518e4f"
 661 |       },
 662 |       "execution_count": null,
 663 |       "outputs": [
 664 |         {
 665 |           "output_type": "stream",
 666 |           "name": "stdout",
 667 |           "text": [
 668 |             "['B05_discharge_soh.csv', 'B18_discharge_soh.csv', 'B33_discharge_soh.csv', 'B48_discharge_soh.csv', 'B46_discharge_soh.csv', 'B34_discharge_soh.csv', 'B47_discharge_soh.csv', 'B07_discharge_soh.csv']\n"
 669 |           ]
 670 |         },
 671 |         {
 672 |           "output_type": "execute_result",
 673 |           "data": {
 674 |             "text/plain": [
 675 |               "(23478, 1)"
 676 |             ]
 677 |           },
 678 |           "metadata": {},
 679 |           "execution_count": 9
 680 |         }
 681 |       ]
 682 |     },
 683 |     {
 684 |       "cell_type": "markdown",
 685 |       "source": [
 686 |         "SPLIT TRAIN AND TEST DATA"
 687 |       ],
 688 |       "metadata": {
 689 |         "id": "I9k5UDrCrQzF"
 690 |       }
 691 |     },
 692 |     {
 693 |       "cell_type": "code",
 694 |       "source": [
 695 |         "train_size = int(len(dataset) * 0.5)\n",
 696 |         "# train_size = int(len(dataset) * 0.5)\n",
 697 |         "test_size = len(dataset) - train_size\n",
 698 |         "train, test = dataset[0:train_size,:], dataset[train_size:len(dataset),:]\n",
 699 |         "print(len(train), len(test))"
 700 |       ],
 701 |       "metadata": {
 702 |         "colab": {
 703 |           "base_uri": "https://localhost:8080/"
 704 |         },
 705 |         "id": "Kl3LuTn2rOyF",
 706 |         "outputId": "9d8351c9-0472-478d-c6b2-d7fcf95eaf80"
 707 |       },
 708 |       "execution_count": null,
 709 |       "outputs": [
 710 |         {
 711 |           "output_type": "stream",
 712 |           "name": "stdout",
 713 |           "text": [
 714 |             "11739 11739\n"
 715 |           ]
 716 |         }
 717 |       ]
 718 |     },
 719 |     {
 720 |       "cell_type": "code",
 721 |       "source": [
 722 |         "def create_dataset(dataset, look_back=1):\n",
 723 |         "    dataX, dataY = [], []\n",
 724 |         "    for i in range(len(dataset) - look_back):\n",
 725 |         "        a = dataset[i:(i + look_back), 0]\n",
 726 |         "        dataX.append(a)\n",
 727 |         "        dataY.append(dataset[i + look_back, 0])\n",
 728 |         "    print(len(dataY))\n",
 729 |         "    return np.array(dataX), np.array(dataY)"
 730 |       ],
 731 |       "metadata": {
 732 |         "id": "fDVek3perWHH"
 733 |       },
 734 |       "execution_count": null,
 735 |       "outputs": []
 736 |     },
 737 |     {
 738 |       "cell_type": "code",
 739 |       "source": [
 740 |         "look_back = 1\n",
 741 |         "trainX, trainY = create_dataset(train, look_back)\n",
 742 |         "testX, testY = create_dataset(test, look_back)"
 743 |       ],
 744 |       "metadata": {
 745 |         "colab": {
 746 |           "base_uri": "https://localhost:8080/"
 747 |         },
 748 |         "id": "4WksHGM6rYYE",
 749 |         "outputId": "4ee5c2bb-98bf-4cac-bc1a-2e7ac118ebb7"
 750 |       },
 751 |       "execution_count": null,
 752 |       "outputs": [
 753 |         {
 754 |           "output_type": "stream",
 755 |           "name": "stdout",
 756 |           "text": [
 757 |             "11738\n",
 758 |             "11738\n"
 759 |           ]
 760 |         }
 761 |       ]
 762 |     },
 763 |     {
 764 |       "cell_type": "code",
 765 |       "source": [
 766 |         "trainX = np.reshape(trainX, (trainX.shape[0], 1, trainX.shape[1]))\n",
 767 |         "testX = np.reshape(testX, (testX.shape[0], 1, testX.shape[1]))\n",
 768 |         "\n",
 769 |         "print(trainX.shape)\n",
 770 |         "print(testX.shape)"
 771 |       ],
 772 |       "metadata": {
 773 |         "colab": {
 774 |           "base_uri": "https://localhost:8080/"
 775 |         },
 776 |         "id": "z0z5HcUMraky",
 777 |         "outputId": "e0a9cb61-9d5e-4cdd-cf45-40ab57ee679e"
 778 |       },
 779 |       "execution_count": null,
 780 |       "outputs": [
 781 |         {
 782 |           "output_type": "stream",
 783 |           "name": "stdout",
 784 |           "text": [
 785 |             "(11738, 1, 1)\n",
 786 |             "(11738, 1, 1)\n"
 787 |           ]
 788 |         }
 789 |       ]
 790 |     },
 791 |     {
 792 |       "cell_type": "markdown",
 793 |       "source": [
 794 |         "TRAINING WITH LSTM"
 795 |       ],
 796 |       "metadata": {
 797 |         "id": "xPRQx9bQri3m"
 798 |       }
 799 |     },
 800 |     {
 801 |       "cell_type": "code",
 802 |       "source": [
 803 |         "\n",
 804 |         "model = Sequential()\n",
 805 |         "model.add(LSTM(64, input_shape=(trainX.shape[1], trainX.shape[2])))\n",
 806 |         "\n",
 807 |         "model.add(Dense(1))\n",
 808 |         "\n",
 809 |         "model.compile(loss='mae', optimizer='adam')"
 810 |       ],
 811 |       "metadata": {
 812 |         "id": "UXZ0CBjureyY"
 813 |       },
 814 |       "execution_count": null,
 815 |       "outputs": []
 816 |     },
 817 |     {
 818 |       "cell_type": "code",
 819 |       "source": [
 820 |         "history = model.fit(trainX, trainY, epochs=100, batch_size=20, validation_data=(testX, testY), verbose=1, shuffle = False)"
 821 |       ],
 822 |       "metadata": {
 823 |         "colab": {
 824 |           "base_uri": "https://localhost:8080/"
 825 |         },
 826 |         "id": "eSINwJ3nrddc",
 827 |         "outputId": "c746e7dd-4ff8-40d4-faa2-45620c4e844b"
 828 |       },
 829 |       "execution_count": null,
 830 |       "outputs": [
 831 |         {
 832 |           "output_type": "stream",
 833 |           "name": "stdout",
 834 |           "text": [
 835 |             "Epoch 1/100\n",
 836 |             "587/587 [==============================] - 9s 7ms/step - loss: 0.0613 - val_loss: 0.0087\n",
 837 |             "Epoch 2/100\n",
 838 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0041 - val_loss: 0.0087\n",
 839 |             "Epoch 3/100\n",
 840 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0041 - val_loss: 0.0084\n",
 841 |             "Epoch 4/100\n",
 842 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0036 - val_loss: 0.0084\n",
 843 |             "Epoch 5/100\n",
 844 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0039 - val_loss: 0.0083\n",
 845 |             "Epoch 6/100\n",
 846 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0037 - val_loss: 0.0083\n",
 847 |             "Epoch 7/100\n",
 848 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0044 - val_loss: 0.0082\n",
 849 |             "Epoch 8/100\n",
 850 |             "587/587 [==============================] - 7s 13ms/step - loss: 0.0039 - val_loss: 0.0084\n",
 851 |             "Epoch 9/100\n",
 852 |             "587/587 [==============================] - 6s 11ms/step - loss: 0.0037 - val_loss: 0.0084\n",
 853 |             "Epoch 10/100\n",
 854 |             "587/587 [==============================] - 5s 9ms/step - loss: 0.0043 - val_loss: 0.0081\n",
 855 |             "Epoch 11/100\n",
 856 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0037 - val_loss: 0.0085\n",
 857 |             "Epoch 12/100\n",
 858 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0034 - val_loss: 0.0081\n",
 859 |             "Epoch 13/100\n",
 860 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0037 - val_loss: 0.0084\n",
 861 |             "Epoch 14/100\n",
 862 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0040 - val_loss: 0.0081\n",
 863 |             "Epoch 15/100\n",
 864 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0037 - val_loss: 0.0081\n",
 865 |             "Epoch 16/100\n",
 866 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0035 - val_loss: 0.0084\n",
 867 |             "Epoch 17/100\n",
 868 |             "587/587 [==============================] - 6s 9ms/step - loss: 0.0034 - val_loss: 0.0080\n",
 869 |             "Epoch 18/100\n",
 870 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0040 - val_loss: 0.0080\n",
 871 |             "Epoch 19/100\n",
 872 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0040 - val_loss: 0.0079\n",
 873 |             "Epoch 20/100\n",
 874 |             "587/587 [==============================] - 5s 9ms/step - loss: 0.0040 - val_loss: 0.0079\n",
 875 |             "Epoch 21/100\n",
 876 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0034 - val_loss: 0.0079\n",
 877 |             "Epoch 22/100\n",
 878 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0036 - val_loss: 0.0081\n",
 879 |             "Epoch 23/100\n",
 880 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0035 - val_loss: 0.0079\n",
 881 |             "Epoch 24/100\n",
 882 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0035 - val_loss: 0.0079\n",
 883 |             "Epoch 25/100\n",
 884 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0033 - val_loss: 0.0078\n",
 885 |             "Epoch 26/100\n",
 886 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0035 - val_loss: 0.0083\n",
 887 |             "Epoch 27/100\n",
 888 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0032 - val_loss: 0.0078\n",
 889 |             "Epoch 28/100\n",
 890 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0036 - val_loss: 0.0081\n",
 891 |             "Epoch 29/100\n",
 892 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0035 - val_loss: 0.0078\n",
 893 |             "Epoch 30/100\n",
 894 |             "587/587 [==============================] - 4s 8ms/step - loss: 0.0034 - val_loss: 0.0078\n",
 895 |             "Epoch 31/100\n",
 896 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0036 - val_loss: 0.0080\n",
 897 |             "Epoch 32/100\n",
 898 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0039 - val_loss: 0.0077\n",
 899 |             "Epoch 33/100\n",
 900 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0034 - val_loss: 0.0077\n",
 901 |             "Epoch 34/100\n",
 902 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0040 - val_loss: 0.0077\n",
 903 |             "Epoch 35/100\n",
 904 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0033 - val_loss: 0.0078\n",
 905 |             "Epoch 36/100\n",
 906 |             "587/587 [==============================] - 6s 10ms/step - loss: 0.0033 - val_loss: 0.0077\n",
 907 |             "Epoch 37/100\n",
 908 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0033 - val_loss: 0.0079\n",
 909 |             "Epoch 38/100\n",
 910 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0033 - val_loss: 0.0079\n",
 911 |             "Epoch 39/100\n",
 912 |             "587/587 [==============================] - 5s 9ms/step - loss: 0.0034 - val_loss: 0.0079\n",
 913 |             "Epoch 40/100\n",
 914 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0033 - val_loss: 0.0076\n",
 915 |             "Epoch 41/100\n",
 916 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0038 - val_loss: 0.0077\n",
 917 |             "Epoch 42/100\n",
 918 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0038 - val_loss: 0.0076\n",
 919 |             "Epoch 43/100\n",
 920 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0033 - val_loss: 0.0076\n",
 921 |             "Epoch 44/100\n",
 922 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0035 - val_loss: 0.0076\n",
 923 |             "Epoch 45/100\n",
 924 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0035 - val_loss: 0.0076\n",
 925 |             "Epoch 46/100\n",
 926 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0031 - val_loss: 0.0079\n",
 927 |             "Epoch 47/100\n",
 928 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0031 - val_loss: 0.0076\n",
 929 |             "Epoch 48/100\n",
 930 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0039 - val_loss: 0.0076\n",
 931 |             "Epoch 49/100\n",
 932 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0036 - val_loss: 0.0075\n",
 933 |             "Epoch 50/100\n",
 934 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0036 - val_loss: 0.0076\n",
 935 |             "Epoch 51/100\n",
 936 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0031 - val_loss: 0.0075\n",
 937 |             "Epoch 52/100\n",
 938 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0032 - val_loss: 0.0081\n",
 939 |             "Epoch 53/100\n",
 940 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0029 - val_loss: 0.0076\n",
 941 |             "Epoch 54/100\n",
 942 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0038 - val_loss: 0.0075\n",
 943 |             "Epoch 55/100\n",
 944 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0034 - val_loss: 0.0078\n",
 945 |             "Epoch 56/100\n",
 946 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0033 - val_loss: 0.0076\n",
 947 |             "Epoch 57/100\n",
 948 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0031 - val_loss: 0.0075\n",
 949 |             "Epoch 58/100\n",
 950 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0031 - val_loss: 0.0078\n",
 951 |             "Epoch 59/100\n",
 952 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0032 - val_loss: 0.0077\n",
 953 |             "Epoch 60/100\n",
 954 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0036 - val_loss: 0.0076\n",
 955 |             "Epoch 61/100\n",
 956 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0037 - val_loss: 0.0081\n",
 957 |             "Epoch 62/100\n",
 958 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0033 - val_loss: 0.0075\n",
 959 |             "Epoch 63/100\n",
 960 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0040 - val_loss: 0.0075\n",
 961 |             "Epoch 64/100\n",
 962 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0030 - val_loss: 0.0075\n",
 963 |             "Epoch 65/100\n",
 964 |             "587/587 [==============================] - 5s 9ms/step - loss: 0.0031 - val_loss: 0.0074\n",
 965 |             "Epoch 66/100\n",
 966 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0035 - val_loss: 0.0077\n",
 967 |             "Epoch 67/100\n",
 968 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0030 - val_loss: 0.0075\n",
 969 |             "Epoch 68/100\n",
 970 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0035 - val_loss: 0.0077\n",
 971 |             "Epoch 69/100\n",
 972 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0034 - val_loss: 0.0074\n",
 973 |             "Epoch 70/100\n",
 974 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0035 - val_loss: 0.0074\n",
 975 |             "Epoch 71/100\n",
 976 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0034 - val_loss: 0.0074\n",
 977 |             "Epoch 72/100\n",
 978 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0031 - val_loss: 0.0076\n",
 979 |             "Epoch 73/100\n",
 980 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0034 - val_loss: 0.0076\n",
 981 |             "Epoch 74/100\n",
 982 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0031 - val_loss: 0.0074\n",
 983 |             "Epoch 75/100\n",
 984 |             "587/587 [==============================] - 6s 10ms/step - loss: 0.0033 - val_loss: 0.0074\n",
 985 |             "Epoch 76/100\n",
 986 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0032 - val_loss: 0.0074\n",
 987 |             "Epoch 77/100\n",
 988 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0031 - val_loss: 0.0074\n",
 989 |             "Epoch 78/100\n",
 990 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0036 - val_loss: 0.0075\n",
 991 |             "Epoch 79/100\n",
 992 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0036 - val_loss: 0.0074\n",
 993 |             "Epoch 80/100\n",
 994 |             "587/587 [==============================] - 5s 9ms/step - loss: 0.0035 - val_loss: 0.0075\n",
 995 |             "Epoch 81/100\n",
 996 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0031 - val_loss: 0.0077\n",
 997 |             "Epoch 82/100\n",
 998 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0034 - val_loss: 0.0074\n",
 999 |             "Epoch 83/100\n",
1000 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0033 - val_loss: 0.0076\n",
1001 |             "Epoch 84/100\n",
1002 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0034 - val_loss: 0.0074\n",
1003 |             "Epoch 85/100\n",
1004 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0034 - val_loss: 0.0074\n",
1005 |             "Epoch 86/100\n",
1006 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0033 - val_loss: 0.0076\n",
1007 |             "Epoch 87/100\n",
1008 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0029 - val_loss: 0.0074\n",
1009 |             "Epoch 88/100\n",
1010 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0035 - val_loss: 0.0075\n",
1011 |             "Epoch 89/100\n",
1012 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0032 - val_loss: 0.0074\n",
1013 |             "Epoch 90/100\n",
1014 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0032 - val_loss: 0.0074\n",
1015 |             "Epoch 91/100\n",
1016 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0031 - val_loss: 0.0074\n",
1017 |             "Epoch 92/100\n",
1018 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0036 - val_loss: 0.0074\n",
1019 |             "Epoch 93/100\n",
1020 |             "587/587 [==============================] - 4s 8ms/step - loss: 0.0029 - val_loss: 0.0074\n",
1021 |             "Epoch 94/100\n",
1022 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0031 - val_loss: 0.0075\n",
1023 |             "Epoch 95/100\n",
1024 |             "587/587 [==============================] - 4s 6ms/step - loss: 0.0030 - val_loss: 0.0078\n",
1025 |             "Epoch 96/100\n",
1026 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0031 - val_loss: 0.0073\n",
1027 |             "Epoch 97/100\n",
1028 |             "587/587 [==============================] - 4s 7ms/step - loss: 0.0029 - val_loss: 0.0077\n",
1029 |             "Epoch 98/100\n",
1030 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0034 - val_loss: 0.0073\n",
1031 |             "Epoch 99/100\n",
1032 |             "587/587 [==============================] - 3s 6ms/step - loss: 0.0033 - val_loss: 0.0076\n",
1033 |             "Epoch 100/100\n",
1034 |             "587/587 [==============================] - 5s 8ms/step - loss: 0.0030 - val_loss: 0.0076\n"
1035 |           ]
1036 |         }
1037 |       ]
1038 |     },
1039 |     {
1040 |       "cell_type": "code",
1041 |       "source": [
1042 |         "# save model architecture\n",
1043 |         "model_json = model.to_json()\n",
1044 |         "open('50%/model/B48_model.json', 'w').write(model_json)\n",
1045 |         "\n",
1046 |         "# save model's learned weights\n",
1047 |         "model.save_weights('50%/model/B48_weights.h5', overwrite=True)"
1048 |       ],
1049 |       "metadata": {
1050 |         "id": "LvMED7V1r3Y6"
1051 |       },
1052 |       "execution_count": null,
1053 |       "outputs": []
1054 |     },
1055 |     {
1056 |       "cell_type": "code",
1057 |       "source": [
1058 |         "# Load trained model\n",
1059 |         "from keras.models import model_from_json \n",
1060 |         "\n",
1061 |         "json_file = open(\"70%/model/B48_model.json\", \"r\")\n",
1062 |         "loaded_model_json = json_file.read() \n",
1063 |         "json_file.close()\n",
1064 |         "loaded_model = model_from_json(loaded_model_json)\n",
1065 |         " \n",
1066 |         "# model weight load \n",
1067 |         "loaded_model.load_weights(\"70%/model/B48_weights.h5\")\n",
1068 |         "print(\"Loaded model from disk\")"
1069 |       ],
1070 |       "metadata": {
1071 |         "colab": {
1072 |           "base_uri": "https://localhost:8080/"
1073 |         },
1074 |         "id": "S_3Zt4BAvIp_",
1075 |         "outputId": "5ff1abf3-d5d0-446f-d081-ac95463f4f75"
1076 |       },
1077 |       "execution_count": null,
1078 |       "outputs": [
1079 |         {
1080 |           "output_type": "stream",
1081 |           "name": "stdout",
1082 |           "text": [
1083 |             "Loaded model from disk\n"
1084 |           ]
1085 |         }
1086 |       ]
1087 |     },
1088 |     {
1089 |       "cell_type": "code",
1090 |       "source": [
1091 |         "plt.plot(history.history['loss'], label='train')\n",
1092 |         "plt.plot(history.history['val_loss'], label='test')\n",
1093 |         "plt.axis([-5, 100, 0, 0.07])\n",
1094 |         "plt.legend() \n",
1095 |         "plt.show()"
1096 |       ],
1097 |       "metadata": {
1098 |         "colab": {
1099 |           "base_uri": "https://localhost:8080/",
1100 |           "height": 269
1101 |         },
1102 |         "id": "9nAYe5N4vJ0j",
1103 |         "outputId": "4273ab9d-444c-48c4-df0e-d9aa976ad4aa"
1104 |       },
1105 |       "execution_count": null,
1106 |       "outputs": [
1107 |         {
1108 |           "output_type": "display_data",
1109 |           "data": {
1110 |             "text/plain": [
1111 |               "<Figure size 432x288 with 1 Axes>"
1112 |             ],
1113 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAD8CAYAAACVZ8iyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAAA1bklEQVR4nO3dfXhU5Z3/8fd5mElmkplMAskkQAgCUZAEfKzg09bQmEpUQEJ/7bZusaXt9qq1rBWr2/2xlq5u+ystYltbkdXWbbttscW2xi7FREVtrKiFKKIYIZCQZALJJJlkJvNw5vz+GBgNEZLJg4GZ7+u6vC6HOefM/Z0zOZ9z32fOPYppmiZCCCFSmjrRDRBCCDHxJAyEEEJIGAghhJAwEEIIgYSBEEIIJAyEEEIwzDDYuXMnFRUVlJeXs3nz5kHPh0Ih1qxZQ3l5OStXrqS5uRmAP/7xjyxdujT+35w5c9i3b9/YViCEEGLUlKHuMzAMg4qKCh599FHcbjdVVVX84Ac/YPbs2fFlfvnLX/L222+zfv16qqur2bFjB/fff/+A7bz99tt85Stf4emnnx6XQoQQQozckD2D+vp6ioqKKCwsxGq1UllZSU1NzYBlamtrWb58OQAVFRXU1dVxcsZUV1dTWVk5hk0XQggxVvShFvB4POTn58cfu91u6uvrBy1TUFAQ26Cu43A48Hq95OTkxJd56qmnePDBB4dskGmajOaeaEVhVOuf7aR+qT9V60/l2gFUVRnV+kOGwVjYs2cPNpuNc889d8hlI5EoXV3+Eb+Wy2Uf1fpnO6lf6k/V+lO5doDcXMeo1h9ymMjtdtPW1hZ/7PF4cLvdg5ZpbW0FIBKJ4PP5yM7Ojj8vQ0RCCHFmGzIMSktLaWxspKmpiVAoRHV1NWVlZQOWKSsrY9u2bQBs376dhQsXoiixLks0GuXPf/6zhIEQQpzBhhwm0nWddevWsXr1agzDYMWKFRQXF7Np0yZKSkpYvHgxVVVVrF27lvLycrKysti4cWN8/V27dlFQUEBhYeG4FiKEEGLkhvxq6YctHDbkmsEoSP1Sf6rWn5lpobHxMJFIaKKbMq503Up2di6aNvBcfrTXDD6UC8hCCDHejhw5Qnq6nYyM/PgwdbIxTZO+vh683qNMnlwwptuW6SiEEEkhGAySkeFM2iAAUBSFjAznuPR+JAyEEEnBNM2kDoITxqtGCQMhhBASBkIIMRZ8Ph+///3WhNe7447b8Pl849CixEgYCCHEGOjt9bFt2+AwiEQip11vw4YHcDhG902gsSDfJhJCiDHw05/+kCNHjrBq1T+i6zpWqxWHw8GhQ4f49a9/z913fx2Px0MoFGLlyk+ydOlNAFRV3cCWLf9NIODnjjtuY/78C3j99Xpyc3P5zne+T1pa+ofSfgkDIUTSqd7r4Y9vtA29YAJuLMmncp77lM//8z9/lQMH3uVnP/sVr732CnfeuYbHHvsNU6ZMBeDuu9fhdGYRDPazevU/8dGPlpGV5RqwjebmJu65516+8Y1/4//+37t49tlaKiqWjGkdpyJhIIQQ42Du3HnxIADYuvXX7Nz5LADt7R6ampoGhUFBwRSKi88D4Lzz5tDa2vJhNVfCQAiRfCrnuU97Fv9hsNls8f9/7bVXeOWVl3nooUdJT0/n1lu/SCgUHLSOxWKJ/7+qahjG4GXGi1xAFkKIMWC32/H7P3gqkL6+XhwOJ+np6Rw61Mibb77xIbduaNIzEEKIMZCV5aK0dAE33/wJ0tLSB/y412WXXc4TT/yeT3+6iunTizj//JIJbOkHk4nqkozUL/Wnav0ez2Hc7ukT3YwPRVvbIfLziwb827j/uI0QQojkJ2EghBBCwkAIIYSEgRBCCCQMhBBCIGEghBACCQMhhBgTI53CGuC3v/0V/f39Y9yixCRVGPQGI+xp7proZgghUtCpprAejt/+9n8mPAyS6g7kP77Rxo+eP8jOr16BriVVzgkhznDvn8L60ksvIzs7m9rapwmHQ1x99TV8/vNfIhAIsG7dXbS3txONGqxatZrOzk6OHTvKbbd9iawsFz/84UMT0v5hhcHOnTu59957iUajrFy5ki9+8YsDng+FQtx5553s3bsXl8vFxo0bmTZtGgBvvfUW//7v/05vby+qqvL444+TlpY29pUAYcMkbJhEoia6Ni4vIYQ4C6S99Tjp+349ptvsn/tJgnOqTvn8+6ewfvnll3jmmRoefvjnmKbJXXfdzu7dr9HV5WXy5Fy+971NAPT29pKZmclvfvNLHnjgIVwu15i2ORFDhoFhGKxfv55HH30Ut9tNVVUVZWVlzJ49O77M1q1bcTqd7Nixg+rqajZs2MD9999PJBJh7dq1fO9732POnDl4vV50ffw6I7oa+6HoSPSMmmFDCJFiXn75JXbteolbbvk0AIGAn+bmw8yffyE/+tH9PPjgA1xxxVUsWHDhBLf0PUMemevr6ykqKqKwsBCAyspKampqBoRBbW0tt956KwAVFRWsX78e0zR58cUXOe+885gzZw4A2dnZ41FDnISBEAIgOKfqtGfx4800TT7zmVUsW7Zi0HOPPPIL6upe5OGHf8LFF1/KLbd8YQJaONiQYeDxeMjPz48/drvd1NfXD1qmoKAgtkFdx+Fw4PV6OXjwIIqi8PnPf57Ozk6WLFnCF75w+sI1TcHlso+kFjIzY8NPGZnpuBzjMxR1ptM0dcTvXzKQ+lO3/vZ2BW0CrxU6HJkEAn40TWXRosvZvPknXHddJXa7nfb2dnRdxzAMnE4nS5Zcj9Pp5E9/egJNU7HbMwgGA2haztAvBCjKyI+TpzKuF5ANw+DVV1/l8ccfx2azsWrVKkpKSli0aNFp1jFHPOtiuD8MQGeXH4thjGgbZ7tUnrUSpP5Urt80TQwjOmGvn5nppKRkAf/4j1UsXHgFH/tYBV/4wmcBsNnsrFv3bZqbm3jwwU0oioqu69xxx10YRpQbb1zGmjVfYfLk3GFdQDbNwcfJ0c5aOmQYuN1u2tre+y1Rj8eD2+0etExrayv5+flEIhF8Ph/Z2dnk5+dz6aWXxuf1vvrqq9m7d+9pw2A0tPgw0cR9IIQQqeuee+4d8PgTn/jUgMdTp07jsssGH/+qqj5JVdUnx7VtQxmyT1VaWkpjYyNNTU2EQiGqq6spKysbsExZWRnbtm0DYPv27SxcuBBFUbjyyivZv38/gUCASCTCrl27BlxrGGu6djwMDLlmIIQQiRiyZ6DrOuvWrWP16tUYhsGKFSsoLi5m06ZNlJSUsHjxYqqqqli7di3l5eVkZWWxceNGALKysli1ahVVVVUoisLVV1/NRz/60fErRo1lm3Fm/V6PEEKc8ZLql85q3znGN/74Jr+8+SLOzcsc45adHVJ5zBik/lSuv729idzcaSiKMtFNGVemaeLxHJZfOjudE18tlZ6BEKknLS2Nvr4ezrDz2zFlmiZ9fT3ounXMt51U01HELyDLNQMhUs7UqVNpbDxMb2/XRDdlXOm6lezs3LHf7phvcQLJTWdCpC5dtzB5csFEN+OslZzDRBIGQgiRkKQMA7nPQAghEpOUYTCBNyEKIcRZKanCQO5AFkKIkUmqMDhx05lcQBZCiMQkVRhocgFZCCFGJKnCQL5aKoQQI5NcYaBJGAghxEgkVRhoioSBEEKMRFKFgUxhLYQQI5NUYXCiZyAT1QkhRGKSKgze6xnIfQZCCJGI5AqDEz9uI9cMhBAiIUkVBpp8tVQIIUYkucLg+A8cSRgIIURikioMFEVBVxUZJhJCiAQlVRhA7CKy9AyEECIxyRcGqio9AyGESFAShoH0DIQQIlHDCoOdO3dSUVFBeXk5mzdvHvR8KBRizZo1lJeXs3LlSpqbmwFobm5m/vz5LF26lKVLl7Ju3bqxbf0HiA0TyX0GQgiRCH2oBQzDYP369Tz66KO43W6qqqooKytj9uzZ8WW2bt2K0+lkx44dVFdXs2HDBu6//34Apk+fzh/+8IdxK+BkmlxAFkKIhA3ZM6ivr6eoqIjCwkKsViuVlZXU1NQMWKa2tpbly5cDUFFRQV1dHeYETQlhUVUZJhJCiAQN2TPweDzk5+fHH7vdburr6wctU1BQENugruNwOPB6vUBsqGjZsmVkZmayZs0aLrnkktO+nqYpuFz2hAs5QdcUVE0b1TbOZpqmpmztIPWncv2pXPtYGDIMRiMvL49nnnmG7Oxs3njjDb7yla9QXV1NZmbmKdcxDJOuLv+IX1NTFQLB8Ki2cTZzuewpWztI/alcfyrXDpCb6xjV+kMOE7ndbtra2uKPPR4Pbrd70DKtra0ARCIRfD4f2dnZWK1WsrOzASgpKWH69OkcPHhwVA0eikW+WiqEEAkbMgxKS0tpbGykqamJUChEdXU1ZWVlA5YpKytj27ZtAGzfvp2FCxeiKAqdnZ0YhgFAU1MTjY2NFBYWjkMZ79HkpjMhhEjYkMNEuq6zbt06Vq9ejWEYrFixguLiYjZt2kRJSQmLFy+mqqqKtWvXUl5eTlZWFhs3bgRg165dPPDAA+i6jqqqfOtb38Llco1vQaoiP24jhBAJUsyJ+trPKYTDxqjG/f55az0K8JOV88euUWeRVB83lfpTt/5Urh0+hGsGZxtdUzHkx22EECIhyRcGqkJEskAIIRKSpGEgaSCEEIlIvjDQ5A5kIYRIVNKFgcxNJIQQiUu6MLDIFNZCCJGwpAsDXZM7kIUQIlFJFwaa9AyEECJhSRcGFpmOQgghEpZ0YSAXkIUQInFJFwa6qsp9BkIIkaCkCwOLJj0DIYRIVNKFgVxAFkKIxCVdGOiqKlNYCyFEgpIwDBRMIHpmzcwthBBntOQLA00BkN6BEEIkIAnDIFaSIT0DIYQYtuQLA1V6BkIIkajkDQO510AIIYYt6cJAOx4Gcq+BEEIMX9KFgeX4NQO510AIIYYv6cLgvWEiCQMhhBiuYYXBzp07qaiooLy8nM2bNw96PhQKsWbNGsrLy1m5ciXNzc0Dnm9paeHCCy/kv/7rv8am1aehSRgIIUTChgwDwzBYv349W7Zsobq6mieffJKGhoYBy2zduhWn08mOHTtYtWoVGzZsGPD8d77zHa666qqxbfkpyDCREEIkbsgwqK+vp6ioiMLCQqxWK5WVldTU1AxYpra2luXLlwNQUVFBXV0d5vHv+T/99NNMnTqV4uLicWj+YHIBWQghEjdkGHg8HvLz8+OP3W43Ho9n0DIFBQUA6LqOw+HA6/XS19fHww8/zK233jrGzT61+B3IEgZCCDFs+nhu/Ec/+hGf/exnycjIGPY6mqbgctlH/JrWjgAAdrt1VNs5W2mampJ1nyD1p279qVz7WBgyDNxuN21tbfHHHo8Ht9s9aJnW1lby8/OJRCL4fD6ys7PZs2cP27dvZ8OGDfT09KCqKmlpaXzmM5855esZhklXl3/EBZ3o6nh7AnR1pY14O2crl8s+qvfvbCf1p279qVw7QG6uY1TrDxkGpaWlNDY20tTUhNvtprq6mu9///sDlikrK2Pbtm1ceOGFbN++nYULF6IoCr/61a/iy/zwhz/EbrefNgjGwolhIrlmIIQQwzdkGOi6zrp161i9ejWGYbBixQqKi4vZtGkTJSUlLF68mKqqKtauXUt5eTlZWVls3Ljxw2j7B7dXvloqhBAJU0zzzJreMxw2RtXVO9wbZsVDdfxg2TyumjVpDFt2dkj1rrLUn7r1p3LtMPphoqS7A1m+WiqEEIlLujCwyFdLhRAiYUkXBrr0DIQQImFJFwaaTEchhBAJS7owsMiP2wghRMKSLgzkArIQQiQu6cJAl2EiIYRIWNKFgUVuOhNCiIQlXRjIMJEQQiQu6cJAhomEECJxyRcGMkwkhBAJS7ow0FQFBQkDIYRIRNKFAcSmsZZrBkIIMXxJGQaaohAxJAyEEGK4kjIMdE2RO5CFECIBSRkGmiLDREIIkYikDANdU+UCshBCJCA5w0CVnoEQQiQiKcNAUxXpGQghRAKSMgx0CQMhhEhIUoaBJsNEQgiRkKQMA+kZCCFEYoYVBjt37qSiooLy8nI2b9486PlQKMSaNWsoLy9n5cqVNDc3A1BfX8/SpUtZunQpN954Izt27Bjb1p+CXEAWQojE6EMtYBgG69ev59FHH8XtdlNVVUVZWRmzZ8+OL7N161acTic7duygurqaDRs2cP/991NcXMzvfvc7dF2nvb2dpUuXcs0116DrQ77s6IpS5aYzIYRIxJA9g/r6eoqKiigsLMRqtVJZWUlNTc2AZWpra1m+fDkAFRUV1NXVYZomNpstfuAPBoMoijIOJQwmw0RCCJGYIcPA4/GQn58ff+x2u/F4PIOWKSgoAEDXdRwOB16vF4A9e/ZQWVnJjTfeyLe+9a1x7xWAXEAWQohEjfuRecGCBVRXV/Puu+/yjW98g6uvvpq0tLRTLq9pCi6XfcSvp2kq6Wk6vv7IqLZzttI0NSXrPkHqT936U7n2sTBkGLjdbtra2uKPPR4Pbrd70DKtra3k5+cTiUTw+XxkZ2cPWGbWrFnY7Xb2799PaWnpKV/PMEy6uvyJ1hHnctkxDZNQ2BjVds5WLpc9Jes+QepP3fpTuXaA3FzHqNYfcpiotLSUxsZGmpqaCIVCVFdXU1ZWNmCZsrIytm3bBsD27dtZuHAhiqLQ1NREJBIB4MiRIxw4cICpU6eOqsHDIdcMhBAiMUP2DHRdZ926daxevRrDMFixYgXFxcVs2rSJkpISFi9eTFVVFWvXrqW8vJysrCw2btwIwKuvvsrDDz+Mruuoqso999xDTk7O+Belye8ZCCFEIhTTNM+oo2Z4lMM7Lpedr/ziVd5q7+V3n7t0DFt2dkj1rrLUn7r1p3Lt8CEME52NYj0Duc9ACCGGKynDQFPkmoEQQiQiKcMg9rOXEgZCCDFcyRkGqio3nQkhRAKSMgzkx22EECIxSRkGcp+BEEIkJinDQOYmEkKIxCRlGJzoGZxht1AIIcQZK2nDAEA6B0IIMTxJGQba8TCQ6wZCCDE8SRkGejwM5C5kIYQYjqQMgxM9A7mILIQQw5OUYaCrsbJkmEgIIYYnOcNAk56BEEIkIjnDQJELyEIIkYjkDIPjPQP5gRshhBiepAwDTZFhIiGESERShkG8ZyBhIIQQw5KcYSBfLRVCiIQkZRhoctOZEEIkJCnDQJfpKIQQIiFJGQYyN5EQQiQmKcNA7kAWQojEDCsMdu7cSUVFBeXl5WzevHnQ86FQiDVr1lBeXs7KlStpbm4G4MUXX+Smm27ihhtu4KabbqKurm5sW38KMjeREEIkZsgwMAyD9evXs2XLFqqrq3nyySdpaGgYsMzWrVtxOp3s2LGDVatWsWHDBgCys7P5yU9+wp/+9Ce+853vcOedd45PFSeRawZCCJGYIcOgvr6eoqIiCgsLsVqtVFZWUlNTM2CZ2tpali9fDkBFRQV1dXWYpsn555+P2+0GoLi4mGAwSCgUGocyBpIwEEKIxOhDLeDxeMjPz48/drvd1NfXD1qmoKAgtkFdx+Fw4PV6ycnJiS+zfft2zj//fKxW62lfT9MUXC57QkUMXF8lO8sGQHq6ZVTbOhtpmppyNb+f1J+69ady7WNhyDAYC++88w4bNmzgkUceGXJZwzDp6vKP+LVcLjv+viAA3b39o9rW2cjlsqdcze8n9adu/alcO0BurmNU6w85TOR2u2lra4s/9ng88aGf9y/T2toKQCQSwefzkZ2dDUBbWxu33nor3/3ud5k+ffqoGjtccgFZCCESM2QYlJaW0tjYSFNTE6FQiOrqasrKygYsU1ZWxrZt24DYcNDChQtRFIWenh6++MUv8vWvf52LL754fCr4APFrBjJrqRBCDMuQYaDrOuvWrWP16tUsWbKE6667juLiYjZt2hS/kFxVVUVXVxfl5eU8+uij3HHHHQD84he/4PDhw/z4xz9m6dKlLF26lI6OjvGtCNA1uc9ACCESoZimeUYdMcNhY9TXDBpbuin/SR13XDOL/3PR1DFs3Zkv1cdNpf7UrT+Va4cP4ZrB2UimsBZCiMQkZRjIBWQhhEhMUoaB3HQmhBCJScowkJ6BEEIkJinDQFUUVEV+3EYIIYYrKcMAYkNFMkwkhBDDk7RhoEkYCCHEsCVtGOiqKtcMhBBimJI4DKRnIIQQw5W0YSDDREIIMXxJGwbSMxBCiOFL2jDQVEWuGQghxDAlbRjoqiJTWAshxDAlbxhoCsaZNSGrEEKcsZI2DDRFIWLIHchCCDEcSRsGuqbKBWQhhBimpA0DTZELyEIIMVxJGwa6Jl8tFUKI4UraMJCvlgohxPAlbRjITWdCCDF8EgZCCCGSOwxkmEgIIYZnWGGwc+dOKioqKC8vZ/PmzYOeD4VCrFmzhvLyclauXElzczMAXq+Xm2++mQsvvJD169ePbcuHEOsZyH0GQggxHEOGgWEYrF+/ni1btlBdXc2TTz5JQ0PDgGW2bt2K0+lkx44drFq1ig0bNgCQlpbG1772Ne68887xaf1pyAVkIYQYviHDoL6+nqKiIgoLC7FarVRWVlJTUzNgmdraWpYvXw5ARUUFdXV1mKaJ3W7nkksuIS0tbXxafxpyzUAIIYZPH2oBj8dDfn5+/LHb7aa+vn7QMgUFBbEN6joOhwOv10tOTk7CDdI0BZfLnvB6762v4nLZsdusRE1Gta2z0Yn6U5XUn7r1p3LtY2HIMPiwGYZJV5d/xOu7XHa6uvwYYYOwER3Vts5GJ+pPVVJ/6tafyrUD5OY6RrX+kMNEbrebtra2+GOPx4Pb7R60TGtrKwCRSASfz0d2dvaoGjZi/k7UvjYmGUdxRr0T0wYhhDjLDBkGpaWlNDY20tTURCgUorq6mrKysgHLlJWVsW3bNgC2b9/OwoULURRlfFp8GtZ3n8KycTaTfnYJ33z3EzzLF8n+n8XYX/4+2rE3UQKdKP1dKMEeiBqJbdyMxtaJRmL/DXreRD/6OpamnYlv+4RoBIzQyNYVQohRUExz6En/n3vuOe677z4Mw2DFihV8+ctfZtOmTZSUlLB48WKCwSBr165l3759ZGVlsXHjRgoLC4FYUPT29hIOh3E4HDzyyCPMnj37lK8VDhsj7uopwR5crTsI9AV4en8n+5uO8C/TGrC0vozCwDJNFMw0J9H0HMy0LMw0J6Y1k6jFgWnNxLRmgqKieRvQO95G6z6A8r4QiGTNIDz1CsLTLkftbSP9rd+id74NgJE5lUDJzYSKyrC0vYr18LPont0YOecSnrqI0NRFRDPcoFowVR2LZzdpB57CevAvEI0QnH0D/ed/ioj7IkgwVE90lZV+L2q/FyPrnGFtQ2+vx7b7IVB1/Bd/FSP71PvoTJbqQwWpXH8q1w6jHyYaVhh8mEYTBvDeB+LBFw7y2MtNvHT71Sh97VgPP4sS7kU5foavhHtR+ztRAl7UUDdKqPf4fz0ooT6UkA8FE8NZRCTnPIycYkzdFnsR00A/uhdLSx1qyBdrt/si+uesJJqeje2N/8Z65MV4mwxHIeH8i9A730bveOsD2x1NyyI0oxxT1Uh/508oET9GhhtTt4OqgaoTTc8has8lapsEWhqmqoPyvs6dGcUWbCPatAu9uzH22nY34cIrCRdciqlaUUwDTANTS4vXk77vN6Qdfoao1RkLPKOf4HlVBM8px9L6CpbmF9G6DxLJv4RQ4dWEp11BNM0Va5eixN6L1pfRW1/BtGYSmvExQjMWE82ccvqdFTVQgl2YWjpY7LHQMqOovha0rndR+9pQ+7tQgl0o4b7jtaqYFhvhKQsJT/kIaAO/qTaiA4JponUfRG99hWjWdML5l8ZqGwkzSlrDn7C/+kNM3U6g5GaCs6+HE5+dD3gPhnytcAAl4sdMzx64v4/Tj75O+huPYT1UC3OX4l1wG2a6a2TtP4uddWFgmgmf7J2OhMFJTnwgHnqxkS0vHebl268a2ZCVGY0N22jWUy8TjaAf24tpdWC4Zg54Suvcj6XtNcIFl8aeO94GJdCJpe0VlGA3ihEEI4Thmkl46uXx11JCvaQ1/AlLy0uxNphRFCMUCy//UdRAB4oRAtOI91ZMRQUUyMgllLuAcP5FmNYsLC11WJueR+3vPHUZ6Tn4L/gi/SX/BNEw9tcexPb6z1CMIKaWRjj/Yoysc2IHfO87H/x2qTqR3FLUQCdazyEAjMwCUK2YmgUU7fh7cPyAH+hA6e+MhTNgqlai6S7UE+/LSds2LRmx9yFqgBFEMQ1M3RYLhKiB2tuK2teGYrUTyZhCNLOAaFpWPEDAjL/fSjSMqeig6mAaWNpeQettfe/9sE0ieM61GDnnvTc8qKiY+vEAVXWIRlCi4eOfkXRMix2iIWy7H8ZybC+RnPNiJw3eBqJprlg4ZuQTtedhKgoWz250z9/RuhuJOqcTmTyXSM4corYc0O2Yejpa536sR/6K3r4bJRrBVPXjJwOTYycJFlvs83S0HlNPJ1xwGZbm5zHTsui77E5C0z8ae78VFYwgargPJdwLZpSoNdYbVowglqbnsTY9h97+OsakOYSmXUF4ymWowW70o3vROt7EtGYSnnIZ4SmLiDqmfsDfShgl0o/a24LW04TqO4Ia7Dp+guUjapuEkXMekUnnEbW7ATN2MFQ1TEsmaJaTtmmiBLtQ/UdR/Ucx9XSitklEbbmgKLHPSbAbtbc1dpLV+TbWsJdATmnsRCHvArTeI+jH9qJ3vE3UnkvYfQGRyecPCGYl0Il+7A30o2+gBjqI5C0gXHBJ7EQm1IfufQet6wBmWhaGYxqGY1rsxMU8PmSsWoZ/4mCEUf3tsb/JQ89ibXoO05JBYN5n6J/7SUzbJLSOfaS9W43e8RbhqYsInlNB1FmI0t+FpfkFrM0voIR8sc+gotE/9xOEpywEJAwGOREG//XSIX764iHq1lyJriXtrBuDfODZkRlF9R0BTFD02EHZCKFE+lGMIJHsYrAMPHNV+zxo3Y2E8+YP+ONRe1uwtL4CkUDsYGhGMVyzCLsvim3DNNG8DVgbd8SCw4gdMBUzEvvjB1DU472cyURtk2IHkWAXSr8X0+rEcM2M/eeYipmeHTtYvD/Qw36sR+qwHq7F0vI3TEsGRkYB0Qw36WqYSOchVF9L7I/mxEEHYr0pPS0WTKaBYoRjm8ubT3jq5YQLLkXzNpB24M9YG2tQw70Jv/+Gs4i+j3ydYPFSUFQsR/5K+hv/jcXzKqr/aDy8DbubiPsCjOzZqN2H0Dv2oXUdGDCcaSoakdxSwtMuJ2p3o/rbUf3tsWtfET9K2A+KRrD4RvrnrMRMy8IVPID51J1YW15KuN1h9wXox/ahe/ef9Nz02IE32B1rl54ee0/NaOx9NE99p7+p2zAtmShB74Bh1kHLaWnHA9WIfa6MUKwXO9z2292oGTlw7O1B7TEV9b2TDkWLDQFHIyhGGCX63jU6U7XGH0fTXKjBrmG9djQtKzbcbLGjhnpjvdhIIPZ5s9gxNStqf9eAE7KobRKhwn9A7fNgPfJi7GQow43ma8JUVKKZU9B8sZkcDGcRqq8JxYwStTqJ2nLiIxz+S75K/7zPABIGg5w4GP7sb4f58QuNPH/bFaRbRtjlPwuddV3lMTZm9Rvh+NCUqWgoRCESRDkegqZqBc0Sey7SjxIJxHp5OcWn7k2aUZR+L0o0HDs7PrnHagRjQ5RhP0rETzSzANOa2B+4y2Wny9uHpfl51N5WFNMEopiaFdOSGQ9WJdSD2t8NmISmLCTqOie+DbXPg966C9M2icjkeZhpTjCjaB1vYT1Sh9rbEuttHH9vUPXY9S89jWhGAYYzdgZtpmfHnjtem9Z1AL3jLdRAZ6wnqygo0QhK+PiwbDiAqWrHz7YtsZ5ARh5R22QUI4gS6ED1HwVMzLQsomkuTHsukZxzMdOzcbnsdHvasLTuQj/6OoazkMjkeRiuWaiBY+jte9Db61FDPfH9F03PITJ5HpHceZiWTPSON7G07kLrfJuoYxqRnHMxsmaihHvRfM2ovmaUSD+oenzfq0EvSsAbG8qzZGJaHZh6eqzNYT9E+jHTXUTteUTteUTySonklsaH/LTOd0jf+99oPU2EZiwmeM7HMe2T0boOYD34FywtLxOZPJfQ9GuIuC947z09iYTBSU4cDH7xSjObnjvAM7deTmbaGXc7xbiRMJD6U7X+VK4dPoT7DM5Wuho765L5iYQQYmhJGwba8TCQ+YmEEGJoSRsGuoSBEEIMW9KGgSbDREIIMWxJGwbSMxBCiOFL+jCQnoEQQgwt6cNgpD99+e6xPh7922EOewOnXMY0TV5q7GR3c/eIXkMIIc4USfsFfE2N5dzGZw+Qm2klw6pz9awcLivKHjA9RW8wQsQwcdljt8NHjCiP7Wpmy0uHCBsmP32xkY/PzeOWy6YzI+e9H8447A3wg2fe5cWDsbsKr5/n5mv/MBOX7aTb6k/S7gvywoEOprpsXFzoiofW6fQGIzzzzjE0VSHDqmO3qoQNk0DYIBA2WDAli8LsU8x9c4YxoiaHvQGKcmyoEzCzbSpqONpHukVlmuvs+IyIiZG0YTDHncnFhVl0+kMc6QrQFYiwdXcLsydn8KmLpxIxojzzTge7mrowoiZTstKZl+/gsDfA2+29lJ+Xy+cWTqd6r4fHd7fw1Jvt5GZaKXTZyLFbeO7dDqyaytf+YSY9/WEe29XMCwc6+fIVRSw+N5es94WC1x/ihQOd/O++dnYd7opPOJCVrvPR4slcWuhiRo6dohzbgLulTdPkz/va2fTcATr94VPWatEUPntpIasumx7/t3ZfkNaefqZkpTM5w/qhTCneHzbY09LDrsNdvHK4i6hpsrQ0n+vmurFZVJ5p6OCnLzRysNNPaYGTuz42m3PzMkf9uqZpsvtID7/9+xE6+yNEIrHeYHFuBrdcNh23Y+Q/u9rTH+bb2/ez50gP8woczJ/i5Ly8TDLTdNJ1FWe6Tr4zfdjbe+VwF8+928FHpru4rCgbqz7yznnYiPJmm4/+cJSQESVqwpVzdU58gvrDBj95sZH/efUIdqvGxuUlXDgta8SvN1EaO/z4ghEcaTqZaRrOdMuo3rfxFomaHOr0M3OSfUKm8h+ppL0D+WShSJTtb7Xzy1ebefdY7PlCVzrXFE/GZbOwt83HG60+TNPkjrLZXFM8Ob6u1x/iyb0eDnT4afIGaO3p59LpLm696hwmZ8YONA1H+7h3x37eaPWhKXDhtCzOzcvk783dvOXpxQSmZqVz3dw8Fp+XS5M3QM3+ozz/bif+8HtzsLgdaUzNSmdKVjpN3gB7WnqYl+/gXz46kxy7lb5QhL6QgUVTsVs0UOBnfzvM9reOMs2VztXn5vLXhmM0dr43vJWmqxQ407BZNNJ0lTRdZUpWeiyAsu1094d595ifAx19WDWVBVOdXDQti3RdY+e7HTz3bgd723zoqkK6rpJu0Zh6fP3CbBut3f283trD/qN9GFETTVUoyXcQCBvsP9pHhlUj35nGu8f8zMixce2cPH779xZ8/WE+ceFUFs7IJt2iYrNoGFETf8jAHzKImiYWTcWqqxhRk6O9Qdp9Ibr7w2SlW5iUYUFXVZ54vZXXW324bBbOn+IkEo6tu/tID5qq8IkLplAxJ4+GY3282eajsdOPYZpETdAU+EhRNhVz8piSNfCg/pbHxzf++CbtvSGuKZ5Mw9E+DnYO/mxdUpjF5xcWcXFh1mn/+LfVt/Ldp98haoIJZFg1rjgnh6mudCbZrUzKsDI7N4OibNtpt9PcFWBbfRtP7m0bdJKgqQoXT8ti4YxsttW30tTVz7LSfHYf6aa1J8h/Xj+Xq2ZNGrBOS3c/v9vTSl1jJ1fNzOGTF00l2x6bUqPhWB9/eqON7v4IuRlWJmdYiQKHO/0c9gbwBsJk2yzkZFjJy7Qyf4qTi6a5cKTrRKIm+9t7qW/pwZmuc9G0rHhwBiNRGo724g2EmTkpgwJn2qCaGzv8/OTFRmrfOTboPciwamTbLUzPtlExJ49riidTkOugo7OPhqN97PX4sKjK8QDR6QsZdPhDdPSF6AsaRKKxALVqKnmZabgdaczIsXNuXka8HaZpUvvOMR556TDTs20sOd/NohnZp5zrzDRNnmvo4McvHKSxM0DFnFzu+ljxqGZAaOnup2b/UayaSk6GlUkZFubkObBbB0+xI9NRnGSoW9JN06S+pYcMq86syWOb3FHTZJ+nl+cajvFsQweHO/2UTnGycEY2i2bkMNedOej1QpEoh7sCNHb4Odjpp7krQEt3P0e6+wH40uVF3FCSP+SQyt8Oefl/NQ0c7Q1xwVQnHynKZkaOjdaeIM1dATy+IP3hKEEjSiBk0NwVoLv/vYnDdFWhKMdGIGTQ0jNw1tDz8jK5pNCFosT+iPtCEZq8/TR2xs7YbBaVefkOSgqcXDA1iwumOcmw6pimyRutPrbubuFAh59PXDiFJee70VWF7kCYB19oZFt9K4l+ADOsGn2h9wJ0alY6n75kGjfMc5Of64jv/5bufjbXHeKpvZ74a9gsKjMnZWDVVRSgL2TwdntsQrr5U5yck2MnTVeJmiZ/fKMNl83Cd288n5ICJwBdgTCNHX78YYP+SJQmb4Bfv3aEY30hSgscuB3peAMhOv1h8h1pXDkzh8vPyWFbfRuP7Wpi0Yxs1i+Zw942HzVvH+Vvh7x09IUw3vcmONN1zs93MDnDiqqAoij0BQ3ae4Mc7Q3S2hNEU+CqWZNYcr6bHLsFi6YSNqK82uKj+vVWDnsDTM1K59+uPZdLprvo8oe57fevs7+9l89eNp10XcUfMmg41seLBzpRFTjP7WBfmw+rrrLk/DwOHPOzp6UHi6aQbbPQ4Q/Hv5CRmaZRlG0n226hOxCmwx/maG+QsGGiAOdMstPWExxwogMwzZWOzaJxoMM/4MsdGVaNmZPsuGwWnOk6wYhJ7TtHSdc1Pn3JVOblO+kNRvAFI/T0R/AGwnj9IV5v6aGlJ4jNorJgmou9LT34gqeeEO/EZ8Cixf7rDxsDPkvFuRksn1/AgilOfvzCQf560Ms5OXa6AuF48BVm2wgb74XJpAwrk+xWDnT08Xqrjxk5Ni4ryubx3S24HWmsXzKHKVnpNHb6OdQZQFXAZbPERw+8/ti2jahJXqYVtyONQDjK1t0tPNtwjJO/A5Omq1w5M4fy83K5auakeC9JwuAkZ9L8JJGoOaxrAmPFNE0cThu9vv5hLd/lD3PI68eRrjPdZYuf8bT19LP7SA/+UITLz8k55TCIaZp090fITNNHXGdrTz/HekPHr39E0VUFm1Ulw6KjqhAyTMKRKIoCuZlp5GZasWgqESNKpz9MTzDCjBx7/PU/aP8f6Ojj7fZezs3NZEaOPX4PyglHugP85a2j1O4/Roc/RCgSpT8S5ZJCF//+8XPjZ8mnEoxE+eMbbWz9ewtR0yTHHvtDP9DhH/AFhBULCrijbPag9ypqmnQHwrT3hnjL4+P1Vh9vtvnw9UeImiYmYLNo5GVayTt+Blt5vpu8Dxj+crnseL19HOnuZ3KGdcCwY18owp1/eJOXD3cBsV7E5AwrlfPc3DS/ALcjjcYOPz/f1cSf97UzxZnGTQumcP35blx2S7ydEDuYfdCJzRttPbx6uJv61h6mZqVz0bQsFkzNojsQ5pWmLl5r6iZoRJnrzmSu20GO3cK7x/rYf7SPQ51+evpjB/xgJMq1c/K45bJCck7z/kdNkz1Heqh+08PbR/s4LzeDSwpdlE6JHRh7+w18wQgZaRqT7FZy7JZBZ/a9wQgeX5A9LT38fk9r/OTAbtH45ytnsPKCKWCa1DV62f5WO53+MFZNxaIphIwonX1hOvwhrJrKqo8Ucn1JPrqq8HpLD//21Fu0dA/v7/FkznSd5fMLWLGggDRdpbMvjKc3yIsHOqnZf5ROf5h/urSQr14dm2RQwuAkZ1IYTASp/8yq/1CnnxcPdjLJbuXaObnjPoY8nJ5xT38Em0XDoimnbE9/2MCqq2fVRf6x2PemafJmm4/XmrupmJP3gYGbiN5ghN/tacVm0Thnko3p2XZUJdbD9PrDKApk26y4bDqaqtDeG6LdFyRkRLninJxTzrhsRE1eb+lhqiud3OND1RIGJznTDgYfNqlf6k/V+lO5dpBZS4UQQowBCQMhhBASBkIIISQMhBBCMMww2LlzJxUVFZSXl7N58+ZBz4dCIdasWUN5eTkrV66kubk5/txDDz1EeXk5FRUVPP/882PXciGEEGNmyDAwDIP169ezZcsWqqurefLJJ2loaBiwzNatW3E6nezYsYNVq1axYcMGABoaGqiurqa6upotW7bwrW99C8MwPuhlhBBCTKAhw6C+vp6ioiIKCwuxWq1UVlZSU1MzYJna2lqWL18OQEVFBXV1dZimSU1NDZWVlVitVgoLCykqKqK+vn58KhFCCDFiQ06a4fF4yM/Pjz92u92DDugej4eCgoLYBnUdh8OB1+vF4/GwYMGCAet6PJ7Tvp7Foo36+7KjXf9sJ/VL/akqlWsfLbmALIQQYugwcLvdtLW1xR97PB7cbvegZVpbWwGIRCL4fD6ys7OHta4QQoiJN2QYlJaW0tjYSFNTE6FQiOrqasrKygYsU1ZWxrZt2wDYvn07CxcuRFEUysrKqK6uJhQK0dTURGNjI/Pnzx+fSoQQQozYkNcMdF1n3bp1rF69GsMwWLFiBcXFxWzatImSkhIWL15MVVUVa9eupby8nKysLDZu3AhAcXEx1113HUuWLEHTNNatW4emffDES0IIISbOGTdRnRBCiA+fXEAWQgghYSCEECLJwmCoaTOSSWtrKzfffDNLliyhsrKSn//85wB0dXVxyy23cO2113LLLbfQ3d09wS0dX4ZhsGzZMr70pS8B0NTUxMqVKykvL2fNmjWEQqEJbuH46enp4bbbbuPjH/841113HX//+99Tav//7Gc/o7Kykuuvv57bb7+dYDCY1Pv/7rvvZtGiRVx//fXxfzvV/jZNk//4j/+gvLycG264gb179w65/aQJg+FMm5FMNE3jrrvu4qmnnuI3v/kNv/rVr2hoaGDz5s0sWrSIv/zlLyxatCjpQ/Gxxx5j1qxZ8ccbNmxg1apV7NixA6fTyeOPPz6BrRtf9957L1dddRX/+7//yx/+8AdmzZqVMvvf4/Hw2GOP8bvf/Y4nn3wSwzCorq5O6v1/0003sWXLlgH/dqr9vXPnThobG/nLX/7Ct7/9be65554ht580YTCcaTOSSV5eHvPmzQMgMzOTmTNn4vF4qKmpYdmyZQAsW7aMp59+egJbOb7a2tp49tlnqaqqAmJnQy+99BIVFRUALF++PGk/Az6fj127dsVrt1qtOJ3OlNr/hmHQ399PJBKhv7+f3NzcpN7/l156KVlZWQP+7VT7+8S/K4rCBRdcQE9PD+3t7afdftKEwQdNmzHU1BfJorm5mX379rFgwQI6OjrIy8sDIDc3l46Ojglu3fi57777WLt2Laoa+xh7vV6cTie6HvvGdH5+ftJ+Bpqbm8nJyeHuu+9m2bJlfPOb38Tv96fM/ne73Xzuc5/jmmuu4corryQzM5N58+alzP4/4VT7++Tj4XDei6QJg1TV19fHbbfdxr/+67+SmZk54DlFOfUPnp/tnnnmGXJycigpKZnopkyISCTCm2++yac+9SmeeOIJbDbboCGhZN7/3d3d1NTUUFNTw/PPP08gEEj5KfJHu7+HvOnsbJGKU1+Ew2Fuu+02brjhBq699loAJk2aRHt7O3l5ebS3t5OTkzPBrRwfr732GrW1tezcuZNgMEhvby/33nsvPT09RCIRdF2nra0taT8D+fn55OfnxyeC/PjHP87mzZtTZv//9a9/Zdq0afH6rr32Wl577bWU2f8nnGp/n3w8HM57kTQ9g+FMm5FMTNPkm9/8JjNnzuSWW26J/3tZWRlPPPEEAE888QSLFy+eoBaOr69//evs3LmT2tpafvCDH7Bw4UK+//3vc9lll7F9+3YAtm3blrSfgdzcXPLz8zlw4AAAdXV1zJo1K2X2/5QpU9izZw+BQADTNKmrq2P27Nkps/9PONX+PvHvpmmye/duHA5HfDjpVJLqDuTnnnuO++67Lz5txpe//OWJbtK4eeWVV/j0pz/NueeeGx8zv/3225k/fz5r1qyhtbWVKVOmcP/99+NyuSa2sePsb3/7G4888ggPPfQQTU1N/Mu//Avd3d3MnTuXDRs2YLVaJ7qJ42Lfvn1885vfJBwOU1hYyH/+538SjUZTZv8/8MADPPXUU+i6zty5c7n33nvxeDxJu/9vv/12Xn75ZbxeL5MmTeKrX/0qH/vYxz5wf5umyfr163n++eex2Wzcd999lJaWnnb7SRUGQgghRiZphomEEEKMnISBEEIICQMhhBASBkIIIZAwEEIIgYSBEEIIJAyEEEIA/x9k5kTwhaz3rQAAAABJRU5ErkJggg==\n"
1114 |           },
1115 |           "metadata": {}
1116 |         }
1117 |       ]
1118 |     },
1119 |     {
1120 |       "cell_type": "code",
1121 |       "source": [
1122 |         "yhat = model.predict(testX)\n",
1123 |         "tyhat = model.predict(trainX)\n",
1124 |         "plt.plot(yhat, label='Prediction')\n",
1125 |         "plt.plot(testY, label='Real data')\n",
1126 |         "plt.legend()\n",
1127 |         "plt.show()"
1128 |       ],
1129 |       "metadata": {
1130 |         "colab": {
1131 |           "base_uri": "https://localhost:8080/",
1132 |           "height": 301
1133 |         },
1134 |         "id": "bT7ze93vvNC1",
1135 |         "outputId": "9c768592-b5f6-4007-f8e3-82414b675df3"
1136 |       },
1137 |       "execution_count": null,
1138 |       "outputs": [
1139 |         {
1140 |           "output_type": "stream",
1141 |           "name": "stdout",
1142 |           "text": [
1143 |             "367/367 [==============================] - 1s 2ms/step\n",
1144 |             "367/367 [==============================] - 1s 2ms/step\n"
1145 |           ]
1146 |         },
1147 |         {
1148 |           "output_type": "display_data",
1149 |           "data": {
1150 |             "text/plain": [
1151 |               "<Figure size 432x288 with 1 Axes>"
1152 |             ],
1153 |             "image/png": "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\n"
1154 |           },
1155 |           "metadata": {}
1156 |         }
1157 |       ]
1158 |     },
1159 |     {
1160 |       "cell_type": "code",
1161 |       "source": [
1162 |         "print(\"Shape of train data :\", trainX.shape)\n",
1163 |         "print(\"Shape of prediction :\", yhat.shape)\n",
1164 |         "print(\"Shape of real data :\", testY.shape)"
1165 |       ],
1166 |       "metadata": {
1167 |         "colab": {
1168 |           "base_uri": "https://localhost:8080/"
1169 |         },
1170 |         "id": "BIy2IVvhvREd",
1171 |         "outputId": "ea1f85b6-568d-43e5-c247-1c75f9aa252b"
1172 |       },
1173 |       "execution_count": null,
1174 |       "outputs": [
1175 |         {
1176 |           "output_type": "stream",
1177 |           "name": "stdout",
1178 |           "text": [
1179 |             "Shape of train data : (11738, 1, 1)\n",
1180 |             "Shape of prediction : (11738, 1)\n",
1181 |             "Shape of real data : (11738,)\n"
1182 |           ]
1183 |         }
1184 |       ]
1185 |     },
1186 |     {
1187 |       "cell_type": "code",
1188 |       "source": [
1189 |         "rmse = math.sqrt(mean_squared_error(testY, yhat))\n",
1190 |         "mae = mean_absolute_error(testY, yhat)\n",
1191 |         "print('Test RMSE: %.3f' % rmse)\n",
1192 |         "print('Test MAE: %.3f' % mae)"
1193 |       ],
1194 |       "metadata": {
1195 |         "colab": {
1196 |           "base_uri": "https://localhost:8080/"
1197 |         },
1198 |         "id": "2eRVod_3vbL9",
1199 |         "outputId": "8424fc99-9fd8-4320-f6fd-a53ae6e6b510"
1200 |       },
1201 |       "execution_count": null,
1202 |       "outputs": [
1203 |         {
1204 |           "output_type": "stream",
1205 |           "name": "stdout",
1206 |           "text": [
1207 |             "Test RMSE: 0.009\n",
1208 |             "Test MAE: 0.008\n"
1209 |           ]
1210 |         }
1211 |       ]
1212 |     },
1213 |     {
1214 |       "cell_type": "markdown",
1215 |       "source": [
1216 |         "VISUALIZATION\n"
1217 |       ],
1218 |       "metadata": {
1219 |         "id": "S7NLXiFOvgni"
1220 |       }
1221 |     },
1222 |     {
1223 |       "cell_type": "code",
1224 |       "source": [
1225 |         "print(\"Shape of cycle :\", np.shape(cycle))\n",
1226 |         "\n",
1227 |         "cycle1 = cycle[0:train_size-1]\n",
1228 |         "cycle2 = cycle[train_size+1:len(dataset)]\n",
1229 |         "\n",
1230 |         "# cycle1 = cycle[0:train_size]\n",
1231 |         "# cycle2 = cycle[train_size:len(dataset)]\n",
1232 |         "\n",
1233 |         "print(np.shape(cycle1), np.shape(cycle2))"
1234 |       ],
1235 |       "metadata": {
1236 |         "colab": {
1237 |           "base_uri": "https://localhost:8080/"
1238 |         },
1239 |         "id": "tWv949wtvd8F",
1240 |         "outputId": "c2b8165c-61f7-47b2-e0e1-4769b8482b53"
1241 |       },
1242 |       "execution_count": null,
1243 |       "outputs": [
1244 |         {
1245 |           "output_type": "stream",
1246 |           "name": "stdout",
1247 |           "text": [
1248 |             "Shape of cycle : (23478,)\n",
1249 |             "(11738,) (11738,)\n"
1250 |           ]
1251 |         }
1252 |       ]
1253 |     },
1254 |     {
1255 |       "cell_type": "code",
1256 |       "source": [
1257 |         "sns.set_style(\"darkgrid\")\n",
1258 |         "plt.figure(figsize=(12, 8))\n",
1259 |         "\n",
1260 |         "plt.plot(cycle1, trainX[:,0],label='Used real data', linewidth = 3, color='r')\n",
1261 |         "plt.plot(cycle2, testY,label='Real data', linewidth = 3, color='b')\n",
1262 |         "# plt.plot(cycle2, yhat[:,0],label='LSTM Prediction', linewidth = 3, color='g')\n",
1263 |         "plt.plot(cycle2, yhat[:,0],label='LSTM Prediction', linewidth = 3, color='g')\n",
1264 |         "plt.legend(prop={'size': 16})\n",
1265 |         "\n",
1266 |         "plt.ylabel('SoH', fontsize = 15)\n",
1267 |         "plt.xlabel('Discharge cycle', fontsize = 15)\n",
1268 |         "plt.title(str(num[7]) + \" SOH Prediction\", fontsize = 15)\n",
1269 |         "\n",
1270 |         "plt.show()"
1271 |       ],
1272 |       "metadata": {
1273 |         "colab": {
1274 |           "base_uri": "https://localhost:8080/",
1275 |           "height": 520
1276 |         },
1277 |         "id": "fIUgI2JEvmLv",
1278 |         "outputId": "4eb4d8f6-4663-4cae-9c38-0240bdec6fae"
1279 |       },
1280 |       "execution_count": null,
1281 |       "outputs": [
1282 |         {
1283 |           "output_type": "display_data",
1284 |           "data": {
1285 |             "text/plain": [
1286 |               "<Figure size 864x576 with 1 Axes>"
1287 |             ],
1288 |             "image/png": "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\n"
1289 |           },
1290 |           "metadata": {}
1291 |         }
1292 |       ]
1293 |     },
1294 |     {
1295 |       "cell_type": "code",
1296 |       "source": [],
1297 |       "metadata": {
1298 |         "id": "SeB8FvqSvswL"
1299 |       },
1300 |       "execution_count": null,
1301 |       "outputs": []
1302 |     }
1303 |   ]
1304 | }


--------------------------------------------------------------------------------
/batterysoc.mlx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/harshika0926/Battery-management-system-using-machine-learning/02982225a3166e573e936d214f8425d483b6a78f/batterysoc.mlx


--------------------------------------------------------------------------------
/neuralnetwork.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |   "nbformat": 4,
  3 |   "nbformat_minor": 0,
  4 |   "metadata": {
  5 |     "colab": {
  6 |       "provenance": []
  7 |     },
  8 |     "kernelspec": {
  9 |       "name": "python3",
 10 |       "display_name": "Python 3"
 11 |     },
 12 |     "language_info": {
 13 |       "name": "python"
 14 |     }
 15 |   },
 16 |   "cells": [
 17 |     {
 18 |       "cell_type": "code",
 19 |       "execution_count": null,
 20 |       "metadata": {
 21 |         "id": "sp43pAyow7LY"
 22 |       },
 23 |       "outputs": [],
 24 |       "source": [
 25 |         "2import numpy as np  \n",
 26 |         "import pandas as pd  \n",
 27 |         "from matplotlib import pyplot as plt \n",
 28 |         "import torch \n",
 29 |         "import torch.nn as nn\n",
 30 |         "import tensorflow as tf\n",
 31 |         "import time"
 32 |       ]
 33 |     },
 34 |     {
 35 |       "cell_type": "code",
 36 |       "source": [
 37 |         "test = pd.read_csv(\"test.csv\")\n",
 38 |         "train=pd.read_csv(\"train.csv\")\n",
 39 |         "num_of_test_data = test.shape[0]\n",
 40 |         "num_of_train_data = train.shape[0]\n",
 41 |         "target = train[\"Capacity(State of Charge)\"]\n",
 42 |         "train.drop([\"Capacity(State of Charge)\"],axis = 1 , inplace = True)\n",
 43 |         "val = test[\"Capacity(State of Charge)\"]\n",
 44 |         "test.drop([\"Capacity(State of Charge)\"],axis = 1 , inplace = True) \n",
 45 |         "combined = train.append(test)\n",
 46 |         "combined.reset_index(inplace=True)\n",
 47 |         "combined.drop(['index'], inplace=True, axis=1)\n",
 48 |         "train_features = torch.tensor(combined[:num_of_train_data].values, dtype=torch.float)\n",
 49 |         "train_labels = torch.tensor(target.values, dtype=torch.float).view(-1, 1)\n",
 50 |         "test_features = torch.tensor(combined[num_of_train_data:].values, dtype=torch.float)\n",
 51 |         "test_true_value = torch.tensor(val.values, dtype=torch.float).view(-1, 1)\n",
 52 |         "print(\"train data size: \", train_features.shape)\n",
 53 |         "print(\"train label data size: \", train_labels.shape)\n",
 54 |         "#print(\"valid data size: \", val_features.shape)\n",
 55 |         "#print(\"valid label data size: \", val_labels.shape)\n",
 56 |         "print(\"test data size: \", test_features.shape)\n",
 57 |         "print(\"test label data size: \", test_true_value.shape)"
 58 |       ],
 59 |       "metadata": {
 60 |         "colab": {
 61 |           "base_uri": "https://localhost:8080/"
 62 |         },
 63 |         "id": "3aDIzMChxCFl",
 64 |         "outputId": "a14df882-7216-4e71-d5c9-fd36dd0ee34c"
 65 |       },
 66 |       "execution_count": null,
 67 |       "outputs": [
 68 |         {
 69 |           "output_type": "stream",
 70 |           "name": "stdout",
 71 |           "text": [
 72 |             "train data size:  torch.Size([118, 6])\n",
 73 |             "train label data size:  torch.Size([118, 1])\n",
 74 |             "test data size:  torch.Size([50, 6])\n",
 75 |             "test label data size:  torch.Size([50, 1])\n"
 76 |           ]
 77 |         }
 78 |       ]
 79 |     },
 80 |     {
 81 |       "cell_type": "code",
 82 |       "source": [
 83 |         "class Net(nn.Module):\n",
 84 |         "  \n",
 85 |         "    def __init__(self, features):\n",
 86 |         "        super(Net, self).__init__()\n",
 87 |         "        \n",
 88 |         "        self.linear_relu1 = nn.Linear(features, 128)\n",
 89 |         "        self.linear_relu2 = nn.Linear(128, 256)\n",
 90 |         "        self.linear_relu3 = nn.Linear(256, 256)\n",
 91 |         "        self.linear_relu4 = nn.Linear(256, 256)\n",
 92 |         "        self.linear5 = nn.Linear(256, 1)\n",
 93 |         "        \n",
 94 |         "    def forward(self, x):\n",
 95 |         "        \n",
 96 |         "        y_pred = self.linear_relu1(x)\n",
 97 |         "        y_pred = nn.functional.relu(y_pred)\n",
 98 |         "\n",
 99 |         "        y_pred = self.linear_relu2(y_pred)\n",
100 |         "        y_pred = nn.functional.relu(y_pred)\n",
101 |         "\n",
102 |         "        y_pred = self.linear_relu3(y_pred)\n",
103 |         "        y_pred = nn.functional.relu(y_pred)\n",
104 |         "\n",
105 |         "        y_pred = self.linear_relu4(y_pred)\n",
106 |         "        y_pred = nn.functional.relu(y_pred)\n",
107 |         "\n",
108 |         "        y_pred = self.linear5(y_pred)\n",
109 |         "        return y_pred\n",
110 |         "\n",
111 |         "model = Net(features=train_features.shape[1])\n",
112 |         "\n",
113 |         "# loss function\n",
114 |         "criterion = nn.MSELoss(reduction='mean')\n",
115 |         "\n",
116 |         "optimizer = torch.optim.Adam(model.parameters(), lr=1e-4)\n"
117 |       ],
118 |       "metadata": {
119 |         "id": "CFYTZn-2yGv0"
120 |       },
121 |       "execution_count": null,
122 |       "outputs": []
123 |     },
124 |     {
125 |       "cell_type": "code",
126 |       "source": [
127 |         "# train model\n",
128 |         "epoch = 1000\n",
129 |         "train_losses = []\n",
130 |         "val_losses = []\n",
131 |         "s_time = time.time()\n",
132 |         "for t in range(epoch):\n",
133 |         "    model.train()\n",
134 |         "    y_pred = model(train_features)\n",
135 |         "    \n",
136 |         "    tr_loss = criterion(y_pred, train_labels)\n",
137 |         "    train_losses.append(tr_loss.item())\n",
138 |         "    \n",
139 |         "    if torch.isnan(tr_loss):\n",
140 |         "        break\n",
141 |         "    \n",
142 |         "    optimizer.zero_grad()\n",
143 |         "    \n",
144 |         "    # backward\n",
145 |         "    tr_loss.backward()\n",
146 |         "    optimizer.step()\n",
147 |         "    \n",
148 |         "    \"\"\"\n",
149 |         "    model.eval()\n",
150 |         "    with torch.no_grad():\n",
151 |         "        val_pred = model(val_features)\n",
152 |         "        val_loss = criterion(val_pred, val_labels)\n",
153 |         "        # print(t, loss.item())\n",
154 |         "        val_losses.append(val_loss.item())\n",
155 |         "    \"\"\"\n",
156 |         "\n",
157 |         "    if t % 100 == 0:  \n",
158 |         "        print(f\"Epoch: {t}, Train loss :{tr_loss:.3f}\") #, Val loss :{val_loss:.3f}\n",
159 |         "c_time = time.time()\n",
160 |         "ss_time = c_time - s_time \n",
161 |         "print('%.3f秒'%ss_time)\n",
162 |         "    "
163 |       ],
164 |       "metadata": {
165 |         "colab": {
166 |           "base_uri": "https://localhost:8080/"
167 |         },
168 |         "id": "wT1wUWR6yaF9",
169 |         "outputId": "8b5b18f9-8db0-4d56-8da8-8537cde1da1c"
170 |       },
171 |       "execution_count": null,
172 |       "outputs": [
173 |         {
174 |           "output_type": "stream",
175 |           "name": "stdout",
176 |           "text": [
177 |             "Epoch: 0, Train loss :0.000\n",
178 |             "Epoch: 100, Train loss :0.000\n",
179 |             "Epoch: 200, Train loss :0.000\n",
180 |             "Epoch: 300, Train loss :0.000\n",
181 |             "Epoch: 400, Train loss :0.000\n",
182 |             "Epoch: 500, Train loss :0.000\n",
183 |             "Epoch: 600, Train loss :0.000\n",
184 |             "Epoch: 700, Train loss :0.000\n",
185 |             "Epoch: 800, Train loss :0.000\n",
186 |             "Epoch: 900, Train loss :0.000\n",
187 |             "7.740秒\n"
188 |           ]
189 |         }
190 |       ]
191 |     },
192 |     {
193 |       "cell_type": "code",
194 |       "source": [
195 |         "plt.plot(np.arange(1000), train_losses)  \n",
196 |         "#plt.plot(np.arange(1000), val_losses)\n",
197 |         "plt.show()  \n",
198 |         "\n",
199 |         "plt.plot(np.arange(num_of_train_data), train_labels,\"-\",label = 'train') \n",
200 |         "plt.plot(np.arange(num_of_train_data), y_pred.detach().numpy(),\".\",label = 'pred')  \n",
201 |         "plt.legend() "
202 |       ],
203 |       "metadata": {
204 |         "colab": {
205 |           "base_uri": "https://localhost:8080/",
206 |           "height": 530
207 |         },
208 |         "id": "K7fcZi2_yuh0",
209 |         "outputId": "a85996dc-e0a0-4713-d3d6-40b38c94df11"
210 |       },
211 |       "execution_count": null,
212 |       "outputs": [
213 |         {
214 |           "output_type": "display_data",
215 |           "data": {
216 |             "text/plain": [
217 |               "<Figure size 432x288 with 1 Axes>"
218 |             ],
219 |             "image/png": "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\n"
220 |           },
221 |           "metadata": {
222 |             "needs_background": "light"
223 |           }
224 |         },
225 |         {
226 |           "output_type": "execute_result",
227 |           "data": {
228 |             "text/plain": [
229 |               "<matplotlib.legend.Legend at 0x7f7d4e20fb80>"
230 |             ]
231 |           },
232 |           "metadata": {},
233 |           "execution_count": 11
234 |         },
235 |         {
236 |           "output_type": "display_data",
237 |           "data": {
238 |             "text/plain": [
239 |               "<Figure size 432x288 with 1 Axes>"
240 |             ],
241 |             "image/png": "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\n"
242 |           },
243 |           "metadata": {
244 |             "needs_background": "light"
245 |           }
246 |         }
247 |       ]
248 |     },
249 |     {
250 |       "cell_type": "code",
251 |       "source": [
252 |         "predictions = model(test_features).detach().numpy()\n",
253 |         "plt.plot(np.arange(num_of_test_data), test_true_value,\"-\",label = 'test')  \n",
254 |         "plt.plot(np.arange(num_of_test_data), predictions,\".\",label = 'pred') \n",
255 |         "plt.legend()"
256 |       ],
257 |       "metadata": {
258 |         "colab": {
259 |           "base_uri": "https://localhost:8080/",
260 |           "height": 282
261 |         },
262 |         "id": "ySTXcUe5zb2L",
263 |         "outputId": "86383415-fbd4-4600-f546-2ec9c5d2c8cc"
264 |       },
265 |       "execution_count": null,
266 |       "outputs": [
267 |         {
268 |           "output_type": "execute_result",
269 |           "data": {
270 |             "text/plain": [
271 |               "<matplotlib.legend.Legend at 0x7f7d4e18b4f0>"
272 |             ]
273 |           },
274 |           "metadata": {},
275 |           "execution_count": 12
276 |         },
277 |         {
278 |           "output_type": "display_data",
279 |           "data": {
280 |             "text/plain": [
281 |               "<Figure size 432x288 with 1 Axes>"
282 |             ],
283 |             "image/png": "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\n"
284 |           },
285 |           "metadata": {
286 |             "needs_background": "light"
287 |           }
288 |         }
289 |       ]
290 |     }
291 |   ]
292 | }


--------------------------------------------------------------------------------
/ridgemsoc.mlx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/harshika0926/Battery-management-system-using-machine-learning/02982225a3166e573e936d214f8425d483b6a78f/ridgemsoc.mlx


--------------------------------------------------------------------------------
/smdatasoc.mlx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/harshika0926/Battery-management-system-using-machine-learning/02982225a3166e573e936d214f8425d483b6a78f/smdatasoc.mlx


--------------------------------------------------------------------------------