├── README.md
└── Credit_card_dataset.ipynb


/README.md:
--------------------------------------------------------------------------------
1 | # Credit-Card-Fraud-Detection-Dataset-
2 | Classification Problem to detect credit card fraud
3 | 


--------------------------------------------------------------------------------
/Credit_card_dataset.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |   "nbformat": 4,
   3 |   "nbformat_minor": 0,
   4 |   "metadata": {
   5 |     "colab": {
   6 |       "name": "Credit_card_dataset.ipynb",
   7 |       "provenance": [],
   8 |       "collapsed_sections": [],
   9 |       "authorship_tag": "ABX9TyMUlfzY4apPBDRCwmR09ep4",
  10 |       "include_colab_link": true
  11 |     },
  12 |     "kernelspec": {
  13 |       "name": "python3",
  14 |       "display_name": "Python 3"
  15 |     }
  16 |   },
  17 |   "cells": [
  18 |     {
  19 |       "cell_type": "markdown",
  20 |       "metadata": {
  21 |         "id": "view-in-github",
  22 |         "colab_type": "text"
  23 |       },
  24 |       "source": [
  25 |         "<a href=\"https://colab.research.google.com/github/Apurva-tech/Credit-Card-Fraud-Detection-Dataset-/blob/master/Credit_card_dataset.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
  26 |       ]
  27 |     },
  28 |     {
  29 |       "cell_type": "code",
  30 |       "metadata": {
  31 |         "id": "XWffzKmi8ee4",
  32 |         "colab_type": "code",
  33 |         "outputId": "09bf68f6-4e5d-4a15-944a-81aa6748ce12",
  34 |         "colab": {
  35 |           "base_uri": "https://localhost:8080/",
  36 |           "height": 34
  37 |         }
  38 |       },
  39 |       "source": [
  40 |         "from google.colab import drive\n",
  41 |         "drive.mount('/content/drive')"
  42 |       ],
  43 |       "execution_count": 1,
  44 |       "outputs": [
  45 |         {
  46 |           "output_type": "stream",
  47 |           "text": [
  48 |             "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
  49 |           ],
  50 |           "name": "stdout"
  51 |         }
  52 |       ]
  53 |     },
  54 |     {
  55 |       "cell_type": "code",
  56 |       "metadata": {
  57 |         "id": "pdBTzVNC84ar",
  58 |         "colab_type": "code",
  59 |         "colab": {}
  60 |       },
  61 |       "source": [
  62 |         "import pandas as pd\n",
  63 |         "df = pd.read_csv('/content/drive/My Drive/CreditCard/creditcard.csv')"
  64 |       ],
  65 |       "execution_count": 0,
  66 |       "outputs": []
  67 |     },
  68 |     {
  69 |       "cell_type": "code",
  70 |       "metadata": {
  71 |         "id": "-u3D13mROVDI",
  72 |         "colab_type": "code",
  73 |         "outputId": "3b7b1fd8-9651-40c5-b347-138a378ffa19",
  74 |         "colab": {
  75 |           "base_uri": "https://localhost:8080/",
  76 |           "height": 34
  77 |         }
  78 |       },
  79 |       "source": [
  80 |         "df.shape"
  81 |       ],
  82 |       "execution_count": 3,
  83 |       "outputs": [
  84 |         {
  85 |           "output_type": "execute_result",
  86 |           "data": {
  87 |             "text/plain": [
  88 |               "(284807, 31)"
  89 |             ]
  90 |           },
  91 |           "metadata": {
  92 |             "tags": []
  93 |           },
  94 |           "execution_count": 3
  95 |         }
  96 |       ]
  97 |     },
  98 |     {
  99 |       "cell_type": "code",
 100 |       "metadata": {
 101 |         "id": "bUBNSbjxAQYd",
 102 |         "colab_type": "code",
 103 |         "outputId": "83b9786b-d96c-44f9-a0ea-41021efd2775",
 104 |         "colab": {
 105 |           "base_uri": "https://localhost:8080/",
 106 |           "height": 224
 107 |         }
 108 |       },
 109 |       "source": [
 110 |         "df.head()"
 111 |       ],
 112 |       "execution_count": 4,
 113 |       "outputs": [
 114 |         {
 115 |           "output_type": "execute_result",
 116 |           "data": {
 117 |             "text/html": [
 118 |               "<div>\n",
 119 |               "<style scoped>\n",
 120 |               "    .dataframe tbody tr th:only-of-type {\n",
 121 |               "        vertical-align: middle;\n",
 122 |               "    }\n",
 123 |               "\n",
 124 |               "    .dataframe tbody tr th {\n",
 125 |               "        vertical-align: top;\n",
 126 |               "    }\n",
 127 |               "\n",
 128 |               "    .dataframe thead th {\n",
 129 |               "        text-align: right;\n",
 130 |               "    }\n",
 131 |               "</style>\n",
 132 |               "<table border=\"1\" class=\"dataframe\">\n",
 133 |               "  <thead>\n",
 134 |               "    <tr style=\"text-align: right;\">\n",
 135 |               "      <th></th>\n",
 136 |               "      <th>Time</th>\n",
 137 |               "      <th>V1</th>\n",
 138 |               "      <th>V2</th>\n",
 139 |               "      <th>V3</th>\n",
 140 |               "      <th>V4</th>\n",
 141 |               "      <th>V5</th>\n",
 142 |               "      <th>V6</th>\n",
 143 |               "      <th>V7</th>\n",
 144 |               "      <th>V8</th>\n",
 145 |               "      <th>V9</th>\n",
 146 |               "      <th>V10</th>\n",
 147 |               "      <th>V11</th>\n",
 148 |               "      <th>V12</th>\n",
 149 |               "      <th>V13</th>\n",
 150 |               "      <th>V14</th>\n",
 151 |               "      <th>V15</th>\n",
 152 |               "      <th>V16</th>\n",
 153 |               "      <th>V17</th>\n",
 154 |               "      <th>V18</th>\n",
 155 |               "      <th>V19</th>\n",
 156 |               "      <th>V20</th>\n",
 157 |               "      <th>V21</th>\n",
 158 |               "      <th>V22</th>\n",
 159 |               "      <th>V23</th>\n",
 160 |               "      <th>V24</th>\n",
 161 |               "      <th>V25</th>\n",
 162 |               "      <th>V26</th>\n",
 163 |               "      <th>V27</th>\n",
 164 |               "      <th>V28</th>\n",
 165 |               "      <th>Amount</th>\n",
 166 |               "      <th>Class</th>\n",
 167 |               "    </tr>\n",
 168 |               "  </thead>\n",
 169 |               "  <tbody>\n",
 170 |               "    <tr>\n",
 171 |               "      <th>0</th>\n",
 172 |               "      <td>0.0</td>\n",
 173 |               "      <td>-1.359807</td>\n",
 174 |               "      <td>-0.072781</td>\n",
 175 |               "      <td>2.536347</td>\n",
 176 |               "      <td>1.378155</td>\n",
 177 |               "      <td>-0.338321</td>\n",
 178 |               "      <td>0.462388</td>\n",
 179 |               "      <td>0.239599</td>\n",
 180 |               "      <td>0.098698</td>\n",
 181 |               "      <td>0.363787</td>\n",
 182 |               "      <td>0.090794</td>\n",
 183 |               "      <td>-0.551600</td>\n",
 184 |               "      <td>-0.617801</td>\n",
 185 |               "      <td>-0.991390</td>\n",
 186 |               "      <td>-0.311169</td>\n",
 187 |               "      <td>1.468177</td>\n",
 188 |               "      <td>-0.470401</td>\n",
 189 |               "      <td>0.207971</td>\n",
 190 |               "      <td>0.025791</td>\n",
 191 |               "      <td>0.403993</td>\n",
 192 |               "      <td>0.251412</td>\n",
 193 |               "      <td>-0.018307</td>\n",
 194 |               "      <td>0.277838</td>\n",
 195 |               "      <td>-0.110474</td>\n",
 196 |               "      <td>0.066928</td>\n",
 197 |               "      <td>0.128539</td>\n",
 198 |               "      <td>-0.189115</td>\n",
 199 |               "      <td>0.133558</td>\n",
 200 |               "      <td>-0.021053</td>\n",
 201 |               "      <td>149.62</td>\n",
 202 |               "      <td>0</td>\n",
 203 |               "    </tr>\n",
 204 |               "    <tr>\n",
 205 |               "      <th>1</th>\n",
 206 |               "      <td>0.0</td>\n",
 207 |               "      <td>1.191857</td>\n",
 208 |               "      <td>0.266151</td>\n",
 209 |               "      <td>0.166480</td>\n",
 210 |               "      <td>0.448154</td>\n",
 211 |               "      <td>0.060018</td>\n",
 212 |               "      <td>-0.082361</td>\n",
 213 |               "      <td>-0.078803</td>\n",
 214 |               "      <td>0.085102</td>\n",
 215 |               "      <td>-0.255425</td>\n",
 216 |               "      <td>-0.166974</td>\n",
 217 |               "      <td>1.612727</td>\n",
 218 |               "      <td>1.065235</td>\n",
 219 |               "      <td>0.489095</td>\n",
 220 |               "      <td>-0.143772</td>\n",
 221 |               "      <td>0.635558</td>\n",
 222 |               "      <td>0.463917</td>\n",
 223 |               "      <td>-0.114805</td>\n",
 224 |               "      <td>-0.183361</td>\n",
 225 |               "      <td>-0.145783</td>\n",
 226 |               "      <td>-0.069083</td>\n",
 227 |               "      <td>-0.225775</td>\n",
 228 |               "      <td>-0.638672</td>\n",
 229 |               "      <td>0.101288</td>\n",
 230 |               "      <td>-0.339846</td>\n",
 231 |               "      <td>0.167170</td>\n",
 232 |               "      <td>0.125895</td>\n",
 233 |               "      <td>-0.008983</td>\n",
 234 |               "      <td>0.014724</td>\n",
 235 |               "      <td>2.69</td>\n",
 236 |               "      <td>0</td>\n",
 237 |               "    </tr>\n",
 238 |               "    <tr>\n",
 239 |               "      <th>2</th>\n",
 240 |               "      <td>1.0</td>\n",
 241 |               "      <td>-1.358354</td>\n",
 242 |               "      <td>-1.340163</td>\n",
 243 |               "      <td>1.773209</td>\n",
 244 |               "      <td>0.379780</td>\n",
 245 |               "      <td>-0.503198</td>\n",
 246 |               "      <td>1.800499</td>\n",
 247 |               "      <td>0.791461</td>\n",
 248 |               "      <td>0.247676</td>\n",
 249 |               "      <td>-1.514654</td>\n",
 250 |               "      <td>0.207643</td>\n",
 251 |               "      <td>0.624501</td>\n",
 252 |               "      <td>0.066084</td>\n",
 253 |               "      <td>0.717293</td>\n",
 254 |               "      <td>-0.165946</td>\n",
 255 |               "      <td>2.345865</td>\n",
 256 |               "      <td>-2.890083</td>\n",
 257 |               "      <td>1.109969</td>\n",
 258 |               "      <td>-0.121359</td>\n",
 259 |               "      <td>-2.261857</td>\n",
 260 |               "      <td>0.524980</td>\n",
 261 |               "      <td>0.247998</td>\n",
 262 |               "      <td>0.771679</td>\n",
 263 |               "      <td>0.909412</td>\n",
 264 |               "      <td>-0.689281</td>\n",
 265 |               "      <td>-0.327642</td>\n",
 266 |               "      <td>-0.139097</td>\n",
 267 |               "      <td>-0.055353</td>\n",
 268 |               "      <td>-0.059752</td>\n",
 269 |               "      <td>378.66</td>\n",
 270 |               "      <td>0</td>\n",
 271 |               "    </tr>\n",
 272 |               "    <tr>\n",
 273 |               "      <th>3</th>\n",
 274 |               "      <td>1.0</td>\n",
 275 |               "      <td>-0.966272</td>\n",
 276 |               "      <td>-0.185226</td>\n",
 277 |               "      <td>1.792993</td>\n",
 278 |               "      <td>-0.863291</td>\n",
 279 |               "      <td>-0.010309</td>\n",
 280 |               "      <td>1.247203</td>\n",
 281 |               "      <td>0.237609</td>\n",
 282 |               "      <td>0.377436</td>\n",
 283 |               "      <td>-1.387024</td>\n",
 284 |               "      <td>-0.054952</td>\n",
 285 |               "      <td>-0.226487</td>\n",
 286 |               "      <td>0.178228</td>\n",
 287 |               "      <td>0.507757</td>\n",
 288 |               "      <td>-0.287924</td>\n",
 289 |               "      <td>-0.631418</td>\n",
 290 |               "      <td>-1.059647</td>\n",
 291 |               "      <td>-0.684093</td>\n",
 292 |               "      <td>1.965775</td>\n",
 293 |               "      <td>-1.232622</td>\n",
 294 |               "      <td>-0.208038</td>\n",
 295 |               "      <td>-0.108300</td>\n",
 296 |               "      <td>0.005274</td>\n",
 297 |               "      <td>-0.190321</td>\n",
 298 |               "      <td>-1.175575</td>\n",
 299 |               "      <td>0.647376</td>\n",
 300 |               "      <td>-0.221929</td>\n",
 301 |               "      <td>0.062723</td>\n",
 302 |               "      <td>0.061458</td>\n",
 303 |               "      <td>123.50</td>\n",
 304 |               "      <td>0</td>\n",
 305 |               "    </tr>\n",
 306 |               "    <tr>\n",
 307 |               "      <th>4</th>\n",
 308 |               "      <td>2.0</td>\n",
 309 |               "      <td>-1.158233</td>\n",
 310 |               "      <td>0.877737</td>\n",
 311 |               "      <td>1.548718</td>\n",
 312 |               "      <td>0.403034</td>\n",
 313 |               "      <td>-0.407193</td>\n",
 314 |               "      <td>0.095921</td>\n",
 315 |               "      <td>0.592941</td>\n",
 316 |               "      <td>-0.270533</td>\n",
 317 |               "      <td>0.817739</td>\n",
 318 |               "      <td>0.753074</td>\n",
 319 |               "      <td>-0.822843</td>\n",
 320 |               "      <td>0.538196</td>\n",
 321 |               "      <td>1.345852</td>\n",
 322 |               "      <td>-1.119670</td>\n",
 323 |               "      <td>0.175121</td>\n",
 324 |               "      <td>-0.451449</td>\n",
 325 |               "      <td>-0.237033</td>\n",
 326 |               "      <td>-0.038195</td>\n",
 327 |               "      <td>0.803487</td>\n",
 328 |               "      <td>0.408542</td>\n",
 329 |               "      <td>-0.009431</td>\n",
 330 |               "      <td>0.798278</td>\n",
 331 |               "      <td>-0.137458</td>\n",
 332 |               "      <td>0.141267</td>\n",
 333 |               "      <td>-0.206010</td>\n",
 334 |               "      <td>0.502292</td>\n",
 335 |               "      <td>0.219422</td>\n",
 336 |               "      <td>0.215153</td>\n",
 337 |               "      <td>69.99</td>\n",
 338 |               "      <td>0</td>\n",
 339 |               "    </tr>\n",
 340 |               "  </tbody>\n",
 341 |               "</table>\n",
 342 |               "</div>"
 343 |             ],
 344 |             "text/plain": [
 345 |               "   Time        V1        V2        V3  ...       V27       V28  Amount  Class\n",
 346 |               "0   0.0 -1.359807 -0.072781  2.536347  ...  0.133558 -0.021053  149.62      0\n",
 347 |               "1   0.0  1.191857  0.266151  0.166480  ... -0.008983  0.014724    2.69      0\n",
 348 |               "2   1.0 -1.358354 -1.340163  1.773209  ... -0.055353 -0.059752  378.66      0\n",
 349 |               "3   1.0 -0.966272 -0.185226  1.792993  ...  0.062723  0.061458  123.50      0\n",
 350 |               "4   2.0 -1.158233  0.877737  1.548718  ...  0.219422  0.215153   69.99      0\n",
 351 |               "\n",
 352 |               "[5 rows x 31 columns]"
 353 |             ]
 354 |           },
 355 |           "metadata": {
 356 |             "tags": []
 357 |           },
 358 |           "execution_count": 4
 359 |         }
 360 |       ]
 361 |     },
 362 |     {
 363 |       "cell_type": "markdown",
 364 |       "metadata": {
 365 |         "id": "nmJWytttASTi",
 366 |         "colab_type": "text"
 367 |       },
 368 |       "source": [
 369 |         "df.info()"
 370 |       ]
 371 |     },
 372 |     {
 373 |       "cell_type": "code",
 374 |       "metadata": {
 375 |         "id": "tbmUSMgdAXct",
 376 |         "colab_type": "code",
 377 |         "outputId": "cdfe2903-a07c-4f42-8fcc-3447330c518a",
 378 |         "colab": {
 379 |           "base_uri": "https://localhost:8080/",
 380 |           "height": 561
 381 |         }
 382 |       },
 383 |       "source": [
 384 |         "df.isnull().sum()"
 385 |       ],
 386 |       "execution_count": 5,
 387 |       "outputs": [
 388 |         {
 389 |           "output_type": "execute_result",
 390 |           "data": {
 391 |             "text/plain": [
 392 |               "Time      0\n",
 393 |               "V1        0\n",
 394 |               "V2        0\n",
 395 |               "V3        0\n",
 396 |               "V4        0\n",
 397 |               "V5        0\n",
 398 |               "V6        0\n",
 399 |               "V7        0\n",
 400 |               "V8        0\n",
 401 |               "V9        0\n",
 402 |               "V10       0\n",
 403 |               "V11       0\n",
 404 |               "V12       0\n",
 405 |               "V13       0\n",
 406 |               "V14       0\n",
 407 |               "V15       0\n",
 408 |               "V16       0\n",
 409 |               "V17       0\n",
 410 |               "V18       0\n",
 411 |               "V19       0\n",
 412 |               "V20       0\n",
 413 |               "V21       0\n",
 414 |               "V22       0\n",
 415 |               "V23       0\n",
 416 |               "V24       0\n",
 417 |               "V25       0\n",
 418 |               "V26       0\n",
 419 |               "V27       0\n",
 420 |               "V28       0\n",
 421 |               "Amount    0\n",
 422 |               "Class     0\n",
 423 |               "dtype: int64"
 424 |             ]
 425 |           },
 426 |           "metadata": {
 427 |             "tags": []
 428 |           },
 429 |           "execution_count": 5
 430 |         }
 431 |       ]
 432 |     },
 433 |     {
 434 |       "cell_type": "code",
 435 |       "metadata": {
 436 |         "id": "wSI27YVtAhxk",
 437 |         "colab_type": "code",
 438 |         "outputId": "f93acd80-e8de-452c-d9e6-0c16c1e1d131",
 439 |         "colab": {
 440 |           "base_uri": "https://localhost:8080/",
 441 |           "height": 317
 442 |         }
 443 |       },
 444 |       "source": [
 445 |         "df.describe()"
 446 |       ],
 447 |       "execution_count": 6,
 448 |       "outputs": [
 449 |         {
 450 |           "output_type": "execute_result",
 451 |           "data": {
 452 |             "text/html": [
 453 |               "<div>\n",
 454 |               "<style scoped>\n",
 455 |               "    .dataframe tbody tr th:only-of-type {\n",
 456 |               "        vertical-align: middle;\n",
 457 |               "    }\n",
 458 |               "\n",
 459 |               "    .dataframe tbody tr th {\n",
 460 |               "        vertical-align: top;\n",
 461 |               "    }\n",
 462 |               "\n",
 463 |               "    .dataframe thead th {\n",
 464 |               "        text-align: right;\n",
 465 |               "    }\n",
 466 |               "</style>\n",
 467 |               "<table border=\"1\" class=\"dataframe\">\n",
 468 |               "  <thead>\n",
 469 |               "    <tr style=\"text-align: right;\">\n",
 470 |               "      <th></th>\n",
 471 |               "      <th>Time</th>\n",
 472 |               "      <th>V1</th>\n",
 473 |               "      <th>V2</th>\n",
 474 |               "      <th>V3</th>\n",
 475 |               "      <th>V4</th>\n",
 476 |               "      <th>V5</th>\n",
 477 |               "      <th>V6</th>\n",
 478 |               "      <th>V7</th>\n",
 479 |               "      <th>V8</th>\n",
 480 |               "      <th>V9</th>\n",
 481 |               "      <th>V10</th>\n",
 482 |               "      <th>V11</th>\n",
 483 |               "      <th>V12</th>\n",
 484 |               "      <th>V13</th>\n",
 485 |               "      <th>V14</th>\n",
 486 |               "      <th>V15</th>\n",
 487 |               "      <th>V16</th>\n",
 488 |               "      <th>V17</th>\n",
 489 |               "      <th>V18</th>\n",
 490 |               "      <th>V19</th>\n",
 491 |               "      <th>V20</th>\n",
 492 |               "      <th>V21</th>\n",
 493 |               "      <th>V22</th>\n",
 494 |               "      <th>V23</th>\n",
 495 |               "      <th>V24</th>\n",
 496 |               "      <th>V25</th>\n",
 497 |               "      <th>V26</th>\n",
 498 |               "      <th>V27</th>\n",
 499 |               "      <th>V28</th>\n",
 500 |               "      <th>Amount</th>\n",
 501 |               "      <th>Class</th>\n",
 502 |               "    </tr>\n",
 503 |               "  </thead>\n",
 504 |               "  <tbody>\n",
 505 |               "    <tr>\n",
 506 |               "      <th>count</th>\n",
 507 |               "      <td>284807.000000</td>\n",
 508 |               "      <td>2.848070e+05</td>\n",
 509 |               "      <td>2.848070e+05</td>\n",
 510 |               "      <td>2.848070e+05</td>\n",
 511 |               "      <td>2.848070e+05</td>\n",
 512 |               "      <td>2.848070e+05</td>\n",
 513 |               "      <td>2.848070e+05</td>\n",
 514 |               "      <td>2.848070e+05</td>\n",
 515 |               "      <td>2.848070e+05</td>\n",
 516 |               "      <td>2.848070e+05</td>\n",
 517 |               "      <td>2.848070e+05</td>\n",
 518 |               "      <td>2.848070e+05</td>\n",
 519 |               "      <td>2.848070e+05</td>\n",
 520 |               "      <td>2.848070e+05</td>\n",
 521 |               "      <td>2.848070e+05</td>\n",
 522 |               "      <td>2.848070e+05</td>\n",
 523 |               "      <td>2.848070e+05</td>\n",
 524 |               "      <td>2.848070e+05</td>\n",
 525 |               "      <td>2.848070e+05</td>\n",
 526 |               "      <td>2.848070e+05</td>\n",
 527 |               "      <td>2.848070e+05</td>\n",
 528 |               "      <td>2.848070e+05</td>\n",
 529 |               "      <td>2.848070e+05</td>\n",
 530 |               "      <td>2.848070e+05</td>\n",
 531 |               "      <td>2.848070e+05</td>\n",
 532 |               "      <td>2.848070e+05</td>\n",
 533 |               "      <td>2.848070e+05</td>\n",
 534 |               "      <td>2.848070e+05</td>\n",
 535 |               "      <td>2.848070e+05</td>\n",
 536 |               "      <td>284807.000000</td>\n",
 537 |               "      <td>284807.000000</td>\n",
 538 |               "    </tr>\n",
 539 |               "    <tr>\n",
 540 |               "      <th>mean</th>\n",
 541 |               "      <td>94813.859575</td>\n",
 542 |               "      <td>3.919560e-15</td>\n",
 543 |               "      <td>5.688174e-16</td>\n",
 544 |               "      <td>-8.769071e-15</td>\n",
 545 |               "      <td>2.782312e-15</td>\n",
 546 |               "      <td>-1.552563e-15</td>\n",
 547 |               "      <td>2.010663e-15</td>\n",
 548 |               "      <td>-1.694249e-15</td>\n",
 549 |               "      <td>-1.927028e-16</td>\n",
 550 |               "      <td>-3.137024e-15</td>\n",
 551 |               "      <td>1.768627e-15</td>\n",
 552 |               "      <td>9.170318e-16</td>\n",
 553 |               "      <td>-1.810658e-15</td>\n",
 554 |               "      <td>1.693438e-15</td>\n",
 555 |               "      <td>1.479045e-15</td>\n",
 556 |               "      <td>3.482336e-15</td>\n",
 557 |               "      <td>1.392007e-15</td>\n",
 558 |               "      <td>-7.528491e-16</td>\n",
 559 |               "      <td>4.328772e-16</td>\n",
 560 |               "      <td>9.049732e-16</td>\n",
 561 |               "      <td>5.085503e-16</td>\n",
 562 |               "      <td>1.537294e-16</td>\n",
 563 |               "      <td>7.959909e-16</td>\n",
 564 |               "      <td>5.367590e-16</td>\n",
 565 |               "      <td>4.458112e-15</td>\n",
 566 |               "      <td>1.453003e-15</td>\n",
 567 |               "      <td>1.699104e-15</td>\n",
 568 |               "      <td>-3.660161e-16</td>\n",
 569 |               "      <td>-1.206049e-16</td>\n",
 570 |               "      <td>88.349619</td>\n",
 571 |               "      <td>0.001727</td>\n",
 572 |               "    </tr>\n",
 573 |               "    <tr>\n",
 574 |               "      <th>std</th>\n",
 575 |               "      <td>47488.145955</td>\n",
 576 |               "      <td>1.958696e+00</td>\n",
 577 |               "      <td>1.651309e+00</td>\n",
 578 |               "      <td>1.516255e+00</td>\n",
 579 |               "      <td>1.415869e+00</td>\n",
 580 |               "      <td>1.380247e+00</td>\n",
 581 |               "      <td>1.332271e+00</td>\n",
 582 |               "      <td>1.237094e+00</td>\n",
 583 |               "      <td>1.194353e+00</td>\n",
 584 |               "      <td>1.098632e+00</td>\n",
 585 |               "      <td>1.088850e+00</td>\n",
 586 |               "      <td>1.020713e+00</td>\n",
 587 |               "      <td>9.992014e-01</td>\n",
 588 |               "      <td>9.952742e-01</td>\n",
 589 |               "      <td>9.585956e-01</td>\n",
 590 |               "      <td>9.153160e-01</td>\n",
 591 |               "      <td>8.762529e-01</td>\n",
 592 |               "      <td>8.493371e-01</td>\n",
 593 |               "      <td>8.381762e-01</td>\n",
 594 |               "      <td>8.140405e-01</td>\n",
 595 |               "      <td>7.709250e-01</td>\n",
 596 |               "      <td>7.345240e-01</td>\n",
 597 |               "      <td>7.257016e-01</td>\n",
 598 |               "      <td>6.244603e-01</td>\n",
 599 |               "      <td>6.056471e-01</td>\n",
 600 |               "      <td>5.212781e-01</td>\n",
 601 |               "      <td>4.822270e-01</td>\n",
 602 |               "      <td>4.036325e-01</td>\n",
 603 |               "      <td>3.300833e-01</td>\n",
 604 |               "      <td>250.120109</td>\n",
 605 |               "      <td>0.041527</td>\n",
 606 |               "    </tr>\n",
 607 |               "    <tr>\n",
 608 |               "      <th>min</th>\n",
 609 |               "      <td>0.000000</td>\n",
 610 |               "      <td>-5.640751e+01</td>\n",
 611 |               "      <td>-7.271573e+01</td>\n",
 612 |               "      <td>-4.832559e+01</td>\n",
 613 |               "      <td>-5.683171e+00</td>\n",
 614 |               "      <td>-1.137433e+02</td>\n",
 615 |               "      <td>-2.616051e+01</td>\n",
 616 |               "      <td>-4.355724e+01</td>\n",
 617 |               "      <td>-7.321672e+01</td>\n",
 618 |               "      <td>-1.343407e+01</td>\n",
 619 |               "      <td>-2.458826e+01</td>\n",
 620 |               "      <td>-4.797473e+00</td>\n",
 621 |               "      <td>-1.868371e+01</td>\n",
 622 |               "      <td>-5.791881e+00</td>\n",
 623 |               "      <td>-1.921433e+01</td>\n",
 624 |               "      <td>-4.498945e+00</td>\n",
 625 |               "      <td>-1.412985e+01</td>\n",
 626 |               "      <td>-2.516280e+01</td>\n",
 627 |               "      <td>-9.498746e+00</td>\n",
 628 |               "      <td>-7.213527e+00</td>\n",
 629 |               "      <td>-5.449772e+01</td>\n",
 630 |               "      <td>-3.483038e+01</td>\n",
 631 |               "      <td>-1.093314e+01</td>\n",
 632 |               "      <td>-4.480774e+01</td>\n",
 633 |               "      <td>-2.836627e+00</td>\n",
 634 |               "      <td>-1.029540e+01</td>\n",
 635 |               "      <td>-2.604551e+00</td>\n",
 636 |               "      <td>-2.256568e+01</td>\n",
 637 |               "      <td>-1.543008e+01</td>\n",
 638 |               "      <td>0.000000</td>\n",
 639 |               "      <td>0.000000</td>\n",
 640 |               "    </tr>\n",
 641 |               "    <tr>\n",
 642 |               "      <th>25%</th>\n",
 643 |               "      <td>54201.500000</td>\n",
 644 |               "      <td>-9.203734e-01</td>\n",
 645 |               "      <td>-5.985499e-01</td>\n",
 646 |               "      <td>-8.903648e-01</td>\n",
 647 |               "      <td>-8.486401e-01</td>\n",
 648 |               "      <td>-6.915971e-01</td>\n",
 649 |               "      <td>-7.682956e-01</td>\n",
 650 |               "      <td>-5.540759e-01</td>\n",
 651 |               "      <td>-2.086297e-01</td>\n",
 652 |               "      <td>-6.430976e-01</td>\n",
 653 |               "      <td>-5.354257e-01</td>\n",
 654 |               "      <td>-7.624942e-01</td>\n",
 655 |               "      <td>-4.055715e-01</td>\n",
 656 |               "      <td>-6.485393e-01</td>\n",
 657 |               "      <td>-4.255740e-01</td>\n",
 658 |               "      <td>-5.828843e-01</td>\n",
 659 |               "      <td>-4.680368e-01</td>\n",
 660 |               "      <td>-4.837483e-01</td>\n",
 661 |               "      <td>-4.988498e-01</td>\n",
 662 |               "      <td>-4.562989e-01</td>\n",
 663 |               "      <td>-2.117214e-01</td>\n",
 664 |               "      <td>-2.283949e-01</td>\n",
 665 |               "      <td>-5.423504e-01</td>\n",
 666 |               "      <td>-1.618463e-01</td>\n",
 667 |               "      <td>-3.545861e-01</td>\n",
 668 |               "      <td>-3.171451e-01</td>\n",
 669 |               "      <td>-3.269839e-01</td>\n",
 670 |               "      <td>-7.083953e-02</td>\n",
 671 |               "      <td>-5.295979e-02</td>\n",
 672 |               "      <td>5.600000</td>\n",
 673 |               "      <td>0.000000</td>\n",
 674 |               "    </tr>\n",
 675 |               "    <tr>\n",
 676 |               "      <th>50%</th>\n",
 677 |               "      <td>84692.000000</td>\n",
 678 |               "      <td>1.810880e-02</td>\n",
 679 |               "      <td>6.548556e-02</td>\n",
 680 |               "      <td>1.798463e-01</td>\n",
 681 |               "      <td>-1.984653e-02</td>\n",
 682 |               "      <td>-5.433583e-02</td>\n",
 683 |               "      <td>-2.741871e-01</td>\n",
 684 |               "      <td>4.010308e-02</td>\n",
 685 |               "      <td>2.235804e-02</td>\n",
 686 |               "      <td>-5.142873e-02</td>\n",
 687 |               "      <td>-9.291738e-02</td>\n",
 688 |               "      <td>-3.275735e-02</td>\n",
 689 |               "      <td>1.400326e-01</td>\n",
 690 |               "      <td>-1.356806e-02</td>\n",
 691 |               "      <td>5.060132e-02</td>\n",
 692 |               "      <td>4.807155e-02</td>\n",
 693 |               "      <td>6.641332e-02</td>\n",
 694 |               "      <td>-6.567575e-02</td>\n",
 695 |               "      <td>-3.636312e-03</td>\n",
 696 |               "      <td>3.734823e-03</td>\n",
 697 |               "      <td>-6.248109e-02</td>\n",
 698 |               "      <td>-2.945017e-02</td>\n",
 699 |               "      <td>6.781943e-03</td>\n",
 700 |               "      <td>-1.119293e-02</td>\n",
 701 |               "      <td>4.097606e-02</td>\n",
 702 |               "      <td>1.659350e-02</td>\n",
 703 |               "      <td>-5.213911e-02</td>\n",
 704 |               "      <td>1.342146e-03</td>\n",
 705 |               "      <td>1.124383e-02</td>\n",
 706 |               "      <td>22.000000</td>\n",
 707 |               "      <td>0.000000</td>\n",
 708 |               "    </tr>\n",
 709 |               "    <tr>\n",
 710 |               "      <th>75%</th>\n",
 711 |               "      <td>139320.500000</td>\n",
 712 |               "      <td>1.315642e+00</td>\n",
 713 |               "      <td>8.037239e-01</td>\n",
 714 |               "      <td>1.027196e+00</td>\n",
 715 |               "      <td>7.433413e-01</td>\n",
 716 |               "      <td>6.119264e-01</td>\n",
 717 |               "      <td>3.985649e-01</td>\n",
 718 |               "      <td>5.704361e-01</td>\n",
 719 |               "      <td>3.273459e-01</td>\n",
 720 |               "      <td>5.971390e-01</td>\n",
 721 |               "      <td>4.539234e-01</td>\n",
 722 |               "      <td>7.395934e-01</td>\n",
 723 |               "      <td>6.182380e-01</td>\n",
 724 |               "      <td>6.625050e-01</td>\n",
 725 |               "      <td>4.931498e-01</td>\n",
 726 |               "      <td>6.488208e-01</td>\n",
 727 |               "      <td>5.232963e-01</td>\n",
 728 |               "      <td>3.996750e-01</td>\n",
 729 |               "      <td>5.008067e-01</td>\n",
 730 |               "      <td>4.589494e-01</td>\n",
 731 |               "      <td>1.330408e-01</td>\n",
 732 |               "      <td>1.863772e-01</td>\n",
 733 |               "      <td>5.285536e-01</td>\n",
 734 |               "      <td>1.476421e-01</td>\n",
 735 |               "      <td>4.395266e-01</td>\n",
 736 |               "      <td>3.507156e-01</td>\n",
 737 |               "      <td>2.409522e-01</td>\n",
 738 |               "      <td>9.104512e-02</td>\n",
 739 |               "      <td>7.827995e-02</td>\n",
 740 |               "      <td>77.165000</td>\n",
 741 |               "      <td>0.000000</td>\n",
 742 |               "    </tr>\n",
 743 |               "    <tr>\n",
 744 |               "      <th>max</th>\n",
 745 |               "      <td>172792.000000</td>\n",
 746 |               "      <td>2.454930e+00</td>\n",
 747 |               "      <td>2.205773e+01</td>\n",
 748 |               "      <td>9.382558e+00</td>\n",
 749 |               "      <td>1.687534e+01</td>\n",
 750 |               "      <td>3.480167e+01</td>\n",
 751 |               "      <td>7.330163e+01</td>\n",
 752 |               "      <td>1.205895e+02</td>\n",
 753 |               "      <td>2.000721e+01</td>\n",
 754 |               "      <td>1.559499e+01</td>\n",
 755 |               "      <td>2.374514e+01</td>\n",
 756 |               "      <td>1.201891e+01</td>\n",
 757 |               "      <td>7.848392e+00</td>\n",
 758 |               "      <td>7.126883e+00</td>\n",
 759 |               "      <td>1.052677e+01</td>\n",
 760 |               "      <td>8.877742e+00</td>\n",
 761 |               "      <td>1.731511e+01</td>\n",
 762 |               "      <td>9.253526e+00</td>\n",
 763 |               "      <td>5.041069e+00</td>\n",
 764 |               "      <td>5.591971e+00</td>\n",
 765 |               "      <td>3.942090e+01</td>\n",
 766 |               "      <td>2.720284e+01</td>\n",
 767 |               "      <td>1.050309e+01</td>\n",
 768 |               "      <td>2.252841e+01</td>\n",
 769 |               "      <td>4.584549e+00</td>\n",
 770 |               "      <td>7.519589e+00</td>\n",
 771 |               "      <td>3.517346e+00</td>\n",
 772 |               "      <td>3.161220e+01</td>\n",
 773 |               "      <td>3.384781e+01</td>\n",
 774 |               "      <td>25691.160000</td>\n",
 775 |               "      <td>1.000000</td>\n",
 776 |               "    </tr>\n",
 777 |               "  </tbody>\n",
 778 |               "</table>\n",
 779 |               "</div>"
 780 |             ],
 781 |             "text/plain": [
 782 |               "                Time            V1  ...         Amount          Class\n",
 783 |               "count  284807.000000  2.848070e+05  ...  284807.000000  284807.000000\n",
 784 |               "mean    94813.859575  3.919560e-15  ...      88.349619       0.001727\n",
 785 |               "std     47488.145955  1.958696e+00  ...     250.120109       0.041527\n",
 786 |               "min         0.000000 -5.640751e+01  ...       0.000000       0.000000\n",
 787 |               "25%     54201.500000 -9.203734e-01  ...       5.600000       0.000000\n",
 788 |               "50%     84692.000000  1.810880e-02  ...      22.000000       0.000000\n",
 789 |               "75%    139320.500000  1.315642e+00  ...      77.165000       0.000000\n",
 790 |               "max    172792.000000  2.454930e+00  ...   25691.160000       1.000000\n",
 791 |               "\n",
 792 |               "[8 rows x 31 columns]"
 793 |             ]
 794 |           },
 795 |           "metadata": {
 796 |             "tags": []
 797 |           },
 798 |           "execution_count": 6
 799 |         }
 800 |       ]
 801 |     },
 802 |     {
 803 |       "cell_type": "code",
 804 |       "metadata": {
 805 |         "id": "uAVPswFQTlC8",
 806 |         "colab_type": "code",
 807 |         "outputId": "63d84050-fc59-46b9-8cea-7a82448e4a8a",
 808 |         "colab": {
 809 |           "base_uri": "https://localhost:8080/",
 810 |           "height": 561
 811 |         }
 812 |       },
 813 |       "source": [
 814 |         "df.var()"
 815 |       ],
 816 |       "execution_count": 7,
 817 |       "outputs": [
 818 |         {
 819 |           "output_type": "execute_result",
 820 |           "data": {
 821 |             "text/plain": [
 822 |               "Time      2.255124e+09\n",
 823 |               "V1        3.836489e+00\n",
 824 |               "V2        2.726820e+00\n",
 825 |               "V3        2.299029e+00\n",
 826 |               "V4        2.004684e+00\n",
 827 |               "V5        1.905081e+00\n",
 828 |               "V6        1.774946e+00\n",
 829 |               "V7        1.530401e+00\n",
 830 |               "V8        1.426479e+00\n",
 831 |               "V9        1.206992e+00\n",
 832 |               "V10       1.185594e+00\n",
 833 |               "V11       1.041855e+00\n",
 834 |               "V12       9.984034e-01\n",
 835 |               "V13       9.905708e-01\n",
 836 |               "V14       9.189055e-01\n",
 837 |               "V15       8.378034e-01\n",
 838 |               "V16       7.678191e-01\n",
 839 |               "V17       7.213734e-01\n",
 840 |               "V18       7.025394e-01\n",
 841 |               "V19       6.626619e-01\n",
 842 |               "V20       5.943254e-01\n",
 843 |               "V21       5.395255e-01\n",
 844 |               "V22       5.266428e-01\n",
 845 |               "V23       3.899507e-01\n",
 846 |               "V24       3.668084e-01\n",
 847 |               "V25       2.717308e-01\n",
 848 |               "V26       2.325429e-01\n",
 849 |               "V27       1.629192e-01\n",
 850 |               "V28       1.089550e-01\n",
 851 |               "Amount    6.256007e+04\n",
 852 |               "Class     1.724507e-03\n",
 853 |               "dtype: float64"
 854 |             ]
 855 |           },
 856 |           "metadata": {
 857 |             "tags": []
 858 |           },
 859 |           "execution_count": 7
 860 |         }
 861 |       ]
 862 |     },
 863 |     {
 864 |       "cell_type": "code",
 865 |       "metadata": {
 866 |         "id": "SS3licyeTsjo",
 867 |         "colab_type": "code",
 868 |         "colab": {}
 869 |       },
 870 |       "source": [
 871 |         "from sklearn.preprocessing import StandardScaler"
 872 |       ],
 873 |       "execution_count": 0,
 874 |       "outputs": []
 875 |     },
 876 |     {
 877 |       "cell_type": "code",
 878 |       "metadata": {
 879 |         "id": "qLAyLS62UE8q",
 880 |         "colab_type": "code",
 881 |         "outputId": "155d8241-36cf-4a8d-bf68-b70c92562e27",
 882 |         "colab": {
 883 |           "base_uri": "https://localhost:8080/",
 884 |           "height": 34
 885 |         }
 886 |       },
 887 |       "source": [
 888 |         "df['Class'].unique()"
 889 |       ],
 890 |       "execution_count": 9,
 891 |       "outputs": [
 892 |         {
 893 |           "output_type": "execute_result",
 894 |           "data": {
 895 |             "text/plain": [
 896 |               "array([0, 1])"
 897 |             ]
 898 |           },
 899 |           "metadata": {
 900 |             "tags": []
 901 |           },
 902 |           "execution_count": 9
 903 |         }
 904 |       ]
 905 |     },
 906 |     {
 907 |       "cell_type": "code",
 908 |       "metadata": {
 909 |         "id": "lwi8m9vyUW7b",
 910 |         "colab_type": "code",
 911 |         "colab": {}
 912 |       },
 913 |       "source": [
 914 |         "df.dropna(inplace =True)"
 915 |       ],
 916 |       "execution_count": 0,
 917 |       "outputs": []
 918 |     },
 919 |     {
 920 |       "cell_type": "code",
 921 |       "metadata": {
 922 |         "id": "n9RYiRJnWCWe",
 923 |         "colab_type": "code",
 924 |         "outputId": "f50a8055-ab08-4521-d487-969817f9dff4",
 925 |         "colab": {
 926 |           "base_uri": "https://localhost:8080/",
 927 |           "height": 34
 928 |         }
 929 |       },
 930 |       "source": [
 931 |         "df['Class'].unique()"
 932 |       ],
 933 |       "execution_count": 11,
 934 |       "outputs": [
 935 |         {
 936 |           "output_type": "execute_result",
 937 |           "data": {
 938 |             "text/plain": [
 939 |               "array([0, 1])"
 940 |             ]
 941 |           },
 942 |           "metadata": {
 943 |             "tags": []
 944 |           },
 945 |           "execution_count": 11
 946 |         }
 947 |       ]
 948 |     },
 949 |     {
 950 |       "cell_type": "markdown",
 951 |       "metadata": {
 952 |         "id": "mhbUxkwB2fMd",
 953 |         "colab_type": "text"
 954 |       },
 955 |       "source": [
 956 |         "## **Exploratory Data Analysis :**"
 957 |       ]
 958 |     },
 959 |     {
 960 |       "cell_type": "code",
 961 |       "metadata": {
 962 |         "id": "RjlN6TNQvU2s",
 963 |         "colab_type": "code",
 964 |         "outputId": "6c4a6d7a-0e27-4e32-a8c6-b51fc041160f",
 965 |         "colab": {
 966 |           "base_uri": "https://localhost:8080/",
 967 |           "height": 357
 968 |         }
 969 |       },
 970 |       "source": [
 971 |         "import seaborn as sns\n",
 972 |         "import matplotlib.pyplot as plt\n",
 973 |         "\n",
 974 |         "correlation = df.corr()\n",
 975 |         "sns.heatmap(correlation,cmap='magma',linecolor='white')"
 976 |       ],
 977 |       "execution_count": 12,
 978 |       "outputs": [
 979 |         {
 980 |           "output_type": "stream",
 981 |           "text": [
 982 |             "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
 983 |             "  import pandas.util.testing as tm\n"
 984 |           ],
 985 |           "name": "stderr"
 986 |         },
 987 |         {
 988 |           "output_type": "execute_result",
 989 |           "data": {
 990 |             "text/plain": [
 991 |               "<matplotlib.axes._subplots.AxesSubplot at 0x7f776997e5f8>"
 992 |             ]
 993 |           },
 994 |           "metadata": {
 995 |             "tags": []
 996 |           },
 997 |           "execution_count": 12
 998 |         },
 999 |         {
1000 |           "output_type": "display_data",
1001 |           "data": {
1002 |             "image/png": "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\n",
1003 |             "text/plain": [
1004 |               "<Figure size 432x288 with 2 Axes>"
1005 |             ]
1006 |           },
1007 |           "metadata": {
1008 |             "tags": [],
1009 |             "needs_background": "light"
1010 |           }
1011 |         }
1012 |       ]
1013 |     },
1014 |     {
1015 |       "cell_type": "code",
1016 |       "metadata": {
1017 |         "id": "_irZZDNFs5AT",
1018 |         "colab_type": "code",
1019 |         "outputId": "969477b5-964a-4f4a-e991-3eb4481fd536",
1020 |         "colab": {
1021 |           "base_uri": "https://localhost:8080/",
1022 |           "height": 296
1023 |         }
1024 |       },
1025 |       "source": [
1026 |         "sns.countplot(x='Class',data=df,palette='GnBu')"
1027 |       ],
1028 |       "execution_count": 13,
1029 |       "outputs": [
1030 |         {
1031 |           "output_type": "execute_result",
1032 |           "data": {
1033 |             "text/plain": [
1034 |               "<matplotlib.axes._subplots.AxesSubplot at 0x7f7765d42ac8>"
1035 |             ]
1036 |           },
1037 |           "metadata": {
1038 |             "tags": []
1039 |           },
1040 |           "execution_count": 13
1041 |         },
1042 |         {
1043 |           "output_type": "display_data",
1044 |           "data": {
1045 |             "image/png": "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\n",
1046 |             "text/plain": [
1047 |               "<Figure size 432x288 with 1 Axes>"
1048 |             ]
1049 |           },
1050 |           "metadata": {
1051 |             "tags": [],
1052 |             "needs_background": "light"
1053 |           }
1054 |         }
1055 |       ]
1056 |     },
1057 |     {
1058 |       "cell_type": "code",
1059 |       "metadata": {
1060 |         "id": "rhW6SV0n313-",
1061 |         "colab_type": "code",
1062 |         "outputId": "f72e2f50-9e41-47aa-b9f9-6cab95e8d9ca",
1063 |         "colab": {
1064 |           "base_uri": "https://localhost:8080/",
1065 |           "height": 296
1066 |         }
1067 |       },
1068 |       "source": [
1069 |         "sns.boxplot(x=df['Time'])"
1070 |       ],
1071 |       "execution_count": 14,
1072 |       "outputs": [
1073 |         {
1074 |           "output_type": "execute_result",
1075 |           "data": {
1076 |             "text/plain": [
1077 |               "<matplotlib.axes._subplots.AxesSubplot at 0x7f7765d9c198>"
1078 |             ]
1079 |           },
1080 |           "metadata": {
1081 |             "tags": []
1082 |           },
1083 |           "execution_count": 14
1084 |         },
1085 |         {
1086 |           "output_type": "display_data",
1087 |           "data": {
1088 |             "image/png": "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\n",
1089 |             "text/plain": [
1090 |               "<Figure size 432x288 with 1 Axes>"
1091 |             ]
1092 |           },
1093 |           "metadata": {
1094 |             "tags": [],
1095 |             "needs_background": "light"
1096 |           }
1097 |         }
1098 |       ]
1099 |     },
1100 |     {
1101 |       "cell_type": "code",
1102 |       "metadata": {
1103 |         "id": "ETLR9Hupws-4",
1104 |         "colab_type": "code",
1105 |         "outputId": "0d2fe307-3571-41af-8132-f66cf5010ce1",
1106 |         "colab": {
1107 |           "base_uri": "https://localhost:8080/",
1108 |           "height": 386
1109 |         }
1110 |       },
1111 |       "source": [
1112 |         "sns.lmplot(x='Amount',y='V25', data= df, hue= 'Class',palette= 'Set1')"
1113 |       ],
1114 |       "execution_count": 15,
1115 |       "outputs": [
1116 |         {
1117 |           "output_type": "execute_result",
1118 |           "data": {
1119 |             "text/plain": [
1120 |               "<seaborn.axisgrid.FacetGrid at 0x7f776584fa90>"
1121 |             ]
1122 |           },
1123 |           "metadata": {
1124 |             "tags": []
1125 |           },
1126 |           "execution_count": 15
1127 |         },
1128 |         {
1129 |           "output_type": "display_data",
1130 |           "data": {
1131 |             "image/png": "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\n",
1132 |             "text/plain": [
1133 |               "<Figure size 402.375x360 with 1 Axes>"
1134 |             ]
1135 |           },
1136 |           "metadata": {
1137 |             "tags": [],
1138 |             "needs_background": "light"
1139 |           }
1140 |         }
1141 |       ]
1142 |     },
1143 |     {
1144 |       "cell_type": "code",
1145 |       "metadata": {
1146 |         "id": "JreC3JPcydgD",
1147 |         "colab_type": "code",
1148 |         "outputId": "d26ef9ad-b83a-44e0-f039-73af469d0c0d",
1149 |         "colab": {
1150 |           "base_uri": "https://localhost:8080/",
1151 |           "height": 296
1152 |         }
1153 |       },
1154 |       "source": [
1155 |         "cmap = sns.cubehelix_palette(dark=.3, light=.8, as_cmap=True)\n",
1156 |         "sns.scatterplot(x='Amount',y='V1',data= df,hue='Class',palette='PuRd',alpha =0.2)"
1157 |       ],
1158 |       "execution_count": 16,
1159 |       "outputs": [
1160 |         {
1161 |           "output_type": "execute_result",
1162 |           "data": {
1163 |             "text/plain": [
1164 |               "<matplotlib.axes._subplots.AxesSubplot at 0x7f776584fd68>"
1165 |             ]
1166 |           },
1167 |           "metadata": {
1168 |             "tags": []
1169 |           },
1170 |           "execution_count": 16
1171 |         },
1172 |         {
1173 |           "output_type": "display_data",
1174 |           "data": {
1175 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEGCAYAAACO8lkDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9a4wl6Xnf93tvVXXO6cvMzi6Xe+GKpEkpFNcxzYsMfRCtyJZJ04kphY5BBklM0TQhWLRhIHAQmUhsIBYsgYBlI0riEJA/0DDF+ItMATHkUIEVG5BlhaRJgbYsUqFIce87t+4+l6p6b/nwnKrununZndnpntt5f8Biu8+lqk739PvU+1z+f5VzplAoFAqFo+i7fQGFQqFQuPcowaFQKBQK11GCQ6FQKBSuowSHQqFQKFxHCQ6FQqFQuA57ty/gNHj44YfzG9/4xrt9GYVCoXBf8eUvf/lizvmRk557IILDG9/4Rr70pS/d7csoFAqF+wql1Hdu9FxJKxUKhULhOkpwKBQKhcJ1lOBQKBQKhet4IGoOhUKhcLN473nmmWdo2/ZuX8odo2kannzySZxzN/2eEhwKhcJG8cwzz7C9vc0b3/hGlFJ3+3LOnJwzly5d4plnnuFNb3rTTb9vY4PDyy+/jM0WFRUYiCZy4cKFu31ZhULhjGnbdmMCA4BSigsXLvDyyy/f0vs2subw8ssvo4MmdYkUEiRQK8Xlly7T9/3dvrxCoXDGbEpgGHgtn3czdw4Jwn5geXlJf9CjtWb3DbuopFhcWjC3c2Y7M+q6vttXWigUCneFjdw5GG9YvLjg6neusnh+wcGzB7zw2y+Qlon+oCctE/OX51x68RK34ncR+kA37whtOMOrLxQK9zsvvPACH/7wh/lDf+gP8a53vYsPfOADfOMb3+Dpp5++25c2spE7h9DKIp67w4U/LAPdfkdKiWpW0Ww3JBJXnr+Csorzrzv/isdcXlnSXmkhQ4yRarti+5FttN3I+FsoFG5Azpkf//Ef5y/8hb/A5z//eQC+9rWv8eKLL97lKzvORq5c1lpI1zyogCy/ON96FpcXLJ9fsv/MPqsXV1x+7jKLq4sTj9e3Pe2VlpQS7UHL6tKKvT/YY+/ZPbp5d+afp1Ao3D/8i3/xL3DO8ZM/+ZPjY3/kj/wR3vCGN4zff/vb3+aHfuiHeOc738k73/lOfuM3fgOA559/nve+97284x3v4Omnn+Zf/at/RYyRj370ozz99NP84T/8h/n5n//5U7nOjdw5MIOdJ3do91pykN3D1qNbmNqQ2kToArnPpJjwrSc9lMhkUkj0Xc/uI7tofRhXYx8hQ+gCYbVOKSV5fHlpiWtc2UEUCgUAvv71r/Oud73rFV/zute9ji9+8Ys0TcM3v/lNPvKRj/ClL32Jz33uc7zvfe/jU5/6FDFGlsslX/3qV3n22Wf5+te/DsDVq1dP5To3Mzj4zOTRCU9Mn6C72oECMzXkkAmrgK40ve8xykACv/Bj4bo76PAHHju1zB6aUTc12sjCn/zx7YjWWo7ZBSpb3Y1PWigU7kO893zyk5/kq1/9KsYYvvGNbwDwnve8h4997GN47/mxH/sx3vGOd/DmN7+Zb33rW/yVv/JX+DN/5s/wp/7UnzqVa7hnb2eVUu9XSv2uUur3lFL//WkeOycgQLfowICuNGRIMWEbizIKYwz9sif6SFgF/MJz8OIBfs8TVoHUJw4uHnD5hcvYxuK2HEoftovVO7X8dBVod8/+mAuFwh3m7W9/O1/+8pdf8TU///M/z6OPPsrXvvY1vvSlL40t9u9973v5l//yX/LEE0/w0Y9+lM9+9rOcP3+er33ta/zwD/8w/+Af/AM+/vGPn8p13pOrllLKAP8L8KeB7wc+opT6/tM6vraa/ef26a52rK6uWF1d0S97qvMVkwsTbG1JPhHagFaaGCL1bk1YBvr9noMXDpi/MCccBPqrPZefuUy2md037DJ93ZStR7dwUwdJgoStNnODVigUrudHfuRH6LqOz3zmM+Njv/3bv813v/vd8fu9vT0ee+wxtNb8o3/0j4gxAvCd73yHRx99lL/0l/4SH//4x/nKV77CxYsXSSnxoQ99iL/9t/82X/nKV07lOu/VVesHgN/LOX8LQCn1eeCDwL8/jYOnLtHPe8IigAI3cSSTxvqDdpqtx7aod2pyzsQQCUvZPTAFYwzdvKNf9tRbNba1EMBXnsn5CTlmfOexlcXNHDnnjRu6KRQKJ6OU4pd/+Zf5a3/tr/FzP/dzNE3DG9/4Rv7e3/t742v+8l/+y3zoQx/is5/9LO9///uZzWYA/Pqv/zqf/vSncc6xtbXFZz/7WZ599ll+4id+gpQkrf13/s7fOZ3rvJU+/juFUurPAe/POX98/f1/DfyxnPMnj7zmE8AnAJ566ql3fec7N/SsuI6L37rIxd+7SGoT2mn80mOcYed7dogh4hpJEWmr6fd7ur2Obtnh557phSkgMw1u6rDO0i07mllD81CDrjWmMsR5JPoojz/cMNmZnOJPqFAovFZ+53d+h7e97W13+zLuOCd9bqXUl3PO7z7p9ffqzuFVyTl/BvgMwLvf/e5binDKKlRW9Kse0xv8ysMEVFJoJQHBNpa+67G1BQuzh2bkC5nuakcKElSUUqyurqSLKfe0By07T+yQQ0YZha413aIjXU7Us8PCNUhnU+wimYypDa6+ebXEQqFQOGvu1eDwLPCGI98/uX7sVMjI4l3PakIXsI2VNtaU0FrjkyfFRFaZ1d4KLNjaop0mtAETDHbL0u13sruY9yiliD6yeHnB9JEprnaEvYBuNKEP7D23x+zhGdZZVgcr2kstMcouxdaWvJ1xU1fST4VC4Z7gXg0O/y/wVqXUm5Cg8GHgvzytg6usMNYQUqDaqghtkB0C0kKmjSanjHaavMiYZIgxkmJicm5CDBGFotlpWF1aHc4wJPmv2+8IWiau7dTipg7jDPMX52CgvdLSHXQoFKY2GGskNbXTSA2jtrKz6CNKKQlezozXn4LkFrWV60wxoY0+1i1VKBQKt8M9GRxyzkEp9UngnwMG+Ic55393eieQo5qJQSmFqQy+80zdFHpZfP3CU+1W+Nazmq/Yev0WyiiWV5Zoq6mmFcYZmvMN6WIidhFtNG7iWO4tUVnR7rXsNDusLq8IrdQoqu2KEIPsNqxCdxrjZNdChtXlFXZisU52KgpF6CWIaa3xKz/OUww7HUAK61OHa0p6qlAo3D73ZHAAyDn/M+CfncWxlVFUs4q4khkGN3FMzk+kCyl1MscQE/1+T7PbsOyW+JXHTRzKKPr9HoWifrzGTix+5bGNxTZSnFYJQghoq+nmHVprYheJfcQvPFuv36JqKvzcs7gokhzTyZRur5NdxEGHdocBKKcsU9gAWVJcMUT83EsgqS1kxsL60dpGoVAovBbu2eBwpmRIfSL4IHUEH8iLjNKK7KW2HftIvVOjncbWFoUihYTKCmUUoQvS7TTR7Dy+I8NybSCnjK0sYS2bkWMmk/G9l1bYIMFJG030kcmFCaEN9MseIqQsuxZbyYKvjZavK4heUltoyDET+oDvvOwqjFxn9LEEh0KhcNts7CoSfURphUKhlMLPPUorWcATNOcaTG3o5z0pJezEUm/XmMpga0u9W5NVpr3Uyu7AaWljtUZqBUZqBUorubNPENtICIEU0mHHU1ZMH5niV575pTmLiwuUUaCko2m8VqeIIdLPe7r9Dr/wY00iR5H96NtiVFQo3A/86q/+Kt/3fd/HW97yFn72Z3/2bl/OiWzkziHGiLIK4wzd1Y6sMpNzE3SjcVNHDJFqp5IF3ln0VMviX2nSPEmqJ2b8ytMddFSzivl8jqkM1U5FJkuaqRYpjn7e4696SRXNpI6RY5ZdxsxiGoNWMh9BkGBQTSuUUqScMMqwurIS8b+UyCkTQiAr6bpq91qMNSivCLNQUkuFwj1MjJGf+qmf4otf/CJPPvkk73nPe/izf/bP8v3ff2oiEKfCRgaHoQaQU2b6uqkMvGlN1VTE7Ui719LPe3Ye3xH5jBQxWhbo1Elw8K0nXAk0Ow3KKVSSTqHli0vpZlKSfiKAbSyThyZiR2oU7eWWarsSmfAI8xfm2InFNRKYtNHSxeQM7bwl60xcSNpqemEq54syRxGDfA7fe+qdmhRFVbaa3rzQXwoiFRJjxFiDbWwJLoXCmkvfusRzX3kOv/C4mePxdz7OhTe/dr/53/qt3+Itb3kLb37zmwH48Ic/zBe+8IUSHO4FssrSoTT3dH0nXUaNo91vib0Uqbcf3aabd/ilRxmF72T2QWlFCIH+oKeaVVRbFSklwiqMaaqwWt/Va4VvPTtP7NCcb1hdXOEqR32+Jqcsiq9L2UUoo1gtVlSzSgKSj2ir0UaTfCLFJJPXfSSFRLfsqOqKnGUHoq0WwyIrnwcZ5JZgsQpj0HGT4/LhOWXag3b0twghkHyi3q3LzEVh47n0rUv8wW/8ATlKLdIvPH/wG38A8JoDxLPPPnvMu+HJJ5/k3/ybf3P7F3vKbOTtodV2VGA1lZFA0ftxZmB2YQYa2kstfu7x+17e5+R9Gin+uonDt57YRmlpDYnQy1Dd7KEZfuXRWo+poMlDE5oLzaG/wzo4VDuVdDINgUIrcpY5C+ukMK2ypMH6eU8OGVc55hfnzF+cs7qyktRSZVBZjXMQOWeZuegCOUrHU3vQSmBaE/pwnfFRiono4x37fRQK9yrPfeW5MTAM5Jh57ivP3aUrunNs5M7BR49C0e63qKzIKdNcaKhmFdlLPj/OI8EH6llNP+9xE4eZGhKJ2IpKazWrCKtAu2hRrWKyOyFnqTegkOPlzOLighTFfjTlJMNrq0zMkcnOBGstySRmD83QTlJF82/POf+m81K41hpVKxIJN3OgIa4iVV0RjXQwxU7STvW2SIXnlMeU0zHSWhdqPQ9x3fNrTtLcGgbzQNppTWWue02h8CDhF/6WHr8ZnnjiiWMKrM888wxPPPHEaz7eWbGRwYEASismD02IPpJzJvmEQtJAMQRSl2l2Ghk0s1KQNrVh2kzHAvCwAJMlb9/Pe0xjZMjOikzH6uIKgGanIXSBvT/Y45H/6BHsliUs1p1LMck1rFUVY4i4yrG6uCLHTHfQifT31FLNKtyWk0G9lSd24lcdgrS1Kq1GH4kbiSqmJHWJFJPUPa5FyS7pKP2yP3S5Q1p9q+2qyJEXHmjczJ0YCNzstQ+bvuc97+Gb3/wmv//7v88TTzzB5z//eT73uc/dzmWeCRuZVjKVEXe3RS+yFBFUpegWUmOwbr1LiLIA55zRRuP3vHhFd7KYdwcdSsuEdQoyN4ECt+UIXWB1aQUZXC3zDmQw1tAf9CxfXIpLXCvzEpOHJtiJJfSyaNuZDNf1854UEylJqketV3NttaSgrKKdt1hnqaaV7IbW7bnW2esW/0w+lCxfBcIyyL+C4V+ChmqrOibFkbNc57XEtqSeCg82j7/zcWksOYIyisff+fhrPqa1ll/4hV/gfe97H29729v483/+z/P2t7/9di/11NnI276cM9uPbbO4skAlRbVTjekcW8n08ezCjEjELI20pk4t7bKlaipZiLcrWYTX0t7aapSVBXl5aYmtRP7CzRxplehXPZNzE1DQ7XWENrB4acHW67dAi8Xo9NyUzkrA8Z2XWoNWuKnDr7ws5BHUZdn1TB+eiptdC7ay0sWk1Vi0zlkK79HL7gYl/zDHaeuBBNV2NeozXVeIvoHm7bDTKRQeVIai82l2KwF84AMf4AMf+MBpXOKZsZHBAcQidNgBqKSYTCa4maPNLf2qBwurFyStU2/XZDIaLYuuzvQv9iitROrCSkdRfa5GVxrTiFaTsYbcZxJJvB6cptvrqM/VpCuJSTNhdXWFRqMbTbVTHYrurWQWo96t6a52MgfhJFDlnOmXogRbzSqq7XWtZB6oHqro5zIMZ6zUBDIZ17gxyAw7Ea31Ycvqehr7JJQWocJri9Sl5lDYBC68+cJtB4P7kc0MDgnmL83RWmMbS1xFDlYHPPzWh+kXPVqLD8P0YZlcJjHqLeUk3UTVViWS3UZRb9ekPo4dQX7hqbdrVv1KOpOMltmKPhLbSOwizjm6RUc1kV1LyoluvwMF9W5NvVPTr3rpYCKjnHQwDZ1FqysrUp+wtcVOLbuP7461i2HSW++I54RCifJsY8cp62EnYWozXuMr4WaOfCDdXCCBoYj8FQoPLhtZc0gp4SqpAyxeWtDv92NB2U2c5OVXieQTWq1lsb3IXdgtmXrOKdOvepReTzav0zFu4phcmOA7z2RncjjR3Ab83FOfq2XYrDEoI4NyburGOkFsIzlkupVMXk/OT9h6/RbNbjNKavjlukU2Sx0i9UmmrrPsUgb5jWsL0mElMwxDNxUZYhcxtRn1okIXTuxg0kbTnGtodhuacw31dl0kwguFB5iN3DkYa4g5jj4LQ6pmmHzOUdzZ+mUv6ZiZk7v0ypLJ0mVUJeppTUpJhufaMMpqDDl+Pz8servGSUDQivNPnWd1eUV9rhZToDZAZryOg+cPqLalTTVG6VyKfcRNHChGY6FKVbRXWiYPTVheWpLC4ezG5PwEm+yhpDcSLBRqNC4aJrbRUgcZdgUoqLfqE9NGRwfoCoXCg8tG/qXHHNl63ZbITiw9KYiJT1bSmmoquZM2xtDsNMQustxbsnh5Qb8n9YgUE9atB+HmPTlnYkjMX5iLWdBaNTXHjF94fOvHQrabOaqtCmMMJFm0U0jkNksHkJUCc7vX4qxj/pKYBDUXGgkwSklRexlw224sQJORTqiLS1Z7K1b7q7FOYGpzbLHXel1EV4rUp8PAAHKcZRHxKxQ2mY0MDkYbur11G+r6bn51ZYVCUU0qQhdEX2nZE9pACAGjDf7Ay/eteDX4VgKLqS0oJcNoQdJC1bYUl7WVQnM1k9qCsYa4ijJrgRLZcDR2Ysk6j7IasYuQoF/1zB6dYY1l+dISow0pi0osGrYf3UYZRUrSdju03qY20e11rA5W+N6DYZzXOIp2+sRupJzy+F+/6qX1tQs3nJ0oFAoPFhsZHDKyQ2ivrLWU2jAK5ZnaUO/W4trWVAS/FrHTcvc9aB35lSerLHMNg29DzDI5PJEBuawkVYWSjp8UkqSkckJZJYN3WqGsotltpDbRBYwxsutImXpXDIXsxMq09WUxB5o9POPc95yT81lzqOlEJgdJTy0vLlleXNJebVleWbL/wr7UOyYiO+5mjnqrPjlVtK5ZtHstYSm1iH7e39ZkaKFQED72sY/xute9jqeffvpuX8oN2cjgQIYc5A54EK0b/BVWF1f4hcwUxBzHIbLYyaKayVRb1bg7mD48pW/70W5UaSXzCSg0mnq3ZuvhLRlyG3yqD0TMTzuNVpLa0Wi2Ht5i57EdspKuoGqrEqXVdXHcWCNzGG1kdXHF4qXFWFCutirsRAKFnVr6lQy5DUX3+bNz9r+7z9539mivig7TYDtK5nhaCaim1RigjhK6MGo3FQqF18ZHP/pRfvVXf/VuX8YrspEF6Zwy9VZN6kQoL5GYnJ+AXtt7Gi2zBl6G4tw5N9px2sayurqi2WnIMUthebuWu/UogaZbdlhnxUmu1jIf0a5TQWrdNdSnsSCsrcZ34vfgtsRn2lhD81CDcw4i9Iue6cNTli8vZTBuS66p2+/AQrPVsF1tY4xhcWlBN++opzUkaPda2RU5Q+ol3aQrjVFHahBGH9ZanBlF/k4iJREfLBQ2gfkX/i1XPv1F4vNXMY+d4/xf/1G2PvhHb+uY733ve/n2t799Ohd4RmxkcIgxknVm8shEdgxWj94KVVORUx5bWkGKvCorunlHv5Q5iNAH0iqJr7RW8rjRIqlhFduPbstxaidFbCMmPjll6X6KYjla79YSKBS4iTv8unE0Ow2uklkC1zgOXjgQaYu1zahSip3Hdrj87ctcfu4ybtcx2ZnItLPWIv/R+1HWO/UJ1Rymt8y2OVZvyDlLQFmjnYbumh+eOhyuKxQedOZf+Ldc+hv/lLySdGp87iqX/sY/BbjtAHGvs5G3f7ayaDTLy0u6/Y7lpSVaaTDIBDEJO7XSltpF2nnLck8kMSbbooGknQy2ZfJYtyAhtYqpuL3ZxrK6vAIjC7+tLCordKWpppVYkVqDbkTUT1txoqt3a5rdBldL66p2WorlWxWudjjnJCgpRbaZ6UNT6q0alaSOMch9KKuomkoMiZQ6FNxLiaykvuE7P3Y0XZtaGixRR9ZKs2W+obApXPn0F8fAMJBXniuf/uJduqI7x0buHEgitLf9+m0R1dPi9paipH78ypOd3OHXOzXdXjfOGGQlgnuDe1pjGpptUVwdpLC11XLXP3HkkMV/eqsmhSQT0Y0EFr/0MpCnHWqmiEvxmLbagoF2/9CERznRbVLIXILWsnMwzqAfErMiv/KEXjqrqpl0SNnK4nuPn3spgFtxkPMLz3x/jptKoNGVZrI9kfc7aZdVSgKSbSwpSc3jfgsMOeVROLFQuFXi81dv6fEHiY0MDjGJuml3tZPuHWPEByHKXfpkMsHPPW7qsLVlGZdUk4pmtxkDwCDSp4wi9IHYRmxtiT7iW89kdwJGOqP6RY+xZhy6czMnHU7npWtqtb+ivdxK+6oCHCxeWjA5NxEpjJUfpcWDD9IOmyLTh6ZUW5UUz7ci7X4rO591qsxORKepntWHHVU5Y63saAbVWQDda6y1UoDW0Gw3YxeTtvq+rDH0C2m/JUttp5pV0j1WKNwk5rFzxOeuDwTmsXN34WruLPffX/wpYLVl8dJCdggx4zvP6mAl+kEhs7i0EMntl1csLy6ldbXzzF+aS7fQMHSGqKlmlUeZa9tYJrsim9Hti/pq8pLO8QvxW7C1HRfmftGP8xM5SVHbLz05yKR16KR4nWMW8T1rxpbZwcxnkAw3xjA5N2H3DbtyDi/GRN1+J14MUeonKcnAXI7ruYosAdO3nn7Vy+dd3t8tq77z4+Q5yGftF32Z0yjcEuf/+o+iJsc1xNTEcf6v/+htHfcjH/kIP/iDP8jv/u7v8uSTT/KLv/iLt3W8s2Ajdw4hBiYPTchRWka1XhekvdxtpjZhakO2UnMY0kN+5fGdiOqFVRjnE/p5j56tVVOzvCf6SL/XYyvL9KGp+DdvicBdTpl2v5VA0HoWlxcyLa3FDhTWU9MxHZPWSCGJeF+U4GSswRiDdlLDSFUSF7j15+oX/TjIlnxi+vAUO7FjsBvSa+1eK5PeePzCHzrZ5eq+9ZFO/fXttjnKz7DsHgo3y1B0Pu1upV/6pV86jcs7U+5KcFBK/RfA3wLeBvxAzvlLR577aeAvAhH4qznnf37a5zdm3dI570a/5uZcI7IYSZROMbIriClisiG0gdmFqTi8+bXMhjLEGJk8PGH+/Fzu8r3cldvGQpDW2LgVRbBPiyDeULzuFrKz0ErTL/pjw3ZZrY2A1tIWgwbSECS0Ef+JQalVGfFyyIhsR7fXiW9DFi0mbbV4T88cy5ekHTalRAhi4jPMZwySIm7m7tvAANzw2u+3mknh7rP1wT/6wHcmncTd2jl8HfjPgf/96INKqe8HPgy8HXgc+DWl1PfmnE/fckwxTjtrp8VLOuex6FrNKrEABXKf0VONcpp6UhNtpN6pWV1dsbq4wmBEqK42dPsd0Ue6vY7moYbYikT2oMBaTSpSTCyvLFleWY6GQXZiMY1MYA86SApF7KIEHERKgyzBTSHT3IMzHEpktbv9bgwmxhqZ1wgR24g8h5s6mnON7HzW6rIgwWVIuQz+DSmm0RjIVOa+KuqaWqbMj7bqDhPuhULh1bkrwSHn/Dtw4t3dB4HP55w74PeVUr8H/ADwr0/1/Ou8fr+SmYXYyR32VtySXcO6l18pJYqrq4hyCq20FH2XHoL4LrhzjpQTqUtiP3pk4jhFkdJ2U4erHKaSBSv4MPpEZCTtU80q6u0aW1nq7Xo8zuBM51uPzVYWvfVuQ1f6WNHYNU46ktbT2vMX5lKYrkTwr5pV5D4zPT+ld/3oT9G3PZNtSbPBOjg4Q7vXjourX3npXLpPPKONkyaD0IoelHa6+E8URnLO9/XO+FZ5LbW2e+0v/QngN498/8z6setQSn0C+ATAU089dUsnUUqRlKRmhkLu4IGQoqSVQgi0V9qxiOxqx2pvJcEjIF1LXmoIrnEEZBGqd2vaS620g2a5A3eVIyUJICEHuWvXihila6pqxKLTVpbmXHNd0dtYg9kyqF35x+wbLxLijfz6tNHSaosEiaHo3ew2+KWXc8wqqkklw3qVFknw9a6gOdeIvtTak2L0eziats+Iv/Y1wWEYsLvRHXkKafSIUEaOfVRGfHxdTGOX1Wnd3Q+T3oXCUZqm4dKlS1y4cGEjAkTOmUuXLtE0zS2978yCg1Lq14DXn/DUp3LOX7jd4+ecPwN8BuDd7373LYXFrLOorxJQtSKrLAtiRhbBlWdybkJ9vsZYg+9F2iK0QRbuFMci8eCk5mqZqLaNRSuNdZZsRJhPGRlOSyaJDhOKHDOTnYkMoGnASNpjwE0d3bwbF2jttAy6aUWT5JccvbS+DnMJgHQxdTKzUU2rcRcx3DU350T2Q1kJXKY2uImko9p5KzMS2yI4OKashp/bemZgONegApuyXKStrPhWrBfkFMTrIoU0OunZqaXZaaS+sqZfiPotID7XE0s1qSgUzoInn3ySZ555hpdffvluX8odo2kannzyyVt6z5kFh5zzn3wNb3sWeMOR759cP3a6JBkiiyvp/NFGun2yymQtQ3FhFVhdWWEqQz2rJa1k9Xj3W52rxjvwFMUIyE6lRbWaiVdDjpl5mMvCp0D1Ul/QtUYF8VHIIWOmMomcQ6bf7/HOU2/VTM5NpN6wDgADQ1H12PTykefqnVrmIYIs/MPks60tbipBIgVRhNVGpMdzOpTOiF0c7VCPMkh2gOyc/NLjvae9JKk2MzE0uw2zh2ZUs2qcMfArPwodhmXAWy87pdoeTpcPZHmNdbYYCxXOBOccb3rTm+72Zdzz3GtppV8BPqeU+rtIQfqtwG+d9kkUMlhWbVej6qpfeVRSTM9NWV1d0S7aw84iBdPtKdW0krmHPtHTM1hJfLUAACAASURBVGkmEli2K5KXuQZtZdHzrdQMtNHU27WY6xiR6HCVIyxFxgLD4XxDK+minMX0Z/rw9EQ3tlf9fOs0GGuZpLHQvPacgOPBJnbX1/tzkkA5dDuhOAwsUTq9+nnPam+Fn3s59oG01y7VUgrgSVpqh8BweHA55zCrcRIxxBIcCoW7yN1qZf1x4H8GHgH+T6XUV3PO78s5/zul1D8B/j0QgJ86k04lLUNkixcWkhLRsP367dGTud/vJdUTkSJsbdFGc+X5KzjjJGeekYKyTxAluPQHomLqJg635cZdxNgxkzn0gKjdWPweFFq7eYe18ivplz1uIfIdt4tS0pXUzbtDZzhnDnWSrlm7QxeIfaSaVSLvMXWye9Lr4+x34lu9kkGz6ONhd1Uv3VUxRBHuWzH6VQMoKy25Q1C+UX3hpLpEoVC4c9ytbqVfBn75Bs/9DPAzZ3sFin7VY7csRKkJ9Mseo2VuwTRG7Dq1FGG1WdcbjCV1iegDxmnoPeGqdAf1qx7lFCorQhfka6NwU7HxzEHy/HZqsZWVHQLSRRNtJCzCsXl1pdU4AX0avfl+6ccCNEDsI15JB5JpDGmxnvhe1wd0LYVto2UmJE/kOnwnKShl1agOS5IOsMHCVCk12pDmqcxdpF50ndzEjT7WsG6RdXps1z36WKFQuHvca2mlO8IgItcuWim6einUJkRuQlsNFaMtZrPTsLq6wk0d7WJJWgVUr/DOiJxFL3pHfiUCd1mJVEO1XZGjyHbrqegT2UY8IQbPZ5BupNhFUYZdMxanr4kLOa2L3K/QZRE6aZXNKUuReOqk5//a1/WBCtkZDS50vvcyaNcnOt/J/IUzxBBFEmSvkw4oo0YxwRRk96WdfL56px7TVtVWhakNvvVS31lbog7pMqXkOIPftjISODahi6RQuJfZyOCgtSb4gLWiNqq03O1rIwVnrcX4xm05mq0G33maHVFeNZUjhYSbuXWxtcdtOZSxLF5cjO2YqRYJjmZH2kSddthtO/o/DIJ5KSVc47DN2rgnI0Y81hxbJFNI9MteFuK134ObXN+3H/pwaNKjOGwj1WqcYxgYjq2UEjXZtclR7qXWkLPMg+htCVr9vB9TRDlkQgxMH5qOirXGyTBgvX08FWacDNQln4gxyk5idthNpbQqMwiFwj3GRgYHpRXNrGHRLcghoytpE80+j5PR9aQmxkhMor5qGun88bZl5/Vb+HkLPqGNwlSWxdVOpq6tmOyouO5McjKDMKiCDtjKXjczYBuxAM1ZWmSH1EvOUi8YF/f1zIEy6rpjxE7eH/soEh9G9J7s1BJXx8s3gy+1X4jUd+ik3VRZdVhEXgeJLL27MjndpHFwEGDnsZ0TO6dAgppvRWAwxTReg5970kOJZrcpu4RC4R5kI4MDQPAiZa1QpJzou54YRDCvntbY2rJ6eUVYBCkeByspksoSD1bktSCenjbo2qG07CCyl1mAofhsjBlz8EfJad2d5NOo5npSwABZYK+964d1x8+1Q2lk+mVP9nn8nNFHmp0Gs2XGzqQh+PTLtaz1cK5OdjxYxjRPvVMfq3u4RoyLUkoy1X2DwOBbKdJ3+514WneBalpJ8Tor2ZVYqU0Ya0p3UqFwD7GRwUFlmUto99pxAXQTR07SQtoj6ZN+X3wYzMSMaRm3XdMrRc4aU2nM7oQUMyklqqbCRy8pq8oyPTc9nEmYHP9R94v+eIF4vYCftEAqJYJ6OcjCn7ykrgYjnmMzEEqNgeHwQQ7TV9cs5EevwThDtNJtVO1UKNTYmgtr/aWjEhvW3DAdNAj4+daP7bOpTwQTxhSUb/0o1eHxMvw2LcNvhcK9wEYGh3Qk167cETXSlae72pG3s2gVTdaLaT7sWhqKroOlp0LugOutWlIxGpxyNLOG5lwjHgy1OXaHH308tigDMvzVBypbHb5mPYymK+kI6g46+kU/vsXNHN1BN5oCwboovDb5IUuaa5z+PoEh8AxfV1NpXzWVua7uMWgVRR/lucbesJMqRdntJJ9G0UE0Y+qqmsmMydH3hzaMbcOFQuHuspHBQSspSA9DXTlmYo6opEZ1UpVlN5FSws+9FGK1WHemTrpzjDHYiXQf6UqsP1OQxdA1jsnO5OQLuJHYx7qbM/pId9CNrwteLEiVUaN2knYiGGisGd3pAKyT67nW+/lGw3S2sYcFbNYT1rP6xDt4bfSxusmxj3RNF5U2emzNHVpX01TmSKxba0IpTcrp0GVuLVg4BIez0FsqFAo3x0YGh+QD9aRicXlJjpmwCtTnarz3VLuVtGeuB+IA6nNiIQrQXe3GyedgAlhZkHPImNnhDmEw/gntemFXh3UF7dYL5zV+NLqSBfBosffwmhO60rh0mMbJ5OtaXZWW1tB+2Y9dSm7qbri4DoN4sY2iFVXdOFV04s8yJvzCj5POgy7S2JG1CrIDinlUSh12I37ppTPqyL/C4Tr7ZX/o5KYkiJWUU6Fw59jI4KC0JltothqijzKPEPKhqJ5dL9AaNJrsxSTHaMPKr6R1dfBaCAlzweCXnvagpZpV4kNdT6SusC4AZzK972FbOpXqrVqc2mIeF78hsFwrr6sQ8TwUJHU4HzG8/tqitKkMk2py0wN0NyqE3wz9QT/WFEB0kbSWOoWrHVuv36KfS30ldIdpoyEQDX4SID8DbbTYo66u0VtahaKyWijcQTYyOCQSqU20e+3onlZtV+LbYBXNtCGFhJ1Y5i/ORUU157HmMHTXxE4kIlKQKWmiFF2rqdwx5yzdSkcZlF2NE7/nFNPY6w+HSqvXLuy2kcExZdR43qGAe6MAcNauZymkY4FhIPbxcALayucEaK+24+sVa4OjtQGPsYcL/0nHHM5XgkOhcGfYyOCgA6Q+4GaO2ImnQlxGqgsy0Wwmhv6FjrCUdNCwIGmnqWtpf1VG0iZkGarrV5K3V0bRHrQYY4g+Mjk/OdbGevROGQ7TKIP2UfIiqzEY8VhrxxkJ44ykVl5lQvqOcaNLuMHj1VY1zjuAtMSelCq6ocWnuQc+c6GwIWxkcMgpowC/9hhIPo0yD4uXF4QQaM7V9HMvcwpaSa58YiVgWNEUUlbc4VQeVOTWQ2hIgMkx08/70YcBjns2HGWYeYB1GslJGqnerqUIPXRY3UMeyNrIJPm1nVc3mnvQVtPsNtftlq7lJL0l7XTZNRQKd5CNDA5aA0d8FapZRdZ5DBKxC/gQqHdq6VwyChK0e+0oHKe0TCcrpUaLUGMN/aIfW0eHnUWK4jZna3vDYu/g+naMtVz2PbFLuAHVrMIbEfXTWmOaV68LvFrnkVISjEMXDgULq6K3VCjcSTYyOKA1OiviMpCyTCrX2zW+E+E8Uxucc9iJuKqhpHvGrzxxFWnONSJL0WfcedFMCr10JZloxGWtXs8BrLuUBk2lk0TxBtOdo3fKwHVzAMAxye17AaVlNoLpKR9XFb2lQuFuspnBIUmbqmksBogx4hcirteFDlNp4soTu4Bau59NdifjDuJom2kOmWq3Ii8yMUn9IoVEsjIEptxaZVSL18HRmYJBFK/ekfbO0Idj7a12cjhklmI61hl01Db0QSTnTPTSDTakr8rOoVC4c2xmcAgZO7GkNhC6OFprKq1E+iJbaXc9EgWijyIvsRbGG7CVFeG6ZSAuROtoWNRiiNS79SiJcVTD6Ohxh0GvQfmVLDuDo4NrfuGPdfEkn/BL8WM4TXIW6e5R86m+O3adR9uAYR0Mt+sSIAqFO8RGBgc1cfh5T8wid5ERQ5rGNUzPTdGVYrXfohS4STUKzJla3M5sZUcvaZTIWsQgcxE5Hba8piTtsrGLN5wsPsrgZX0tOecT7TSDFz+G08Qv/LEgFrpwQ82nsyL28Trr0uTTsRbZQqFwtmzkX1oOSZzcciZlhTYKtMwyuB2HUYYYEqmNZJWl1TVEJucmmPNmvJtOXjqdQhdGr+lh8ernvXxdSbG5O+hENvuazp5hbuKVGDycr52aPm0rzRTT9bubLAFi0Hy6E6R0gzmHG8w/FAqF02czBWt8JC06ZORYXNuIMqQV5oHF5cV6lkGzurwCDdVUFFBnD81w9aEmk1JSfFZWagpkWdy0k1ZXYyU1lFNGK42bSQE6IcY3Oa4ltvPJgkspih+CtsfTXHC90uvtcu0Mxqs9flYMP7NrKZLehcKdYyN3Dip4nM4EY+iuLlC1yDh0V1ZUswaMmNHUuyJAl3Mmx4yfexZpIXWFKBIPg/SEUhIIdKMx2oA/YaZh7eCmjRZhPSMLb1itC9Nbxx3UQhckcA1rc5ZjDgHptDuWtH1lzac7hbZrZdkjEhq2fu0SH4VC4dbZyL82naCu5G4/bNdklQk+oLQECdMYjJKZhXq7lmE3p4khontN27akkFjtrainNW7icFNHs9vQ7DaAqLcevdHXTtRUu4OObtFJYKkP1VNjF0mTQ0XSnGVHcfQYSq0ns89IgE4pUWTtFp0ECHV7uku3QzWtxFAoptHzoVAo3Dk2MjjkxmL6SPWoY//FfYyzKGMIIcrinmG1v6KaVfTznsmFCaENNOfET9pauau11uJbj20sfuWptqsx9VFvrb0PokhUu6kb5TFIQBIxOTSjhHVOGdZrYI75ujt4uMGw3DVEH/FLP8pfu4k71vk01hbS2v/hSJHXVIaJm4he1Hr+4m4xuMQVCoU7z0YGB73V0FcWYzW73/PQOFhWpYTK0M47kcfQipwkpVRtV1IAtoeqqUorsRAdpLiPLGSmOt6KmkIah9yOvi71CRyjV8PAIAg4Dr3Ztd3oqyzWOeVjXhApJLp5R7PbyKBdTLKrGWJMJ6852k11VE+qUChsJhsZHLAGXn+OxUsLQhtJGibnJ7jKklNm1lhiGwlBagrNdkO9VRO9DLFpo6WtcipBwVjRX7rZ7iFtpDA9dDhpI5afQw//MACWYyYsZe4hGGkptc0r/8qCD9ebCWXZTWij5ZzXbD5CF0a57Jtl8Ml+UIfwCoVN564EB6XUp4H/DOiB/w/4iZzz1fVzPw38RcRe56/mnP/5aZ8/p8zq+X2o7HoxzawuJuqthvp8Q7tcEfskFp/r2gRIMTlvZUIMOO1QRu6wU0porQl9GG08bXNcR0lbEY47Kn+hncw1uMnh63zr6Zc9/VzMeuzMSufU2tPhdtM8J3Ye5eMprVfDr8Qbmiw7Gje7sZlQoVC4P7lbf9FfBJ7OOf/HwDeAnwZQSn0/8GHg7cD7gf9VKXXq+Y246EkhERa93K1njTaWySNTVlcWTM5PqXdqqt0KO7E0O824MFfbldhoblWimNpIXtwvPe0VKVTnmPGL9QJ6hGqrkvy+ljRSvVWPgSEncY1rr7Z0+x3dfkd/0NPvy05FWz0GllfCOnv9b1UdGgJpd8KvXN98m2joAn7pRxmRYTdVKBQeLO7KziHn/H8d+fY3gT+3/vqDwOdzzh3w+0qp3wN+APjXp3l+ZbTIRPQRv4xoBbH3qLyLUWuxvSiT0815CQxkpIBrxM94mFq21srXa2vPYYcA0oF0dPcwWGcenWrOSbqSBt2l0AYJRFqPheVqJkHlWuOgEz/b2iY0rAIxiMf0UY0mW1uR9hgmkDXUs5uXpbh2iA/W9ZS1BEihUHgwuBdqDh8D/o/1108gwWLgmfVjp4p2UO00+ItLtFMoZ9h+ZEYioozcDbvaoStN6iS9ZCf22MyB4nDA7WiO/6hb2Y0G247il35cqFNI+JUfbUMHw5/RSvQmh95eyU5TKQkeqZFja6dPR69oA0oPKabxBqBQeNA5s+CglPo14PUnPPWpnPMX1q/5FBCAf/wajv8J4BMATz311C29N1uDsYrpw1NSymilyAZWl1bENuC2xEvAaYeaKPETWHsfhz5IeidDbKMY8az78GMfj+Xeb2Tsc5TQHw56Dd1NsY+YqZEOIgXNbnPqef3XusDZ+noJkNOohdzL5PUU/fC5h/Tig/yZC4UzCw455z/5Ss8rpT4K/KfAn8iHt9jPAm848rIn14+ddPzPAJ8BePe7331r+g4+o5QmrFpCH1FGU+9U6EoT27UntFHEGMFIeqinRxklNp5BTpdjpm97qSU0IsanKz1qLN2UH8ERzSRbW6otsSrFglGGerum3qlf8RB3ElMZqu0jHtnOHCuo3yuM0uansID7lT8WEFNM+IW/p34vhcJpc7e6ld4P/HfAH885L4889SvA55RSfxd4HHgr8Funfv6UafdW1Fs1ukrodceR6jKqWlt/rjuUcszMX5ozuzADxPSnmlYoLdPKKsnrqm2Z6B3SKzebqnGNkwIvkobSVqMbSfW4LUe9fe8tQHdravpmSDHRz6XhALUOZrPqtlJnRxsBUkiknEgpUaWqtPIWHlju1l/4LwA18MX1H+1v5px/Muf875RS/wT490i66adyzq/eonOLaAXOadK6qLy6tMDNxHfBVAaMpITqrZrUp7EWYCeWHA5nEIa7yWjFk4H65oPCgJs4ULI7CT6I9tI65ZO9nOtWF+KcsgSaDUx7+IU/nCLP8nP12t+W5IjSYuAUVmH8nSur8FueanLn1GoLhTvJ3epWessrPPczwM+c5fnNTkO109DuySS0ndZkwE4ctjGELqLRY7G43qnpV72I7XWBjMh4Dx1LOmj8wqOdHhVbbwXXOIw1J0pjxO7mg0POGb/0o2GQMuKPvSnTzjmd7HsR+3hbNqa2sWNH2dHHBuHFTQzChQefezM3cMa4xtHMHH3rSV0idh7bOPpFR+wt9W5NJqPRKCMSGsknjDG4xkktQkPqkrSG+kTf96Agz/I4V3BLRd9TyE6Efu0rsSZHKaQ2u81mOKjd4CPeru+FrazsPNaxW1tpZyYz6lcVCg8aG/mvOh205EsLmnMT8WrYqkg+wHDnmda56q1q/VwaVVVtbam3axHTm7j1EJ14OKz2VviVJ6wC7X57U0NrA9roE+/wb6bjafxc/fU7jxzzTYn1PQgopU6UFzHN7e+cTGVGFd2jwWAjgm5hI9nInUNa9mhn8Hst/UGPMjIX4HYbXG1Q1XpIrg1klal2qjFNMxj45HgoipfJpJgOh+XWHUhhFW4ppVNtVZIW8kEK0hN3Uyml6OPoRZ1zvm7B2qSiaTWtpKusS2PX2FEBxNeKrURv66gbna3txqTsCpvHRgYH1TgSmdxFqomln3t8SESfMI9t47Skk7QW85voRXZ7kMXQ25rmXEN/0I8DakNKI3QSEJRdt8LeynWdMEH9avSrXsT5gBgioQ/U08MOp2vvdO8UsY/4lciGG2twU3fHhsdc7aTd4RRRWlHv1IQ+kGM+bF4oFB5QNjI42HNTcmXJV1bopgYbUUZhp9Ly2O634gBn8yhnoRC5bF3JtHQ1rVAoKVBHkcnWVhO7KEXktSbTaZNiEtOftZz4Ube00ZI0Z5EMd+ZVVVzPghRFJnyY34g+Eg8ik93Jfb2LUVrd3OzKmugjoQ0SINfzIPfz5y9sFhsZHFJMuHNT7F5LTgnrNKa2ZBR+1RPadbdS8qJx1EeabVnoXe1kZ6CkE8hNHO1ee0yCW6FEIvsU7TVTSPSLw/59W1sR0btm/M9Y8ZG4m/MRoT9BNjytd2D1ZvyTSzEd89UIMZBCot65eR2rQuFuspkF6S6Qnr+CRZGBuOxRZGIX6A+82If2gbAKkgrJh/4FKadjeebo4yixHTpZAGxjb3vw6lrGwS6QekYbxkBxLcMAX+HuMbQTH+Wo4VOhcK+zkcFBrXpil8gO6t2Grcd3cDs1OSe0VlgnhcasMm7qqGaVtLY6TTNrxpROv+xZXF6s22EjOUgaqm8PZbZPgxTSsULoQA551F8aeK2zFqfJ0UnxEc1mFW9vIOiSb/REoXCPsRl7/GvJQOXousDBt18mtoHp49tU2zXBy3Tx4PusrMI5x2R3Ig5uUydWnPsdoQv0+/3opBZ9HFVU7cSe3s5hKHhfu64M6SWrZRhP6dNTWb0NtNHUW/WxgvRR2fBNwFTm2MwJIAHSblCALNzXbGRwiBNL7wP9pQXVrKbtI/6gx2w3THdqzFRaSJVVsnOYVJjK4GopKPpWFr3B55ksefZ6qxZJ50q/YodQ7CMpyaJ50u4ixURow2jtObS0hu74YjMUmwePiXuJaz20Nw3jxCHPr/xop1rNihZT4f5hI4MDTpOnE4hzcgg0F6boaYUCTC31Ajd1zM7Pji1wg6nNkOJRWmYRYhtHZzSU9NqftOjnnOnnhzIMHo+bumOqpjln+oN+PEeMMsNQ79RgZNBtGMbbqDTNfYhrnBTgi9d24T5kI4ODNRZqjWkcykd0U9EuexqrySFAqlFZOo5MZWTBXvSHpjxpLQettbioofCdCPO5ibthMTp04TovBL/yx/wQoo/X1xfWdpzVpILJGfxACmeGUmojjJAKDx4bGRySD0x8JGpF10fysqWyhnrisErL/EIbqLZkGC104dBWEwkKoQ+oSqGQwbXpI9PrRNhyziKnsc49D7YV6uhqca0+z43qlaWOWSgU7iD3VqL6TrHsyVf3mezP2XGZrZ2aygIXD8ZFOKeMMrKIH9UsGuSwbWVHuW1gNAA6yqCzRGaUjz6plfGYe5wzJ95plhRSoVC4k2zkziEuOvxvfof2a9+FkDFveQT9vY9i3vY4GEkDVNvVqNWvjCKtEr71IpcB6Fof6yCKfSSG41PA1xaQjTPjxPXo/tZcI+SmxeO5X66lObR4PrxacMhZuqSUViW/XSgUbpvXFByUUj+ac/7iaV/MnSJfWhC+ewllLdlk8sEKHRJqatHTCjMxIrTWx1GNc3lpeWx3oLJIZxwTxnuVKWClFXYmBe+hxfOkjh5TGSbVZOyIerXW1NAF+mUvRXGFFLlvQeahUCgUruW1ppV+8VSv4g4Tlz3m8fOY7RqlQT+yQ//SHN0FtNGiuBoi3UEnUhCsF9yZE12lbWltPSlFdGiHzYlBYuhgqabVq7Z6avPqMwspiqzG4DVAvsYNrVAoFF4DN9w5KKV+5UZPARfO5nLuDPbRbXIXyNMK+9gufcyY2uBjhtYDiuyyKKu2ET2TRfpoauckaWzU8drA0KI6pJdsbY+1rZ4G0ccTi9XRxzumglooFB48Ximt9EPAfwXMr3lcAT9wZld0B2je+nqmP/y9rP7v/0BKCW0N1dNPkDLkLkBTyePo0YvZ1OZYx5JSiuZ8I3WItC5QT911RjDVtLot/+JX44Y7ixIXCoXCbfBKweE3gWXO+f+59gml1O+e3SWdPdpo3Du+h6wV9IlOK9Lck1MLO9PxNcCorKqdHsXuhgG00Ispj9YarfVNSyOkkMhJPAFut3g8zEgcnY1Qa32o4VxDauyWrUsLhcLG8krB4fcBf9ITOef3ns3l3BlyTMRnLhG/exV6j5lUUDnYqlGVxjRi1mMri2tEAsEv/ZhaSikRlyJtoUSjm+gjvvNjhxNI6mlI+wz1haPDdChxf7sZt7cboZSY0PhW6gzaaGwjOkaxj8d8FYbZjds5X6FQ2AxeaZX4XeDTSqnHgH8C/FLO+d/emcs6W9L+CrXoycuOdHkBRmMfP4dptplemOFet31oqJMzvr0mRiaxAB2G5Aayz+QmE3rRRQrLMHYbKa3QlT6WmiJLsBj8IV4rSqsTU1eDrs/R84VlKMGhUCi8KjfMMeSc/37O+QeBPw5cAv6hUuo/KKX+plLqe+/YFZ4BcdETFx0hQZg2pFlNWraYLqCRGsOY7lkPsB1FKUVW11eBlVZ0Bx1+7un3evqDXhZoZHiun/fXX0zizDqLTpL5HqQ/CoVC4ZV41QR0zvk7Oeefyzn/UeAjwI8Bv3PmV3aGpJjxS0+OkXRliX9uj7DwqJ2KhCisjlIXWl1fS1Bc77SmEWE8n8Y0E8h09fD16Dd9zbHOamjtpBpIkYwuFAo3w6vmF5RSFvjTwIeBPwH8OvC3zvSqzprGomoHPmFmFdgG/cgWISryfo+tetBQb9Wj9HI+yOOduKmMDLIFWfgHldQx/ZRlqnqYps5J/u8mbpT4HrC1PTO5bTd1pIM0nh8tjxUKhcKr8UpzDj+K7BQ+APwW8HngEznnxe2eVCn1PwEfREa3XgI+mnN+Tkni/e+vz7lcP/6V2z3ftehJhX5oin3iPOHSHHwkKU3SGjPcxKe1YqqTbqDmXCOBQKmx4yfndcAIyIyDNQTWnUGNpQ8ynKaNSG24mbS6hi7I406/oqfyKNzXSVeUbewtTT5re3jdwG3XNgqFwubwSjuHnwY+B/y3Oecrp3zeT+ec/wcApdRfBf5H4CeRHcpb1//9MeB/W///VLHTCvvELvG5Pcy5CWpSkXen5KtLeN32+LoU0rFhN20PJ5Z96/ELf+y1dmqxtZjyaK2pt2uUkV3F0TbSm5178Msjiq7k8Xy3EiCUUqUAXSgUbpkbrho55x85q5PmnPePfDvjMNHyQeCzWRL+v6mUOqeUeizn/Pxpnl9bTbM7JT6+S9xfovqE0pl0TXpn6DTKSWoUoQ+jNeexrqM1oQ1Mz08xtZE5htvwkc45XyfcN5yj6CYVCoWz5q7dUiqlfgb4b4A94D9ZP/wE8N0jL3tm/dh1wUEp9QngEwBPPfXUa7kAVMzgpR3JAn3OcCTtMuTn/dIfLtRZFujYx+uVUtch7lTktYt/Q6FQuIuc2bisUurXlFJfP+G/DwLknD+Vc34D8I+BT97q8XPOn8k5vzvn/O5HHnnklt6bYqKbd4SDFfhIDhnfReqZw01FXK/ZbcTUpwv4/po5h3yybMVppm+UPjkd9Eo1ikKhUDgtzmylyTn/yZt86T8G/hnwN4FngTccee7J9WOnSugDpIy5sEVe9qTWgzXo8zNcZTETN3o9DxahburQ+jCW6lpjrR2lKYwzp94JNBzvaDrLNiU4FAqFs+eurDRKqbfmnL+5/vaDwH9Yf/0rwCeVUp9HCtF7p11vGMiVJVhLbkBvNSitUZVDb9fHvJ6VUmij8SsvuOFg9gAAEmZJREFU3tDrQQXXiGeCjfZQJ+mUO4GUFgtSl13pMioUCneUu3Ub+rNKqe9DWlm/g3QqgewgPgD8HtLK+hNncXJtRMZCbdXk/UQMCWYG9/g2elLh97tjr7eNPaw5GPGADl3Arzw5iPsaGqrZ2egWlcBQKBTuNHclOOScP3SDxzPwU2d+/phxMwcaogKdFXpmSdaKWF6M9MtepqMrg9YaN3FMzk3olyKcl3OmP+jJKeO2HMYa+nmP3tVnNtR2P1FsSwuF+5uNTGCPswsJqrqSqWUP3UFHjpnk0//f3v3HyHbWdRx/f8+v2b23tLeFiqWttJASrUGh3FSMiGJMgf7hhYQ/SggQNKkaQEg0BuQf/iFRIpJUDaYoEQhaETQ0/kAo1l8xUCrpL8DCBar0WuHyw3J7d3fm/Pj6x/PM7OyeWbq7nd0zO+fz2mz2zDOzs88zZ/Z85zzneZ5vWO+ogXpYh3zSx4oQOIabS2GMZx43o5DyE49JdnoeHOpRzfD8cJK2NBuE1KgicnT08iiWFVlY2iIuelcNw1pKaZoyOj8iSROK4wXpShpmSGfh9mQZim2mU4Pu1AVUj2pG50eM1kczF8RbFu6+GRhgMvR31pwNEVlcvTxzSLKE7FjG2nfWwoXnJBzUR+ujyRyFJN3sHkrTdPJ7llqYOR2/ICyDAXGRvhlzHEZrI6r1zYNjtVGxcuHKUp5hNGWzGRi2lw/a5SKymHoZHCAGiEEWkvBYuMiMM3NG8zgbnJmRr+asfXuNaqMiyUIAGS+Pka1mrf51b3yyBMZEExPvLGFXy47XF5YvDoostd4GByMMEx1+b8hobUSapKQrKUmRkFgyybGQrWRbJp41ZUgTmg0yHMcIZwvbE/+MeeMzZzsva9dSkrXzbY+vO4jI0dHb/1hLjGoU+sKzIgsXlC1kcxs8eTA5oG//JFyVm2cB3oQscX7OJyOgtncrWRqGuW7valnmXM7F8YIqrWjKZrLw4DJ2oYkss14GB2+c0dqIpmpo6jAyqckbVi5aoa5qquHOi9slaTLJplaeLydnAHVV05xrWD2xuiWgmBnF8SJkgZtae2mZF88bd7+x2nVNZJHVo5qmCXnP57IemcxVL4NDWD4jzHKeHpJaroX8DaPzI5pRQ3FB0TpzyFdyhtWQuqwpN0rqYU26kjI8NyQ/llON2oElKzLSEyl1FfJB6B9B+my8JM1012O2ouHOi6aX5/rjYJDmKUmeTLqYGm/IV3MMmxz8t0uLNORpiPMkstWMLMsmyYGmh7VOGy+kp8AgfVeXdWvJ+2qjmiSlksXQyzOHNE8nQ0uLY0XoGrKQFnSctY2GHd+saZ6SDsJCe+NUoFVZ4aOQkMdLn2R9E5GtxoM9WuV1ow9Pe+RNyEt/EEvs9DY45MdzyvUSa4ykSMjJwwXmqgxdTZWHNZVWqpkjbdI0jFCqhhXlRolh5Mc2zzr8nLNyYqWD1okstp0GY+jD1O41dRO6v6dGVe42w+Ru9TI4QLh2kA3Ciqp1XbO2sUa5VlKuha6k4ni43jB6bDQzo1s6SDevLzjUVodF92IAb+qGupyREGiPmqqhXC+pq5o0S8lX86Ue6STLL81T0iKdrHwMYaizzhp2b3Ru60oL1XqFpUY+mN9Al94GBwinYpaGAJAXOV7FtZMsXCMYH4SrUUWRbY3KaZ6ycuEK1bCiaZqwQN+cP/l442yc25gMg61HNXVVs3rRqhazkyPLLMwxqssar0M63bRQYNit8SjLVvlwvqsQ9Do4NHVDtV6x8ehGyO6W2WYXkhPOAnbI+gbh9LjICtIiZbhtme8kf+LD86qyai9F0TBzRJTIUWI2O9Oh7IIxOTa1yueot3tnvOR2U4dx1tVGFZaYtriMRpFMAsTjfapJ85TiSQXleknTNOR5Pp+scMojLT1WjSqqjYqmbsiKLIwk1Bnz5Exr+4ivea9C0Nvg0JSbp2bpIKWuayjBCiPLsrAia5aSrexudq83Yb6EuYVt90nWuP1K85TSyq1BwlDfrCy9uqwZnRtNbo+DxMqFGuQB4ZpomZTUo5okSchWsrl3zfU2ODhbl9keL8mdFRmDCwd7GhpWjSrK85tzIsajlQYX7e15tkvShMEFA0ZrIamQJUZxrDjwUR3uIaeF46RZqk9rcuimL1aPNeV8BnksA7NwLODYwf2N3gaHNEu3rHlkhIxl+bG952tuRjMuDsVlOZ7oGzktUlaL1c0ERQesqZutIyGSMP9D/5ByqNSl2rnejom0xLZOeksI6T7neBCc58H8sPJIV+vV1pEQDZPhvSKHZbxM/jRLTcO4D1Fvzxwgjrc+kT7hWYbpIG1lOkvy9tyIo2B61dmxpm4m3VoihyErMvy4U66Ha25plu7rrF72r9fBYeyJHvTGo5W2j6w4itI0pW629feOh86JHKLxRFX8if+Pyt4pOMxJVmRLMW47W82oq3pLn29xrNAnNumEmemDSUeO/tFM5irNU1YuWpl0k6V5qovRIj2k4CAtSZrMfREvETlajt4V0wW3Uz4HEZGjpNPgYGa/bmZuZk+Jt83MbjGz02Z2n5ld12X99qIclqx9d431764zPDecuTCWiMhR0VlwMLMrgRuA/54qfilwTfy+GXjPQdfDG6fcKBmdH4X0oftQj2rKx8owoc7D7emp/yIiR02XZw7vBn6TrXMhTwEf8ODTwAkzu+ygKtDUDRuPblCeL6k2KkbnRozW9n5QnznVP+ZzEBE5ijoJDmZ2Cjjj7vduu+ty4OtTtx+OZbOe42Yzu9vM7j579uy+6lENq0k+6UnZRqUuIRHpvQMbrWRmdwA/OOOutwG/RehS2jd3vxW4FeDkyZP7ugo8zv+8tTCW72H05k4zpDUEVESOqgMLDu7+87PKzezZwNXAvXFi1RXA58zseuAMcOXUw6+IZQciyZN2l5DtnON2J5MZ0ushK1yWZ/PJ5yAi0pFDn+fg7vcDPzC+bWYPASfd/VtmdjvwBjO7DfgJ4FF3f+Sg6pINMpqyYfjYkKZssNRYPbG/FJzLMkNaRAQWbxLc3wE3AqeBNeB1B/nHzMIy3UmekKThuxpWZIPsSC6aJyIyL50HB3e/amrbgdcf2t9unGpYkViyeY3Bw0Xp4gLNEBaR/ur1x2NvfGZSke0jmERE+qbXwcFSm3l9Icl7/bKIiPQ8OJiF7qOpVyEdpGQrnfe2iYh0qvdHwTRPWT2xSlM1mCkNoYgIKDgA4QxCE9ZERDbpY7KIiLQoOIiISIuCg4iItCg4iIhIi4KDiIi0KDiIiEiLgoOIiLQoOIiISIuCwxRvXIvuiYigGdJACAqjtdEkK1yapxTHi30l/RERWQY6cwDK9ZJ6WIflux3qUU25VnZdLRGRzig4ANWo2lWZiEhfKDgQFt5rFx5+PUREFoWCA8zM35Cv5h3URERkMeiCNJCv5JgZ9bDGcdJBSj5QcBCR/lJwiLJBRjbQyyEiAupWEhGRGRQcRESkRcFBRERaOgkOZvZ2MztjZvfE7xun7nurmZ02swfN7MVd1E9EpO+6vAL7bnf/3ekCM7sWuAn4UeBpwB1m9ix3r7uooIhIXy1at9Ip4DZ3H7r714DTwPUd10lEpHe6DA5vMLP7zOx9ZnZxLLsc+PrUYx6OZS1mdrOZ3W1md589e/ag6yoi0isHFhzM7A4ze2DG9yngPcAzgecAjwDv2uvzu/ut7n7S3U9eeumlc669iEi/Hdg1B3f/+d08zszeC/xNvHkGuHLq7itimYiIHKKuRitdNnXz5cADcft24CYzG5jZ1cA1wF2HXT8Rkb7rarTSO83sOYQMCg8Bvwzg7p83sw8DXwAq4PUaqSQicvg6CQ7u/urvc987gHccYnVERGSbRRvKKiIiC0DBQUREWhQcRESkRcFBRERaFBxERKRFwUFERFoUHEREpEXBQUREWhQcRESkRcFBRERaFBxERKRFwUFERFoUHPbJG8fdu66GiMiB6GrJ7iOrLmvKtZKmbsAgP5aTD/KuqyUiMlcKDnvg7gwfG0IzLoDysZIkSUjztNO6iYjMk7qV9qApm83AMKUeKR+RiCwXBYe9sD2Wi4gcUQoOe5BkCUm27SUzyAr1zonIclFw2AMzo7igIBtkkECapwwuGLQDhojIEaePvHuUpAnFBQUFRddVERE5MPrIKyIiLQoOIiLSouAgIiItCg4iItKi4CAiIi0KDiIi0mLLsLKomZ0F/mufv/4U4FtzrM6i61N71dbl1Ke2wsG29+nufumsO5YiODwRZna3u5/suh6HpU/tVVuXU5/aCt21V91KIiLSouAgIiItCg5wa9cVOGR9aq/aupz61FboqL29v+YgIiJtOnMQEZEWBQcREWnpdXAws5eY2YNmdtrM3tJ1ffbLzB4ys/vN7B4zuzuWXWJmnzSzL8efF8dyM7NbYpvvM7Prpp7ntfHxXzaz13bVnmlm9j4z+6aZPTBVNre2mdnz4mt3Ov5uZ3n9dmjr283sTNy395jZjVP3vTXW+0Eze/FU+cz3tZldbWafieV/YWadrTtvZlea2Z1m9gUz+7yZvSmWL+u+3am9i7t/3b2X30AKfAV4BlAA9wLXdl2vfbblIeAp28reCbwlbr8F+J24fSPw94Tkps8HPhPLLwG+Gn9eHLcvXoC2vRC4DnjgINoG3BUfa/F3X7pgbX078BszHnttfM8OgKvjezn9fu9r4MPATXH7j4Bf7bCtlwHXxe0nAV+KbVrWfbtTexd2//b5zOF64LS7f9XdR8BtwKmO6zRPp4D3x+33Ay+bKv+AB58GTpjZZcCLgU+6+3fc/bvAJ4GXHHalt3P3fwG+s614Lm2L913o7p/28B/1gannOnQ7tHUnp4Db3H3o7l8DThPe0zPf1/FT888BH4m/P/26HTp3f8TdPxe3zwFfBC5nefftTu3dSef7t8/B4XLg61O3H+b776xF5sAnzOw/zOzmWPZUd38kbv8v8NS4vVO7j9LrMa+2XR63t5cvmjfErpT3jbtZ2Htbnwz8n7tX28o7Z2ZXAc8FPkMP9u229sKC7t8+B4dl8gJ3vw54KfB6M3vh9J3xk9NSjlle5rZF7wGeCTwHeAR4V7fVmS8zuwD4KPBmd//e9H3LuG9ntHdh92+fg8MZ4Mqp21fEsiPH3c/En98E/ppw6vmNeGpN/PnN+PCd2n2UXo95te1M3N5evjDc/RvuXrt7A7yXsG9h7239NqErJttW3hkzywkHyg+5+1/F4qXdt7Pau8j7t8/B4bPANfEKfwHcBNzecZ32zMyOm9mTxtvADcADhLaMR268FvhY3L4deE0c/fF84NF4Gv8PwA1mdnE8tb0hli2iubQt3vc9M3t+7LN9zdRzLYTxgTJ6OWHfQmjrTWY2MLOrgWsIF2Bnvq/jp/A7gVfE359+3Q5dfL3/BPiiu//e1F1LuW93au9C79/Dulq/iN+EERBfIlz9f1vX9dlnG55BGLFwL/D5cTsIfZCfAr4M3AFcEssN+MPY5vuBk1PP9YuEC1+ngdd13bZYpz8nnG6XhH7UX5pn24CThH/IrwB/QFw1YIHa+sHYlvsIB4zLph7/tljvB5kaibPT+zq+V+6Kr8FfAoMO2/oCQpfRfcA98fvGJd63O7V3Yfevls8QEZGWPncriYjIDhQcRESkRcFBRERaFBxERKRFwUFERFoUHER2YGYvMzM3sx/usA5vNrNjXf196S8FB5GdvRL4t/izK28GFBzk0Ck4iMwQ18B5AWEi2k2x7GfN7J/N7GNm9lUz+20ze5WZ3RXzBjwzPu4qM/vHuJjap8zsh2L5n5rZK6b+xmNTz/tPZvYRM/tPM/tQnAn8a8DTgDvN7M5Dfgmk5xQcRGY7BXzc3b8EfNvMnhfLfxz4FeBHgFcDz3L364E/Bt4YH/P7wPvd/ceADwG37OLvPZdwlnAtYabrT7n7LcD/AC9y9xfNp1kiu6PgIDLbKwlr5RN/jruWPuthbf4hYfmCT8Ty+4Gr4vZPAn8Wtz9IOAN5PHe5+8MeFmC7Z+q5RDqRPf5DRPrFzC4hJE55tpk5IfuWA38LDKce2kzdbnj8/6eK+IHMzBJCJq+x6eetd/FcIgdKZw4iba8APujuT3f3q9z9SuBrwE/v8vf/nXidAngV8K9x+yFg3D31C0C+i+c6R0grKXKoFBxE2l5JyIsx7aPsftTSG4HXmdl9hOsSb4rl7wV+xszuJXQ9nd/Fc90KfFwXpOWwaVVWERFp0ZmDiIi0KDiIiEiLgoOIiLQoOIiISIuCg4iItCg4iIhIi4KDiIi0/D+iYqVeMiSM1AAAAABJRU5ErkJggg==\n",
1176 |             "text/plain": [
1177 |               "<Figure size 432x288 with 1 Axes>"
1178 |             ]
1179 |           },
1180 |           "metadata": {
1181 |             "tags": [],
1182 |             "needs_background": "light"
1183 |           }
1184 |         }
1185 |       ]
1186 |     },
1187 |     {
1188 |       "cell_type": "code",
1189 |       "metadata": {
1190 |         "id": "2kar1en3J19A",
1191 |         "colab_type": "code",
1192 |         "outputId": "3a8f7ac5-690e-4228-ae27-6afcdd9a6054",
1193 |         "colab": {
1194 |           "base_uri": "https://localhost:8080/",
1195 |           "height": 296
1196 |         }
1197 |       },
1198 |       "source": [
1199 |         "sns.barplot(x='Class',y='Amount',data=df,palette='PuRd')"
1200 |       ],
1201 |       "execution_count": 17,
1202 |       "outputs": [
1203 |         {
1204 |           "output_type": "execute_result",
1205 |           "data": {
1206 |             "text/plain": [
1207 |               "<matplotlib.axes._subplots.AxesSubplot at 0x7f7765799860>"
1208 |             ]
1209 |           },
1210 |           "metadata": {
1211 |             "tags": []
1212 |           },
1213 |           "execution_count": 17
1214 |         },
1215 |         {
1216 |           "output_type": "display_data",
1217 |           "data": {
1218 |             "image/png": "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\n",
1219 |             "text/plain": [
1220 |               "<Figure size 432x288 with 1 Axes>"
1221 |             ]
1222 |           },
1223 |           "metadata": {
1224 |             "tags": [],
1225 |             "needs_background": "light"
1226 |           }
1227 |         }
1228 |       ]
1229 |     },
1230 |     {
1231 |       "cell_type": "code",
1232 |       "metadata": {
1233 |         "id": "aKOIm0hkKWcT",
1234 |         "colab_type": "code",
1235 |         "outputId": "284f1397-b96e-4eb0-ee3b-3ccc1dc4b692",
1236 |         "colab": {
1237 |           "base_uri": "https://localhost:8080/",
1238 |           "height": 296
1239 |         }
1240 |       },
1241 |       "source": [
1242 |         "sns.barplot(x='Class',y='Time',data=df,palette='Purples')"
1243 |       ],
1244 |       "execution_count": 18,
1245 |       "outputs": [
1246 |         {
1247 |           "output_type": "execute_result",
1248 |           "data": {
1249 |             "text/plain": [
1250 |               "<matplotlib.axes._subplots.AxesSubplot at 0x7f77657013c8>"
1251 |             ]
1252 |           },
1253 |           "metadata": {
1254 |             "tags": []
1255 |           },
1256 |           "execution_count": 18
1257 |         },
1258 |         {
1259 |           "output_type": "display_data",
1260 |           "data": {
1261 |             "image/png": "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\n",
1262 |             "text/plain": [
1263 |               "<Figure size 432x288 with 1 Axes>"
1264 |             ]
1265 |           },
1266 |           "metadata": {
1267 |             "tags": [],
1268 |             "needs_background": "light"
1269 |           }
1270 |         }
1271 |       ]
1272 |     },
1273 |     {
1274 |       "cell_type": "code",
1275 |       "metadata": {
1276 |         "id": "aO4f0k-Ovddd",
1277 |         "colab_type": "code",
1278 |         "outputId": "84713212-f5e0-4aad-fd57-16bb76facefc",
1279 |         "colab": {
1280 |           "base_uri": "https://localhost:8080/",
1281 |           "height": 296
1282 |         }
1283 |       },
1284 |       "source": [
1285 |         "sns.stripplot(x=\"Class\", y=\"Time\", data=df , palette= \"Purples\")"
1286 |       ],
1287 |       "execution_count": 19,
1288 |       "outputs": [
1289 |         {
1290 |           "output_type": "execute_result",
1291 |           "data": {
1292 |             "text/plain": [
1293 |               "<matplotlib.axes._subplots.AxesSubplot at 0x7f77656dc208>"
1294 |             ]
1295 |           },
1296 |           "metadata": {
1297 |             "tags": []
1298 |           },
1299 |           "execution_count": 19
1300 |         },
1301 |         {
1302 |           "output_type": "display_data",
1303 |           "data": {
1304 |             "image/png": "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\n",
1305 |             "text/plain": [
1306 |               "<Figure size 432x288 with 1 Axes>"
1307 |             ]
1308 |           },
1309 |           "metadata": {
1310 |             "tags": [],
1311 |             "needs_background": "light"
1312 |           }
1313 |         }
1314 |       ]
1315 |     },
1316 |     {
1317 |       "cell_type": "code",
1318 |       "metadata": {
1319 |         "id": "_Z2t79B7WdYt",
1320 |         "colab_type": "code",
1321 |         "colab": {}
1322 |       },
1323 |       "source": [
1324 |         "import numpy as np\n",
1325 |         "from sklearn.decomposition import PCA\n",
1326 |         "X = df.drop(columns='Class')\n",
1327 |         "y = df['Class']"
1328 |       ],
1329 |       "execution_count": 0,
1330 |       "outputs": []
1331 |     },
1332 |     {
1333 |       "cell_type": "markdown",
1334 |       "metadata": {
1335 |         "id": "p8DrMhXDYCjS",
1336 |         "colab_type": "text"
1337 |       },
1338 |       "source": [
1339 |         "'''pca = PCA(n_components=2)\n",
1340 |         "pca.fit_transform(X)'''"
1341 |       ]
1342 |     },
1343 |     {
1344 |       "cell_type": "code",
1345 |       "metadata": {
1346 |         "id": "LzthJAEabIYX",
1347 |         "colab_type": "code",
1348 |         "outputId": "487e0da9-32f6-41d8-d67d-6d4f25dd5f91",
1349 |         "colab": {
1350 |           "base_uri": "https://localhost:8080/",
1351 |           "height": 224
1352 |         }
1353 |       },
1354 |       "source": [
1355 |         "X.head()"
1356 |       ],
1357 |       "execution_count": 21,
1358 |       "outputs": [
1359 |         {
1360 |           "output_type": "execute_result",
1361 |           "data": {
1362 |             "text/html": [
1363 |               "<div>\n",
1364 |               "<style scoped>\n",
1365 |               "    .dataframe tbody tr th:only-of-type {\n",
1366 |               "        vertical-align: middle;\n",
1367 |               "    }\n",
1368 |               "\n",
1369 |               "    .dataframe tbody tr th {\n",
1370 |               "        vertical-align: top;\n",
1371 |               "    }\n",
1372 |               "\n",
1373 |               "    .dataframe thead th {\n",
1374 |               "        text-align: right;\n",
1375 |               "    }\n",
1376 |               "</style>\n",
1377 |               "<table border=\"1\" class=\"dataframe\">\n",
1378 |               "  <thead>\n",
1379 |               "    <tr style=\"text-align: right;\">\n",
1380 |               "      <th></th>\n",
1381 |               "      <th>Time</th>\n",
1382 |               "      <th>V1</th>\n",
1383 |               "      <th>V2</th>\n",
1384 |               "      <th>V3</th>\n",
1385 |               "      <th>V4</th>\n",
1386 |               "      <th>V5</th>\n",
1387 |               "      <th>V6</th>\n",
1388 |               "      <th>V7</th>\n",
1389 |               "      <th>V8</th>\n",
1390 |               "      <th>V9</th>\n",
1391 |               "      <th>V10</th>\n",
1392 |               "      <th>V11</th>\n",
1393 |               "      <th>V12</th>\n",
1394 |               "      <th>V13</th>\n",
1395 |               "      <th>V14</th>\n",
1396 |               "      <th>V15</th>\n",
1397 |               "      <th>V16</th>\n",
1398 |               "      <th>V17</th>\n",
1399 |               "      <th>V18</th>\n",
1400 |               "      <th>V19</th>\n",
1401 |               "      <th>V20</th>\n",
1402 |               "      <th>V21</th>\n",
1403 |               "      <th>V22</th>\n",
1404 |               "      <th>V23</th>\n",
1405 |               "      <th>V24</th>\n",
1406 |               "      <th>V25</th>\n",
1407 |               "      <th>V26</th>\n",
1408 |               "      <th>V27</th>\n",
1409 |               "      <th>V28</th>\n",
1410 |               "      <th>Amount</th>\n",
1411 |               "    </tr>\n",
1412 |               "  </thead>\n",
1413 |               "  <tbody>\n",
1414 |               "    <tr>\n",
1415 |               "      <th>0</th>\n",
1416 |               "      <td>0.0</td>\n",
1417 |               "      <td>-1.359807</td>\n",
1418 |               "      <td>-0.072781</td>\n",
1419 |               "      <td>2.536347</td>\n",
1420 |               "      <td>1.378155</td>\n",
1421 |               "      <td>-0.338321</td>\n",
1422 |               "      <td>0.462388</td>\n",
1423 |               "      <td>0.239599</td>\n",
1424 |               "      <td>0.098698</td>\n",
1425 |               "      <td>0.363787</td>\n",
1426 |               "      <td>0.090794</td>\n",
1427 |               "      <td>-0.551600</td>\n",
1428 |               "      <td>-0.617801</td>\n",
1429 |               "      <td>-0.991390</td>\n",
1430 |               "      <td>-0.311169</td>\n",
1431 |               "      <td>1.468177</td>\n",
1432 |               "      <td>-0.470401</td>\n",
1433 |               "      <td>0.207971</td>\n",
1434 |               "      <td>0.025791</td>\n",
1435 |               "      <td>0.403993</td>\n",
1436 |               "      <td>0.251412</td>\n",
1437 |               "      <td>-0.018307</td>\n",
1438 |               "      <td>0.277838</td>\n",
1439 |               "      <td>-0.110474</td>\n",
1440 |               "      <td>0.066928</td>\n",
1441 |               "      <td>0.128539</td>\n",
1442 |               "      <td>-0.189115</td>\n",
1443 |               "      <td>0.133558</td>\n",
1444 |               "      <td>-0.021053</td>\n",
1445 |               "      <td>149.62</td>\n",
1446 |               "    </tr>\n",
1447 |               "    <tr>\n",
1448 |               "      <th>1</th>\n",
1449 |               "      <td>0.0</td>\n",
1450 |               "      <td>1.191857</td>\n",
1451 |               "      <td>0.266151</td>\n",
1452 |               "      <td>0.166480</td>\n",
1453 |               "      <td>0.448154</td>\n",
1454 |               "      <td>0.060018</td>\n",
1455 |               "      <td>-0.082361</td>\n",
1456 |               "      <td>-0.078803</td>\n",
1457 |               "      <td>0.085102</td>\n",
1458 |               "      <td>-0.255425</td>\n",
1459 |               "      <td>-0.166974</td>\n",
1460 |               "      <td>1.612727</td>\n",
1461 |               "      <td>1.065235</td>\n",
1462 |               "      <td>0.489095</td>\n",
1463 |               "      <td>-0.143772</td>\n",
1464 |               "      <td>0.635558</td>\n",
1465 |               "      <td>0.463917</td>\n",
1466 |               "      <td>-0.114805</td>\n",
1467 |               "      <td>-0.183361</td>\n",
1468 |               "      <td>-0.145783</td>\n",
1469 |               "      <td>-0.069083</td>\n",
1470 |               "      <td>-0.225775</td>\n",
1471 |               "      <td>-0.638672</td>\n",
1472 |               "      <td>0.101288</td>\n",
1473 |               "      <td>-0.339846</td>\n",
1474 |               "      <td>0.167170</td>\n",
1475 |               "      <td>0.125895</td>\n",
1476 |               "      <td>-0.008983</td>\n",
1477 |               "      <td>0.014724</td>\n",
1478 |               "      <td>2.69</td>\n",
1479 |               "    </tr>\n",
1480 |               "    <tr>\n",
1481 |               "      <th>2</th>\n",
1482 |               "      <td>1.0</td>\n",
1483 |               "      <td>-1.358354</td>\n",
1484 |               "      <td>-1.340163</td>\n",
1485 |               "      <td>1.773209</td>\n",
1486 |               "      <td>0.379780</td>\n",
1487 |               "      <td>-0.503198</td>\n",
1488 |               "      <td>1.800499</td>\n",
1489 |               "      <td>0.791461</td>\n",
1490 |               "      <td>0.247676</td>\n",
1491 |               "      <td>-1.514654</td>\n",
1492 |               "      <td>0.207643</td>\n",
1493 |               "      <td>0.624501</td>\n",
1494 |               "      <td>0.066084</td>\n",
1495 |               "      <td>0.717293</td>\n",
1496 |               "      <td>-0.165946</td>\n",
1497 |               "      <td>2.345865</td>\n",
1498 |               "      <td>-2.890083</td>\n",
1499 |               "      <td>1.109969</td>\n",
1500 |               "      <td>-0.121359</td>\n",
1501 |               "      <td>-2.261857</td>\n",
1502 |               "      <td>0.524980</td>\n",
1503 |               "      <td>0.247998</td>\n",
1504 |               "      <td>0.771679</td>\n",
1505 |               "      <td>0.909412</td>\n",
1506 |               "      <td>-0.689281</td>\n",
1507 |               "      <td>-0.327642</td>\n",
1508 |               "      <td>-0.139097</td>\n",
1509 |               "      <td>-0.055353</td>\n",
1510 |               "      <td>-0.059752</td>\n",
1511 |               "      <td>378.66</td>\n",
1512 |               "    </tr>\n",
1513 |               "    <tr>\n",
1514 |               "      <th>3</th>\n",
1515 |               "      <td>1.0</td>\n",
1516 |               "      <td>-0.966272</td>\n",
1517 |               "      <td>-0.185226</td>\n",
1518 |               "      <td>1.792993</td>\n",
1519 |               "      <td>-0.863291</td>\n",
1520 |               "      <td>-0.010309</td>\n",
1521 |               "      <td>1.247203</td>\n",
1522 |               "      <td>0.237609</td>\n",
1523 |               "      <td>0.377436</td>\n",
1524 |               "      <td>-1.387024</td>\n",
1525 |               "      <td>-0.054952</td>\n",
1526 |               "      <td>-0.226487</td>\n",
1527 |               "      <td>0.178228</td>\n",
1528 |               "      <td>0.507757</td>\n",
1529 |               "      <td>-0.287924</td>\n",
1530 |               "      <td>-0.631418</td>\n",
1531 |               "      <td>-1.059647</td>\n",
1532 |               "      <td>-0.684093</td>\n",
1533 |               "      <td>1.965775</td>\n",
1534 |               "      <td>-1.232622</td>\n",
1535 |               "      <td>-0.208038</td>\n",
1536 |               "      <td>-0.108300</td>\n",
1537 |               "      <td>0.005274</td>\n",
1538 |               "      <td>-0.190321</td>\n",
1539 |               "      <td>-1.175575</td>\n",
1540 |               "      <td>0.647376</td>\n",
1541 |               "      <td>-0.221929</td>\n",
1542 |               "      <td>0.062723</td>\n",
1543 |               "      <td>0.061458</td>\n",
1544 |               "      <td>123.50</td>\n",
1545 |               "    </tr>\n",
1546 |               "    <tr>\n",
1547 |               "      <th>4</th>\n",
1548 |               "      <td>2.0</td>\n",
1549 |               "      <td>-1.158233</td>\n",
1550 |               "      <td>0.877737</td>\n",
1551 |               "      <td>1.548718</td>\n",
1552 |               "      <td>0.403034</td>\n",
1553 |               "      <td>-0.407193</td>\n",
1554 |               "      <td>0.095921</td>\n",
1555 |               "      <td>0.592941</td>\n",
1556 |               "      <td>-0.270533</td>\n",
1557 |               "      <td>0.817739</td>\n",
1558 |               "      <td>0.753074</td>\n",
1559 |               "      <td>-0.822843</td>\n",
1560 |               "      <td>0.538196</td>\n",
1561 |               "      <td>1.345852</td>\n",
1562 |               "      <td>-1.119670</td>\n",
1563 |               "      <td>0.175121</td>\n",
1564 |               "      <td>-0.451449</td>\n",
1565 |               "      <td>-0.237033</td>\n",
1566 |               "      <td>-0.038195</td>\n",
1567 |               "      <td>0.803487</td>\n",
1568 |               "      <td>0.408542</td>\n",
1569 |               "      <td>-0.009431</td>\n",
1570 |               "      <td>0.798278</td>\n",
1571 |               "      <td>-0.137458</td>\n",
1572 |               "      <td>0.141267</td>\n",
1573 |               "      <td>-0.206010</td>\n",
1574 |               "      <td>0.502292</td>\n",
1575 |               "      <td>0.219422</td>\n",
1576 |               "      <td>0.215153</td>\n",
1577 |               "      <td>69.99</td>\n",
1578 |               "    </tr>\n",
1579 |               "  </tbody>\n",
1580 |               "</table>\n",
1581 |               "</div>"
1582 |             ],
1583 |             "text/plain": [
1584 |               "   Time        V1        V2        V3  ...       V26       V27       V28  Amount\n",
1585 |               "0   0.0 -1.359807 -0.072781  2.536347  ... -0.189115  0.133558 -0.021053  149.62\n",
1586 |               "1   0.0  1.191857  0.266151  0.166480  ...  0.125895 -0.008983  0.014724    2.69\n",
1587 |               "2   1.0 -1.358354 -1.340163  1.773209  ... -0.139097 -0.055353 -0.059752  378.66\n",
1588 |               "3   1.0 -0.966272 -0.185226  1.792993  ... -0.221929  0.062723  0.061458  123.50\n",
1589 |               "4   2.0 -1.158233  0.877737  1.548718  ...  0.502292  0.219422  0.215153   69.99\n",
1590 |               "\n",
1591 |               "[5 rows x 30 columns]"
1592 |             ]
1593 |           },
1594 |           "metadata": {
1595 |             "tags": []
1596 |           },
1597 |           "execution_count": 21
1598 |         }
1599 |       ]
1600 |     },
1601 |     {
1602 |       "cell_type": "code",
1603 |       "metadata": {
1604 |         "id": "LCwuXkafPGKk",
1605 |         "colab_type": "code",
1606 |         "outputId": "c5c29036-8577-4060-c771-61cbc22a243a",
1607 |         "colab": {
1608 |           "base_uri": "https://localhost:8080/",
1609 |           "height": 34
1610 |         }
1611 |       },
1612 |       "source": [
1613 |         "df.shape"
1614 |       ],
1615 |       "execution_count": 22,
1616 |       "outputs": [
1617 |         {
1618 |           "output_type": "execute_result",
1619 |           "data": {
1620 |             "text/plain": [
1621 |               "(284807, 31)"
1622 |             ]
1623 |           },
1624 |           "metadata": {
1625 |             "tags": []
1626 |           },
1627 |           "execution_count": 22
1628 |         }
1629 |       ]
1630 |     },
1631 |     {
1632 |       "cell_type": "markdown",
1633 |       "metadata": {
1634 |         "id": "lp0nQkXX11CV",
1635 |         "colab_type": "text"
1636 |       },
1637 |       "source": [
1638 |         "The data is highly imbalanced. Hence,we train our model before and after resampling .\n",
1639 |         "\n",
1640 |         "# Imbalanced Data : "
1641 |       ]
1642 |     },
1643 |     {
1644 |       "cell_type": "code",
1645 |       "metadata": {
1646 |         "id": "nWAEWbo0dCvy",
1647 |         "colab_type": "code",
1648 |         "colab": {}
1649 |       },
1650 |       "source": [
1651 |         "from sklearn.model_selection import train_test_split\n",
1652 |         "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=42)\n"
1653 |       ],
1654 |       "execution_count": 0,
1655 |       "outputs": []
1656 |     },
1657 |     {
1658 |       "cell_type": "code",
1659 |       "metadata": {
1660 |         "id": "uxFAGvAhWSZ3",
1661 |         "colab_type": "code",
1662 |         "outputId": "2c91170f-33b7-4c8a-e09e-2180ee24e06b",
1663 |         "colab": {
1664 |           "base_uri": "https://localhost:8080/",
1665 |           "height": 238
1666 |         }
1667 |       },
1668 |       "source": [
1669 |         "from sklearn.preprocessing import StandardScaler\n",
1670 |         "scaler = StandardScaler()\n",
1671 |         "scaler.fit(X_train)\n",
1672 |         "scaler.transform(X_train)"
1673 |       ],
1674 |       "execution_count": 24,
1675 |       "outputs": [
1676 |         {
1677 |           "output_type": "execute_result",
1678 |           "data": {
1679 |             "text/plain": [
1680 |               "array([[ 0.4628655 , -0.76417848, -0.58517942, ..., -0.05427856,\n",
1681 |               "         0.47278134,  0.27637606],\n",
1682 |               "       [ 0.99884641, -0.43199832,  0.8362486 , ..., -0.21155863,\n",
1683 |               "        -0.17561255, -0.20845219],\n",
1684 |               "       [-1.06243719, -0.5473776 ,  0.36358059, ..., -0.1751781 ,\n",
1685 |               "         0.27717173, -0.30005801],\n",
1686 |               "       ...,\n",
1687 |               "       [-0.31423311, -0.07216301,  0.59345235, ..., -0.29402614,\n",
1688 |               "        -0.59027941, -0.32887389],\n",
1689 |               "       [-0.1428877 , -1.49506753,  1.40403542, ...,  1.21908694,\n",
1690 |               "         1.01135271, -0.34027614],\n",
1691 |               "       [-0.38613248,  0.62850772, -0.46466388, ...,  0.00552523,\n",
1692 |               "         0.11653329,  0.09409522]])"
1693 |             ]
1694 |           },
1695 |           "metadata": {
1696 |             "tags": []
1697 |           },
1698 |           "execution_count": 24
1699 |         }
1700 |       ]
1701 |     },
1702 |     {
1703 |       "cell_type": "code",
1704 |       "metadata": {
1705 |         "id": "7FXiXz8KdeoJ",
1706 |         "colab_type": "code",
1707 |         "outputId": "e51933e8-611a-4b95-e768-ffd03526e0db",
1708 |         "colab": {
1709 |           "base_uri": "https://localhost:8080/",
1710 |           "height": 153
1711 |         }
1712 |       },
1713 |       "source": [
1714 |         "from sklearn.linear_model import LogisticRegression\n",
1715 |         "clf = LogisticRegression(random_state=0).fit(X_train, y_train)\n",
1716 |         "y_pred = clf.predict(X_test)"
1717 |       ],
1718 |       "execution_count": 25,
1719 |       "outputs": [
1720 |         {
1721 |           "output_type": "stream",
1722 |           "text": [
1723 |             "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
1724 |             "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
1725 |             "\n",
1726 |             "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
1727 |             "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
1728 |             "Please also refer to the documentation for alternative solver options:\n",
1729 |             "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
1730 |             "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
1731 |           ],
1732 |           "name": "stderr"
1733 |         }
1734 |       ]
1735 |     },
1736 |     {
1737 |       "cell_type": "code",
1738 |       "metadata": {
1739 |         "id": "W5UT0_Yvdx58",
1740 |         "colab_type": "code",
1741 |         "outputId": "a2ce191b-5f01-422f-a35c-f9996f4f2dfe",
1742 |         "colab": {
1743 |           "base_uri": "https://localhost:8080/",
1744 |           "height": 34
1745 |         }
1746 |       },
1747 |       "source": [
1748 |         "clf.score(X_test,y_test)"
1749 |       ],
1750 |       "execution_count": 26,
1751 |       "outputs": [
1752 |         {
1753 |           "output_type": "execute_result",
1754 |           "data": {
1755 |             "text/plain": [
1756 |               "0.9990256576810653"
1757 |             ]
1758 |           },
1759 |           "metadata": {
1760 |             "tags": []
1761 |           },
1762 |           "execution_count": 26
1763 |         }
1764 |       ]
1765 |     },
1766 |     {
1767 |       "cell_type": "code",
1768 |       "metadata": {
1769 |         "id": "GWYwNPcceVQA",
1770 |         "colab_type": "code",
1771 |         "outputId": "fdfd0355-caee-43c1-d19b-679cfba50d70",
1772 |         "colab": {
1773 |           "base_uri": "https://localhost:8080/",
1774 |           "height": 34
1775 |         }
1776 |       },
1777 |       "source": [
1778 |         "clf.score(X_train,y_train)"
1779 |       ],
1780 |       "execution_count": 27,
1781 |       "outputs": [
1782 |         {
1783 |           "output_type": "execute_result",
1784 |           "data": {
1785 |             "text/plain": [
1786 |               "0.998987617331055"
1787 |             ]
1788 |           },
1789 |           "metadata": {
1790 |             "tags": []
1791 |           },
1792 |           "execution_count": 27
1793 |         }
1794 |       ]
1795 |     },
1796 |     {
1797 |       "cell_type": "code",
1798 |       "metadata": {
1799 |         "id": "JKMYOLD4ejKE",
1800 |         "colab_type": "code",
1801 |         "outputId": "c267c750-3e35-4150-eb49-2212b1f538a3",
1802 |         "colab": {
1803 |           "base_uri": "https://localhost:8080/",
1804 |           "height": 170
1805 |         }
1806 |       },
1807 |       "source": [
1808 |         " from sklearn.metrics import classification_report\n",
1809 |         " target_names = ['not_fraud', 'fraud']\n",
1810 |         "print(classification_report(y_test, y_pred, target_names=target_names))"
1811 |       ],
1812 |       "execution_count": 28,
1813 |       "outputs": [
1814 |         {
1815 |           "output_type": "stream",
1816 |           "text": [
1817 |             "              precision    recall  f1-score   support\n",
1818 |             "\n",
1819 |             "   not_fraud       1.00      1.00      1.00    113732\n",
1820 |             "       fraud       0.73      0.67      0.70       191\n",
1821 |             "\n",
1822 |             "    accuracy                           1.00    113923\n",
1823 |             "   macro avg       0.86      0.83      0.85    113923\n",
1824 |             "weighted avg       1.00      1.00      1.00    113923\n",
1825 |             "\n"
1826 |           ],
1827 |           "name": "stdout"
1828 |         }
1829 |       ]
1830 |     },
1831 |     {
1832 |       "cell_type": "code",
1833 |       "metadata": {
1834 |         "id": "2Yg_RgmVTjZt",
1835 |         "colab_type": "code",
1836 |         "outputId": "422a7c02-5bda-4ee4-8c75-37b89f76ae83",
1837 |         "colab": {
1838 |           "base_uri": "https://localhost:8080/",
1839 |           "height": 34
1840 |         }
1841 |       },
1842 |       "source": [
1843 |         "from sklearn.metrics import log_loss\n",
1844 |         "log_loss(y_test,y_pred)"
1845 |       ],
1846 |       "execution_count": 29,
1847 |       "outputs": [
1848 |         {
1849 |           "output_type": "execute_result",
1850 |           "data": {
1851 |             "text/plain": [
1852 |               "0.033652928385943554"
1853 |             ]
1854 |           },
1855 |           "metadata": {
1856 |             "tags": []
1857 |           },
1858 |           "execution_count": 29
1859 |         }
1860 |       ]
1861 |     },
1862 |     {
1863 |       "cell_type": "code",
1864 |       "metadata": {
1865 |         "id": "NzmzFLoZmV5s",
1866 |         "colab_type": "code",
1867 |         "outputId": "e9b4117d-4d23-45d0-f261-cb8616770c1d",
1868 |         "colab": {
1869 |           "base_uri": "https://localhost:8080/",
1870 |           "height": 221
1871 |         }
1872 |       },
1873 |       "source": [
1874 |         "from sklearn.tree import DecisionTreeClassifier\n",
1875 |         "dtc = DecisionTreeClassifier(random_state=42)\n",
1876 |         "dtc.fit(X_train,y_train)\n",
1877 |         "y_pred = dtc.predict(X_test)\n",
1878 |         "print(\"Test accuracy\",dtc.score(X_test,y_test))\n",
1879 |         "print(\"Train accuracy\",dtc.score(X_train,y_train))\n",
1880 |         "print(\"Classification report :\")\n",
1881 |         "print(classification_report(y_test,y_pred, target_names=target_names))"
1882 |       ],
1883 |       "execution_count": 30,
1884 |       "outputs": [
1885 |         {
1886 |           "output_type": "stream",
1887 |           "text": [
1888 |             "Test accuracy 0.9991222141270858\n",
1889 |             "Train accuracy 1.0\n",
1890 |             "Classification report :\n",
1891 |             "              precision    recall  f1-score   support\n",
1892 |             "\n",
1893 |             "   not_fraud       1.00      1.00      1.00    113732\n",
1894 |             "       fraud       0.71      0.80      0.75       191\n",
1895 |             "\n",
1896 |             "    accuracy                           1.00    113923\n",
1897 |             "   macro avg       0.86      0.90      0.88    113923\n",
1898 |             "weighted avg       1.00      1.00      1.00    113923\n",
1899 |             "\n"
1900 |           ],
1901 |           "name": "stdout"
1902 |         }
1903 |       ]
1904 |     },
1905 |     {
1906 |       "cell_type": "code",
1907 |       "metadata": {
1908 |         "id": "6A62wpyeWDRg",
1909 |         "colab_type": "code",
1910 |         "outputId": "c530c3e8-b7d8-487b-f4f4-363baf300bbc",
1911 |         "colab": {
1912 |           "base_uri": "https://localhost:8080/",
1913 |           "height": 221
1914 |         }
1915 |       },
1916 |       "source": [
1917 |         "from sklearn.ensemble import RandomForestClassifier\n",
1918 |         "from sklearn.datasets import make_classification\n",
1919 |         "rfc = RandomForestClassifier(max_depth=2, random_state=42,n_estimators=10)\n",
1920 |         "rfc.fit(X_train, y_train)\n",
1921 |         "y_pred = rfc.predict(X_test)\n",
1922 |         "print(\"Test accuracy\",rfc.score(X_test,y_test))\n",
1923 |         "print(\"Train accuracy\",rfc.score(X_train,y_train))\n",
1924 |         "print(\"Classification report :\")\n",
1925 |         "print(classification_report(y_test,y_pred, target_names=target_names))"
1926 |       ],
1927 |       "execution_count": 31,
1928 |       "outputs": [
1929 |         {
1930 |           "output_type": "stream",
1931 |           "text": [
1932 |             "Test accuracy 0.9991748812794606\n",
1933 |             "Train accuracy 0.9991573230963695\n",
1934 |             "Classification report :\n",
1935 |             "              precision    recall  f1-score   support\n",
1936 |             "\n",
1937 |             "   not_fraud       1.00      1.00      1.00    113732\n",
1938 |             "       fraud       0.86      0.60      0.71       191\n",
1939 |             "\n",
1940 |             "    accuracy                           1.00    113923\n",
1941 |             "   macro avg       0.93      0.80      0.85    113923\n",
1942 |             "weighted avg       1.00      1.00      1.00    113923\n",
1943 |             "\n"
1944 |           ],
1945 |           "name": "stdout"
1946 |         }
1947 |       ]
1948 |     },
1949 |     {
1950 |       "cell_type": "code",
1951 |       "metadata": {
1952 |         "id": "r-A-rJ16X0Eh",
1953 |         "colab_type": "code",
1954 |         "outputId": "73b7ca6a-45f6-4643-8b92-22aedc595440",
1955 |         "colab": {
1956 |           "base_uri": "https://localhost:8080/",
1957 |           "height": 221
1958 |         }
1959 |       },
1960 |       "source": [
1961 |         "from sklearn.pipeline import make_pipeline\n",
1962 |         "from sklearn.svm import SVC\n",
1963 |         "svc = make_pipeline(StandardScaler(), SVC(gamma='auto',random_state=42,kernel='sigmoid'))\n",
1964 |         "svc.fit(X_train, y_train)\n",
1965 |         "y_pred = svc.predict(X_test)\n",
1966 |         "print(\"Test accuracy\",svc.score(X_test,y_test))\n",
1967 |         "print(\"Train accuracy\",svc.score(X_train,y_train))\n",
1968 |         "print(\"Classification report :\")\n",
1969 |         "print(classification_report(y_pred,y_test, target_names=target_names))"
1970 |       ],
1971 |       "execution_count": 32,
1972 |       "outputs": [
1973 |         {
1974 |           "output_type": "stream",
1975 |           "text": [
1976 |             "Test accuracy 0.9986130983207956\n",
1977 |             "Train accuracy 0.9984024250368672\n",
1978 |             "Classification report :\n",
1979 |             "              precision    recall  f1-score   support\n",
1980 |             "\n",
1981 |             "   not_fraud       1.00      1.00      1.00    113750\n",
1982 |             "       fraud       0.54      0.60      0.57       173\n",
1983 |             "\n",
1984 |             "    accuracy                           1.00    113923\n",
1985 |             "   macro avg       0.77      0.80      0.78    113923\n",
1986 |             "weighted avg       1.00      1.00      1.00    113923\n",
1987 |             "\n"
1988 |           ],
1989 |           "name": "stdout"
1990 |         }
1991 |       ]
1992 |     },
1993 |     {
1994 |       "cell_type": "markdown",
1995 |       "metadata": {
1996 |         "id": "XrAZ1c-b1lma",
1997 |         "colab_type": "text"
1998 |       },
1999 |       "source": [
2000 |         "# **Neural Networks**"
2001 |       ]
2002 |     },
2003 |     {
2004 |       "cell_type": "code",
2005 |       "metadata": {
2006 |         "id": "w1kV1lunr0Ne",
2007 |         "colab_type": "code",
2008 |         "colab": {}
2009 |       },
2010 |       "source": [
2011 |         "#Using neural networks \n",
2012 |         "import tensorflow as tf\n",
2013 |         "from tensorflow.keras.callbacks import EarlyStopping\n",
2014 |         "from tensorflow.keras.layers import Dense,Activation,Flatten\n",
2015 |         "from tensorflow.keras import Sequential"
2016 |       ],
2017 |       "execution_count": 0,
2018 |       "outputs": []
2019 |     },
2020 |     {
2021 |       "cell_type": "code",
2022 |       "metadata": {
2023 |         "id": "DvEVDxuCgHrx",
2024 |         "colab_type": "code",
2025 |         "outputId": "0fbb6f51-28a8-425c-89de-c3fd9c8480cd",
2026 |         "colab": {
2027 |           "base_uri": "https://localhost:8080/",
2028 |           "height": 428
2029 |         }
2030 |       },
2031 |       "source": [
2032 |         "model = Sequential()\n",
2033 |         "model.add(Flatten())\n",
2034 |         "model.add(Dense(2,activation='relu'))\n",
2035 |         "model.add(Dense(2,activation='relu'))\n",
2036 |         "model.add(Dense(2,activation='relu'))\n",
2037 |         "model.add(Dense(2,activation='relu'))\n",
2038 |         "model.add(Dense(1,activation='sigmoid'))\n",
2039 |         "\n",
2040 |         "earlystop = EarlyStopping(monitor='val_loss',patience=2,verbose=0,mode='min')\n",
2041 |         "\n",
2042 |         "model.compile(optimizer='adam',loss = 'binary_crossentropy',metrics = ['accuracy'])\n",
2043 |         "\n",
2044 |         "model.fit(X_train,y_train,epochs =10,validation_data = (X_test,y_test),callbacks = [earlystop])\n"
2045 |       ],
2046 |       "execution_count": 34,
2047 |       "outputs": [
2048 |         {
2049 |           "output_type": "stream",
2050 |           "text": [
2051 |             "Epoch 1/10\n",
2052 |             "WARNING:tensorflow:Layer flatten is casting an input tensor from dtype float64 to the layer's dtype of float32, which is new behavior in TensorFlow 2.  The layer has dtype float32 because it's dtype defaults to floatx.\n",
2053 |             "\n",
2054 |             "If you intended to run this layer in float32, you can safely ignore this warning. If in doubt, this warning is likely only an issue if you are porting a TensorFlow 1.X model to TensorFlow 2.\n",
2055 |             "\n",
2056 |             "To change all layers to have dtype float64 by default, call `tf.keras.backend.set_floatx('float64')`. To change just this layer, pass dtype='float64' to the layer constructor. If you are the author of this layer, you can disable autocasting by passing autocast=False to the base Layer constructor.\n",
2057 |             "\n",
2058 |             "5341/5341 [==============================] - 10s 2ms/step - loss: 0.1994 - accuracy: 0.9982 - val_loss: 0.0415 - val_accuracy: 0.9983\n",
2059 |             "Epoch 2/10\n",
2060 |             "5341/5341 [==============================] - 10s 2ms/step - loss: 0.0223 - accuracy: 0.9982 - val_loss: 0.0138 - val_accuracy: 0.9983\n",
2061 |             "Epoch 3/10\n",
2062 |             "5341/5341 [==============================] - 10s 2ms/step - loss: 0.0133 - accuracy: 0.9982 - val_loss: 0.0125 - val_accuracy: 0.9983\n",
2063 |             "Epoch 4/10\n",
2064 |             "5341/5341 [==============================] - 10s 2ms/step - loss: 0.0129 - accuracy: 0.9982 - val_loss: 0.0124 - val_accuracy: 0.9983\n",
2065 |             "Epoch 5/10\n",
2066 |             "5341/5341 [==============================] - 10s 2ms/step - loss: 0.0129 - accuracy: 0.9982 - val_loss: 0.0124 - val_accuracy: 0.9983\n",
2067 |             "Epoch 6/10\n",
2068 |             "5341/5341 [==============================] - 10s 2ms/step - loss: 0.0129 - accuracy: 0.9982 - val_loss: 0.0124 - val_accuracy: 0.9983\n",
2069 |             "Epoch 7/10\n",
2070 |             "5341/5341 [==============================] - 10s 2ms/step - loss: 0.0129 - accuracy: 0.9982 - val_loss: 0.0124 - val_accuracy: 0.9983\n",
2071 |             "Epoch 8/10\n",
2072 |             "5341/5341 [==============================] - 10s 2ms/step - loss: 0.0129 - accuracy: 0.9982 - val_loss: 0.0124 - val_accuracy: 0.9983\n"
2073 |           ],
2074 |           "name": "stdout"
2075 |         },
2076 |         {
2077 |           "output_type": "execute_result",
2078 |           "data": {
2079 |             "text/plain": [
2080 |               "<tensorflow.python.keras.callbacks.History at 0x7f771cabed68>"
2081 |             ]
2082 |           },
2083 |           "metadata": {
2084 |             "tags": []
2085 |           },
2086 |           "execution_count": 34
2087 |         }
2088 |       ]
2089 |     },
2090 |     {
2091 |       "cell_type": "code",
2092 |       "metadata": {
2093 |         "id": "o_cG0yexswSr",
2094 |         "colab_type": "code",
2095 |         "outputId": "af67a168-2b9a-42a6-c4ee-254ad4ecb427",
2096 |         "colab": {
2097 |           "base_uri": "https://localhost:8080/",
2098 |           "height": 282
2099 |         }
2100 |       },
2101 |       "source": [
2102 |         "loss = pd.DataFrame(model.history.history)\n",
2103 |         "loss.plot()"
2104 |       ],
2105 |       "execution_count": 35,
2106 |       "outputs": [
2107 |         {
2108 |           "output_type": "execute_result",
2109 |           "data": {
2110 |             "text/plain": [
2111 |               "<matplotlib.axes._subplots.AxesSubplot at 0x7f77182bc908>"
2112 |             ]
2113 |           },
2114 |           "metadata": {
2115 |             "tags": []
2116 |           },
2117 |           "execution_count": 35
2118 |         },
2119 |         {
2120 |           "output_type": "display_data",
2121 |           "data": {
2122 |             "image/png": "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\n",
2123 |             "text/plain": [
2124 |               "<Figure size 432x288 with 1 Axes>"
2125 |             ]
2126 |           },
2127 |           "metadata": {
2128 |             "tags": [],
2129 |             "needs_background": "light"
2130 |           }
2131 |         }
2132 |       ]
2133 |     },
2134 |     {
2135 |       "cell_type": "markdown",
2136 |       "metadata": {
2137 |         "id": "Ytnu_pa-yGhD",
2138 |         "colab_type": "text"
2139 |       },
2140 |       "source": [
2141 |         "# Let's try after resampling data \n",
2142 |         "# Oversampling the data : \n",
2143 |         "\n"
2144 |       ]
2145 |     },
2146 |     {
2147 |       "cell_type": "code",
2148 |       "metadata": {
2149 |         "id": "RQyQ-jNitbmo",
2150 |         "colab_type": "code",
2151 |         "outputId": "71832729-e1c1-48d9-9869-2670e9bc4058",
2152 |         "colab": {
2153 |           "base_uri": "https://localhost:8080/",
2154 |           "height": 68
2155 |         }
2156 |       },
2157 |       "source": [
2158 |         "#Since the data is imbalanced ,Resampling :\n",
2159 |         " \n",
2160 |         "#Oversampling \n",
2161 |         " \n",
2162 |         "from sklearn.utils import resample\n",
2163 |         "\n",
2164 |         "# Separate input features and target\n",
2165 |         "y = df.Class\n",
2166 |         "X = df.drop('Class', axis=1)\n",
2167 |         "\n",
2168 |         "# setting up testing and training sets\n",
2169 |         "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=27)\n",
2170 |         "\n",
2171 |         "# concatenate our training data back together\n",
2172 |         "X = pd.concat([X_train, y_train], axis=1)\n",
2173 |         "\n",
2174 |         "# separate minority and majority classes\n",
2175 |         "not_fraud = X[X.Class==0]\n",
2176 |         "fraud = X[X.Class==1]\n",
2177 |         "\n",
2178 |         "# upsample minority\n",
2179 |         "fraud_upsampled = resample(fraud,\n",
2180 |         "                          replace=True, # sample with replacement\n",
2181 |         "                          n_samples=len(not_fraud), # match number in majority class\n",
2182 |         "                          random_state=27) # reproducible results\n",
2183 |         "\n",
2184 |         "# combine majority and upsampled minority\n",
2185 |         "upsampled = pd.concat([not_fraud, fraud_upsampled])\n",
2186 |         "\n",
2187 |         "# check new class counts\n",
2188 |         "upsampled.Class.value_counts()"
2189 |       ],
2190 |       "execution_count": 36,
2191 |       "outputs": [
2192 |         {
2193 |           "output_type": "execute_result",
2194 |           "data": {
2195 |             "text/plain": [
2196 |               "1    213245\n",
2197 |               "0    213245\n",
2198 |               "Name: Class, dtype: int64"
2199 |             ]
2200 |           },
2201 |           "metadata": {
2202 |             "tags": []
2203 |           },
2204 |           "execution_count": 36
2205 |         }
2206 |       ]
2207 |     },
2208 |     {
2209 |       "cell_type": "code",
2210 |       "metadata": {
2211 |         "id": "CniNA7EevVdr",
2212 |         "colab_type": "code",
2213 |         "outputId": "376b977d-2fbf-4890-a01e-6c7d75c3f923",
2214 |         "colab": {
2215 |           "base_uri": "https://localhost:8080/",
2216 |           "height": 357
2217 |         }
2218 |       },
2219 |       "source": [
2220 |         "\n",
2221 |         "# trying logistic regression again with the balanced dataset\n",
2222 |         "y_train = upsampled.Class\n",
2223 |         "X_train = upsampled.drop('Class', axis=1)\n",
2224 |         "\n",
2225 |         "rfc = LogisticRegression()\n",
2226 |         "rfc.fit(X_train, y_train)\n",
2227 |         "y_pred = rfc.predict(X_test)\n",
2228 |         "print(\"Test accuracy\",rfc.score(X_test,y_test))\n",
2229 |         "print(\"Train accuracy\",rfc.score(X_train,y_train))\n",
2230 |         "print(\"Classification report :\")\n",
2231 |         "print(classification_report(y_test,y_pred, target_names=target_names))"
2232 |       ],
2233 |       "execution_count": 37,
2234 |       "outputs": [
2235 |         {
2236 |           "output_type": "stream",
2237 |           "text": [
2238 |             "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
2239 |             "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
2240 |             "\n",
2241 |             "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
2242 |             "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
2243 |             "Please also refer to the documentation for alternative solver options:\n",
2244 |             "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
2245 |             "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
2246 |           ],
2247 |           "name": "stderr"
2248 |         },
2249 |         {
2250 |           "output_type": "stream",
2251 |           "text": [
2252 |             "Test accuracy 0.9677818038819135\n",
2253 |             "Train accuracy 0.9500246195690403\n",
2254 |             "Classification report :\n",
2255 |             "              precision    recall  f1-score   support\n",
2256 |             "\n",
2257 |             "   not_fraud       1.00      0.97      0.98     71070\n",
2258 |             "       fraud       0.05      0.89      0.09       132\n",
2259 |             "\n",
2260 |             "    accuracy                           0.97     71202\n",
2261 |             "   macro avg       0.52      0.93      0.54     71202\n",
2262 |             "weighted avg       1.00      0.97      0.98     71202\n",
2263 |             "\n"
2264 |           ],
2265 |           "name": "stdout"
2266 |         }
2267 |       ]
2268 |     },
2269 |     {
2270 |       "cell_type": "markdown",
2271 |       "metadata": {
2272 |         "id": "hj6c_zsQxkU4",
2273 |         "colab_type": "text"
2274 |       },
2275 |       "source": [
2276 |         "# **Undersampling the data:**"
2277 |       ]
2278 |     },
2279 |     {
2280 |       "cell_type": "code",
2281 |       "metadata": {
2282 |         "id": "KRRVvMFJvlcx",
2283 |         "colab_type": "code",
2284 |         "outputId": "a672ac9c-2030-4f5b-b0b7-679e7ed988dc",
2285 |         "colab": {
2286 |           "base_uri": "https://localhost:8080/",
2287 |           "height": 221
2288 |         }
2289 |       },
2290 |       "source": [
2291 |         "# still using our separated classes fraud and not_fraud from above\n",
2292 |         "\n",
2293 |         "# downsample majority\n",
2294 |         "not_fraud_downsampled = resample(not_fraud,\n",
2295 |         "                                replace = False, # sample without replacement\n",
2296 |         "                                n_samples = len(fraud), # match minority n\n",
2297 |         "                                random_state = 27) # reproducible results\n",
2298 |         "\n",
2299 |         "# combine minority and downsampled majority\n",
2300 |         "downsampled = pd.concat([not_fraud_downsampled, fraud])\n",
2301 |         "\n",
2302 |         "y_train = downsampled.Class\n",
2303 |         "X_train = downsampled.drop('Class', axis=1)\n",
2304 |         "\n",
2305 |         "rfc = RandomForestClassifier(max_depth=2, random_state=42,n_estimators=10)\n",
2306 |         "rfc.fit(X_train, y_train)\n",
2307 |         "y_pred = rfc.predict(X_test)\n",
2308 |         "print(\"Test accuracy\",rfc.score(X_test,y_test))\n",
2309 |         "print(\"Train accuracy\",rfc.score(X_train,y_train))\n",
2310 |         "print(\"Classification report :\")\n",
2311 |         "print(classification_report(y_test,y_pred, target_names=target_names))"
2312 |       ],
2313 |       "execution_count": 38,
2314 |       "outputs": [
2315 |         {
2316 |           "output_type": "stream",
2317 |           "text": [
2318 |             "Test accuracy 0.9829780062357799\n",
2319 |             "Train accuracy 0.9361111111111111\n",
2320 |             "Classification report :\n",
2321 |             "              precision    recall  f1-score   support\n",
2322 |             "\n",
2323 |             "   not_fraud       1.00      0.98      0.99     71070\n",
2324 |             "       fraud       0.09      0.86      0.16       132\n",
2325 |             "\n",
2326 |             "    accuracy                           0.98     71202\n",
2327 |             "   macro avg       0.54      0.92      0.57     71202\n",
2328 |             "weighted avg       1.00      0.98      0.99     71202\n",
2329 |             "\n"
2330 |           ],
2331 |           "name": "stdout"
2332 |         }
2333 |       ]
2334 |     },
2335 |     {
2336 |       "cell_type": "markdown",
2337 |       "metadata": {
2338 |         "id": "cYIWz7rc3NkB",
2339 |         "colab_type": "text"
2340 |       },
2341 |       "source": [
2342 |         "# **Generate synthetic samples using SMOTE**"
2343 |       ]
2344 |     },
2345 |     {
2346 |       "cell_type": "code",
2347 |       "metadata": {
2348 |         "id": "EHgSzYSwwxOE",
2349 |         "colab_type": "code",
2350 |         "outputId": "d4c40d9e-1f35-4998-f248-35af4b0584fd",
2351 |         "colab": {
2352 |           "base_uri": "https://localhost:8080/",
2353 |           "height": 343
2354 |         }
2355 |       },
2356 |       "source": [
2357 |         "from imblearn.over_sampling import SMOTE\n",
2358 |         "\n",
2359 |         "# Separate input features and target\n",
2360 |         "y = df.Class\n",
2361 |         "X = df.drop('Class', axis=1)\n",
2362 |         "\n",
2363 |         "# setting up testing and training sets\n",
2364 |         "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=27)\n",
2365 |         "\n",
2366 |         "sm = SMOTE(random_state=27, ratio=1.0)\n",
2367 |         "X_train, y_train = sm.fit_sample(X_train, y_train)\n",
2368 |         "\n",
2369 |         "\n",
2370 |         "rfc = RandomForestClassifier(max_depth=2, random_state=42,n_estimators=10)\n",
2371 |         "rfc.fit(X_train, y_train)\n",
2372 |         "y_pred = rfc.predict(X_test)\n",
2373 |         "print(\"Test accuracy\",rfc.score(X_test,y_test))\n",
2374 |         "print(\"Train accuracy\",rfc.score(X_train,y_train))\n",
2375 |         "print(\"Classification report :\")\n",
2376 |         "print(classification_report(y_test,y_pred, target_names=target_names))"
2377 |       ],
2378 |       "execution_count": 39,
2379 |       "outputs": [
2380 |         {
2381 |           "output_type": "stream",
2382 |           "text": [
2383 |             "/usr/local/lib/python3.6/dist-packages/sklearn/externals/six.py:31: FutureWarning: The module is deprecated in version 0.21 and will be removed in version 0.23 since we've dropped support for Python 2.7. Please rely on the official version of six (https://pypi.org/project/six/).\n",
2384 |             "  \"(https://pypi.org/project/six/).\", FutureWarning)\n",
2385 |             "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:144: FutureWarning: The sklearn.neighbors.base module is  deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.neighbors. Anything that cannot be imported from sklearn.neighbors is now part of the private API.\n",
2386 |             "  warnings.warn(message, FutureWarning)\n",
2387 |             "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function safe_indexing is deprecated; safe_indexing is deprecated in version 0.22 and will be removed in version 0.24.\n",
2388 |             "  warnings.warn(msg, category=FutureWarning)\n"
2389 |           ],
2390 |           "name": "stderr"
2391 |         },
2392 |         {
2393 |           "output_type": "stream",
2394 |           "text": [
2395 |             "Test accuracy 0.98706497008511\n",
2396 |             "Train accuracy 0.9605993106520668\n",
2397 |             "Classification report :\n",
2398 |             "              precision    recall  f1-score   support\n",
2399 |             "\n",
2400 |             "   not_fraud       1.00      0.99      0.99     71070\n",
2401 |             "       fraud       0.11      0.85      0.20       132\n",
2402 |             "\n",
2403 |             "    accuracy                           0.99     71202\n",
2404 |             "   macro avg       0.56      0.92      0.59     71202\n",
2405 |             "weighted avg       1.00      0.99      0.99     71202\n",
2406 |             "\n"
2407 |           ],
2408 |           "name": "stdout"
2409 |         }
2410 |       ]
2411 |     },
2412 |     {
2413 |       "cell_type": "code",
2414 |       "metadata": {
2415 |         "id": "0EbTf0QGfHLf",
2416 |         "colab_type": "code",
2417 |         "colab": {}
2418 |       },
2419 |       "source": [
2420 |         ""
2421 |       ],
2422 |       "execution_count": 0,
2423 |       "outputs": []
2424 |     }
2425 |   ]
2426 | }


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