├── Labels.xlsx
├── README.md
├── rClassify Trained Models.ipynb
└── rClassify preprocessing Demo.ipynb


/Labels.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/sumitrj/rClassify/1d3e3ebf4e43d4fc05a07e0fcc8ed0c84ac26df0/Labels.xlsx


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # Retinopathy-Detection
 2 | 
 3 | ## Dataset Description:
 4 | Over 200 Images of retina, obtained by Fundus imaging were acquired and analyzed. 100 of them were normal and tagged as 0, and 100 were affected by diabetic retinopathy and tagged as 1. 
 5 | 
 6 | ## Preprocessing
 7 | The images were preprocessed to highlight the sites of microaneurysm and suppress all other features. 
 8 | 
 9 | ## Regression Analysis
10 | These images were then used to train machine learning models based on SVM, Random Forest, Decision Tree and K Nearest Neighbors. The train and test data were split into 60-40, 70-30 and 80-20 combinations of train and test ratios. The best  model was the random forest model obtained by training over 70% of data.
11 | 
12 | ## Dataset is subject to privacy concerns. 
13 | Will be shared soon.
14 | 


--------------------------------------------------------------------------------
/rClassify Trained Models.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |  "cells": [
   3 |   {
   4 |    "cell_type": "markdown",
   5 |    "metadata": {},
   6 |    "source": [
   7 |     "## Imports"
   8 |    ]
   9 |   },
  10 |   {
  11 |    "cell_type": "code",
  12 |    "execution_count": 1,
  13 |    "metadata": {},
  14 |    "outputs": [],
  15 |    "source": [
  16 |     "import matplotlib.pyplot as plt\n",
  17 |     "import numpy as np\n",
  18 |     "import pandas as pd\n",
  19 |     "\n",
  20 |     "import cv2\n",
  21 |     "from scipy.ndimage.filters import gaussian_filter\n",
  22 |     "\n",
  23 |     "from sklearn.model_selection import train_test_split\n",
  24 |     "from sklearn.metrics import classification_report\n",
  25 |     "from pandas_ml import ConfusionMatrix\n",
  26 |     "\n",
  27 |     "from sklearn.svm import SVC\n",
  28 |     "from sklearn.ensemble import RandomForestClassifier\n",
  29 |     "from sklearn.tree import DecisionTreeClassifier\n",
  30 |     "from sklearn.neighbors import KNeighborsClassifier"
  31 |    ]
  32 |   },
  33 |   {
  34 |    "cell_type": "markdown",
  35 |    "metadata": {},
  36 |    "source": [
  37 |     "## Read Dataset"
  38 |    ]
  39 |   },
  40 |   {
  41 |    "cell_type": "code",
  42 |    "execution_count": 2,
  43 |    "metadata": {},
  44 |    "outputs": [
  45 |     {
  46 |      "data": {
  47 |       "text/html": [
  48 |        "<div>\n",
  49 |        "<style scoped>\n",
  50 |        "    .dataframe tbody tr th:only-of-type {\n",
  51 |        "        vertical-align: middle;\n",
  52 |        "    }\n",
  53 |        "\n",
  54 |        "    .dataframe tbody tr th {\n",
  55 |        "        vertical-align: top;\n",
  56 |        "    }\n",
  57 |        "\n",
  58 |        "    .dataframe thead th {\n",
  59 |        "        text-align: right;\n",
  60 |        "    }\n",
  61 |        "</style>\n",
  62 |        "<table border=\"1\" class=\"dataframe\">\n",
  63 |        "  <thead>\n",
  64 |        "    <tr style=\"text-align: right;\">\n",
  65 |        "      <th></th>\n",
  66 |        "      <th>Image_ID</th>\n",
  67 |        "      <th>Label</th>\n",
  68 |        "    </tr>\n",
  69 |        "  </thead>\n",
  70 |        "  <tbody>\n",
  71 |        "    <tr>\n",
  72 |        "      <th>0</th>\n",
  73 |        "      <td>002c21358ce6</td>\n",
  74 |        "      <td>0</td>\n",
  75 |        "    </tr>\n",
  76 |        "    <tr>\n",
  77 |        "      <th>1</th>\n",
  78 |        "      <td>005b95c28852</td>\n",
  79 |        "      <td>0</td>\n",
  80 |        "    </tr>\n",
  81 |        "    <tr>\n",
  82 |        "      <th>2</th>\n",
  83 |        "      <td>0097f532ac9f</td>\n",
  84 |        "      <td>0</td>\n",
  85 |        "    </tr>\n",
  86 |        "    <tr>\n",
  87 |        "      <th>3</th>\n",
  88 |        "      <td>00cc2b75cddd</td>\n",
  89 |        "      <td>0</td>\n",
  90 |        "    </tr>\n",
  91 |        "    <tr>\n",
  92 |        "      <th>4</th>\n",
  93 |        "      <td>00f6c1be5a33</td>\n",
  94 |        "      <td>0</td>\n",
  95 |        "    </tr>\n",
  96 |        "  </tbody>\n",
  97 |        "</table>\n",
  98 |        "</div>"
  99 |       ],
 100 |       "text/plain": [
 101 |        "       Image_ID  Label\n",
 102 |        "0  002c21358ce6      0\n",
 103 |        "1  005b95c28852      0\n",
 104 |        "2  0097f532ac9f      0\n",
 105 |        "3  00cc2b75cddd      0\n",
 106 |        "4  00f6c1be5a33      0"
 107 |       ]
 108 |      },
 109 |      "execution_count": 2,
 110 |      "metadata": {},
 111 |      "output_type": "execute_result"
 112 |     }
 113 |    ],
 114 |    "source": [
 115 |     "df = pd.read_excel(\"Labels.xlsx\")\n",
 116 |     "df.head()"
 117 |    ]
 118 |   },
 119 |   {
 120 |    "cell_type": "code",
 121 |    "execution_count": 3,
 122 |    "metadata": {},
 123 |    "outputs": [
 124 |     {
 125 |      "data": {
 126 |       "text/html": [
 127 |        "<div>\n",
 128 |        "<style scoped>\n",
 129 |        "    .dataframe tbody tr th:only-of-type {\n",
 130 |        "        vertical-align: middle;\n",
 131 |        "    }\n",
 132 |        "\n",
 133 |        "    .dataframe tbody tr th {\n",
 134 |        "        vertical-align: top;\n",
 135 |        "    }\n",
 136 |        "\n",
 137 |        "    .dataframe thead th {\n",
 138 |        "        text-align: right;\n",
 139 |        "    }\n",
 140 |        "</style>\n",
 141 |        "<table border=\"1\" class=\"dataframe\">\n",
 142 |        "  <thead>\n",
 143 |        "    <tr style=\"text-align: right;\">\n",
 144 |        "      <th></th>\n",
 145 |        "      <th>Label</th>\n",
 146 |        "    </tr>\n",
 147 |        "  </thead>\n",
 148 |        "  <tbody>\n",
 149 |        "    <tr>\n",
 150 |        "      <th>count</th>\n",
 151 |        "      <td>200.000000</td>\n",
 152 |        "    </tr>\n",
 153 |        "    <tr>\n",
 154 |        "      <th>mean</th>\n",
 155 |        "      <td>0.500000</td>\n",
 156 |        "    </tr>\n",
 157 |        "    <tr>\n",
 158 |        "      <th>std</th>\n",
 159 |        "      <td>0.501255</td>\n",
 160 |        "    </tr>\n",
 161 |        "    <tr>\n",
 162 |        "      <th>min</th>\n",
 163 |        "      <td>0.000000</td>\n",
 164 |        "    </tr>\n",
 165 |        "    <tr>\n",
 166 |        "      <th>25%</th>\n",
 167 |        "      <td>0.000000</td>\n",
 168 |        "    </tr>\n",
 169 |        "    <tr>\n",
 170 |        "      <th>50%</th>\n",
 171 |        "      <td>0.500000</td>\n",
 172 |        "    </tr>\n",
 173 |        "    <tr>\n",
 174 |        "      <th>75%</th>\n",
 175 |        "      <td>1.000000</td>\n",
 176 |        "    </tr>\n",
 177 |        "    <tr>\n",
 178 |        "      <th>max</th>\n",
 179 |        "      <td>1.000000</td>\n",
 180 |        "    </tr>\n",
 181 |        "  </tbody>\n",
 182 |        "</table>\n",
 183 |        "</div>"
 184 |       ],
 185 |       "text/plain": [
 186 |        "            Label\n",
 187 |        "count  200.000000\n",
 188 |        "mean     0.500000\n",
 189 |        "std      0.501255\n",
 190 |        "min      0.000000\n",
 191 |        "25%      0.000000\n",
 192 |        "50%      0.500000\n",
 193 |        "75%      1.000000\n",
 194 |        "max      1.000000"
 195 |       ]
 196 |      },
 197 |      "execution_count": 3,
 198 |      "metadata": {},
 199 |      "output_type": "execute_result"
 200 |     }
 201 |    ],
 202 |    "source": [
 203 |     "df.describe()"
 204 |    ]
 205 |   },
 206 |   {
 207 |    "cell_type": "markdown",
 208 |    "metadata": {},
 209 |    "source": [
 210 |     "## Function to  read, preprocess and flatten images"
 211 |    ]
 212 |   },
 213 |   {
 214 |    "cell_type": "code",
 215 |    "execution_count": 4,
 216 |    "metadata": {},
 217 |    "outputs": [],
 218 |    "source": [
 219 |     "def read_preprocess(name):\n",
 220 |     "    path = 'ImageData/' + name + '.png'\n",
 221 |     "    frame = cv2.imread(path)\n",
 222 |     "    imG = cv2.equalizeHist(frame[:,:,1])\n",
 223 |     "    blG = gaussian_filter(np.max(imG) - imG, sigma=50)  \n",
 224 |     "    fG = ( blG < np.mean(blG) )* (np.max(imG) - imG)\n",
 225 |     "    pifG = (fG/np.mean(fG))**2\n",
 226 |     "    pifG = (pifG * (255/np.max(pifG))).astype(int)\n",
 227 |     "    frame[:,:,1] = pifG\n",
 228 |     "    fr = cv2.resize(frame,(256,256), interpolation = cv2.INTER_AREA)\n",
 229 |     "    fg = fr[:,:,1]\n",
 230 |     "    fg1 = np.max(fg**2) - fg**2\n",
 231 |     "    fg2 = (fg1>245) * fg1\n",
 232 |     "    fg3 = (fg2>=248) * 255\n",
 233 |     "    \n",
 234 |     "    return (fg3.reshape(256*256))/255  "
 235 |    ]
 236 |   },
 237 |   {
 238 |    "cell_type": "markdown",
 239 |    "metadata": {},
 240 |    "source": []
 241 |   },
 242 |   {
 243 |    "cell_type": "code",
 244 |    "execution_count": 5,
 245 |    "metadata": {},
 246 |    "outputs": [],
 247 |    "source": [
 248 |     "def CR(Ypred,Ytrue):\n",
 249 |     "    rep = classification_report(y_pred = Ypred ,y_true=Ytrue,)\n",
 250 |     "    det = np.array( [ [float(x.strip()) for x in rep.split('\\n')[2].split('     ')[3:6] ], [float(x.strip()) for x in rep.split('\\n')[3].split('     ')[3:6] ]])\n",
 251 |     "    dd = pd.DataFrame()\n",
 252 |     "    dd['Precision'] = det.T[0]\n",
 253 |     "    dd['Recall'] = det.T[1]\n",
 254 |     "    dd['f1_score'] = det.T[2]\n",
 255 |     "    return dd"
 256 |    ]
 257 |   },
 258 |   {
 259 |    "cell_type": "markdown",
 260 |    "metadata": {},
 261 |    "source": [
 262 |     "## Feature Matrix"
 263 |    ]
 264 |   },
 265 |   {
 266 |    "cell_type": "code",
 267 |    "execution_count": 6,
 268 |    "metadata": {},
 269 |    "outputs": [],
 270 |    "source": [
 271 |     "X = np.zeros((len(df['Image_ID']),(256*256)))"
 272 |    ]
 273 |   },
 274 |   {
 275 |    "cell_type": "code",
 276 |    "execution_count": 7,
 277 |    "metadata": {},
 278 |    "outputs": [
 279 |     {
 280 |      "name": "stdout",
 281 |      "output_type": "stream",
 282 |      "text": [
 283 |       "pocessing 0\n",
 284 |       "pocessing 1\n",
 285 |       "pocessing 2\n",
 286 |       "pocessing 3\n",
 287 |       "pocessing 4\n",
 288 |       "pocessing 5\n",
 289 |       "pocessing 6\n",
 290 |       "pocessing 7\n",
 291 |       "pocessing 8\n",
 292 |       "pocessing 9\n",
 293 |       "pocessing 10\n",
 294 |       "pocessing 11\n",
 295 |       "pocessing 12\n",
 296 |       "pocessing 13\n",
 297 |       "pocessing 14\n",
 298 |       "pocessing 15\n",
 299 |       "pocessing 16\n",
 300 |       "pocessing 17\n",
 301 |       "pocessing 18\n",
 302 |       "pocessing 19\n",
 303 |       "pocessing 20\n",
 304 |       "pocessing 21\n",
 305 |       "pocessing 22\n",
 306 |       "pocessing 23\n",
 307 |       "pocessing 24\n",
 308 |       "pocessing 25\n",
 309 |       "pocessing 26\n",
 310 |       "pocessing 27\n",
 311 |       "pocessing 28\n",
 312 |       "pocessing 29\n",
 313 |       "pocessing 30\n",
 314 |       "pocessing 31\n",
 315 |       "pocessing 32\n",
 316 |       "pocessing 33\n",
 317 |       "pocessing 34\n",
 318 |       "pocessing 35\n",
 319 |       "pocessing 36\n",
 320 |       "pocessing 37\n",
 321 |       "pocessing 38\n",
 322 |       "pocessing 39\n",
 323 |       "pocessing 40\n",
 324 |       "pocessing 41\n",
 325 |       "pocessing 42\n",
 326 |       "pocessing 43\n",
 327 |       "pocessing 44\n",
 328 |       "pocessing 45\n",
 329 |       "pocessing 46\n",
 330 |       "pocessing 47\n",
 331 |       "pocessing 48\n",
 332 |       "pocessing 49\n",
 333 |       "pocessing 50\n",
 334 |       "pocessing 51\n",
 335 |       "pocessing 52\n",
 336 |       "pocessing 53\n",
 337 |       "pocessing 54\n",
 338 |       "pocessing 55\n",
 339 |       "pocessing 56\n",
 340 |       "pocessing 57\n",
 341 |       "pocessing 58\n",
 342 |       "pocessing 59\n",
 343 |       "pocessing 60\n",
 344 |       "pocessing 61\n",
 345 |       "pocessing 62\n",
 346 |       "pocessing 63\n",
 347 |       "pocessing 64\n",
 348 |       "pocessing 65\n",
 349 |       "pocessing 66\n",
 350 |       "pocessing 67\n",
 351 |       "pocessing 68\n",
 352 |       "pocessing 69\n",
 353 |       "pocessing 70\n",
 354 |       "pocessing 71\n",
 355 |       "pocessing 72\n",
 356 |       "pocessing 73\n",
 357 |       "pocessing 74\n",
 358 |       "pocessing 75\n",
 359 |       "pocessing 76\n",
 360 |       "pocessing 77\n",
 361 |       "pocessing 78\n",
 362 |       "pocessing 79\n",
 363 |       "pocessing 80\n",
 364 |       "pocessing 81\n",
 365 |       "pocessing 82\n",
 366 |       "pocessing 83\n",
 367 |       "pocessing 84\n",
 368 |       "pocessing 85\n",
 369 |       "pocessing 86\n",
 370 |       "pocessing 87\n",
 371 |       "pocessing 88\n",
 372 |       "pocessing 89\n",
 373 |       "pocessing 90\n",
 374 |       "pocessing 91\n",
 375 |       "pocessing 92\n",
 376 |       "pocessing 93\n",
 377 |       "pocessing 94\n",
 378 |       "pocessing 95\n",
 379 |       "pocessing 96\n",
 380 |       "pocessing 97\n",
 381 |       "pocessing 98\n",
 382 |       "pocessing 99\n",
 383 |       "pocessing 100\n",
 384 |       "pocessing 101\n",
 385 |       "pocessing 102\n",
 386 |       "pocessing 103\n",
 387 |       "pocessing 104\n",
 388 |       "pocessing 105\n",
 389 |       "pocessing 106\n",
 390 |       "pocessing 107\n",
 391 |       "pocessing 108\n",
 392 |       "pocessing 109\n",
 393 |       "pocessing 110\n",
 394 |       "pocessing 111\n",
 395 |       "pocessing 112\n",
 396 |       "pocessing 113\n",
 397 |       "pocessing 114\n",
 398 |       "pocessing 115\n",
 399 |       "pocessing 116\n",
 400 |       "pocessing 117\n",
 401 |       "pocessing 118\n",
 402 |       "pocessing 119\n",
 403 |       "pocessing 120\n",
 404 |       "pocessing 121\n",
 405 |       "pocessing 122\n",
 406 |       "pocessing 123\n",
 407 |       "pocessing 124\n",
 408 |       "pocessing 125\n",
 409 |       "pocessing 126\n",
 410 |       "pocessing 127\n",
 411 |       "pocessing 128\n",
 412 |       "pocessing 129\n",
 413 |       "pocessing 130\n",
 414 |       "pocessing 131\n",
 415 |       "pocessing 132\n",
 416 |       "pocessing 133\n",
 417 |       "pocessing 134\n",
 418 |       "pocessing 135\n",
 419 |       "pocessing 136\n",
 420 |       "pocessing 137\n",
 421 |       "pocessing 138\n",
 422 |       "pocessing 139\n",
 423 |       "pocessing 140\n",
 424 |       "pocessing 141\n",
 425 |       "pocessing 142\n",
 426 |       "pocessing 143\n",
 427 |       "pocessing 144\n",
 428 |       "pocessing 145\n",
 429 |       "pocessing 146\n",
 430 |       "pocessing 147\n",
 431 |       "pocessing 148\n",
 432 |       "pocessing 149\n",
 433 |       "pocessing 150\n",
 434 |       "pocessing 151\n",
 435 |       "pocessing 152\n",
 436 |       "pocessing 153\n",
 437 |       "pocessing 154\n",
 438 |       "pocessing 155\n",
 439 |       "pocessing 156\n",
 440 |       "pocessing 157\n",
 441 |       "pocessing 158\n",
 442 |       "pocessing 159\n",
 443 |       "pocessing 160\n",
 444 |       "pocessing 161\n",
 445 |       "pocessing 162\n",
 446 |       "pocessing 163\n",
 447 |       "pocessing 164\n",
 448 |       "pocessing 165\n",
 449 |       "pocessing 166\n",
 450 |       "pocessing 167\n",
 451 |       "pocessing 168\n",
 452 |       "pocessing 169\n",
 453 |       "pocessing 170\n",
 454 |       "pocessing 171\n",
 455 |       "pocessing 172\n",
 456 |       "pocessing 173\n",
 457 |       "pocessing 174\n",
 458 |       "pocessing 175\n",
 459 |       "pocessing 176\n",
 460 |       "pocessing 177\n",
 461 |       "pocessing 178\n",
 462 |       "pocessing 179\n",
 463 |       "pocessing 180\n",
 464 |       "pocessing 181\n",
 465 |       "pocessing 182\n",
 466 |       "pocessing 183\n",
 467 |       "pocessing 184\n",
 468 |       "pocessing 185\n",
 469 |       "pocessing 186\n",
 470 |       "pocessing 187\n",
 471 |       "pocessing 188\n",
 472 |       "pocessing 189\n",
 473 |       "pocessing 190\n",
 474 |       "pocessing 191\n",
 475 |       "pocessing 192\n",
 476 |       "pocessing 193\n",
 477 |       "pocessing 194\n",
 478 |       "pocessing 195\n",
 479 |       "pocessing 196\n",
 480 |       "pocessing 197\n",
 481 |       "pocessing 198\n",
 482 |       "pocessing 199\n"
 483 |      ]
 484 |     }
 485 |    ],
 486 |    "source": [
 487 |     "for i in range(len(df['Image_ID'])):\n",
 488 |     "    print(\"pocessing\", i)\n",
 489 |     "    X[i] = read_preprocess(df['Image_ID'][i])"
 490 |    ]
 491 |   },
 492 |   {
 493 |    "cell_type": "markdown",
 494 |    "metadata": {},
 495 |    "source": [
 496 |     "## Labels"
 497 |    ]
 498 |   },
 499 |   {
 500 |    "cell_type": "code",
 501 |    "execution_count": 8,
 502 |    "metadata": {},
 503 |    "outputs": [],
 504 |    "source": [
 505 |     "y = np.array(df['Label'])"
 506 |    ]
 507 |   },
 508 |   {
 509 |    "cell_type": "markdown",
 510 |    "metadata": {},
 511 |    "source": [
 512 |     "## Train : Test = 60:40"
 513 |    ]
 514 |   },
 515 |   {
 516 |    "cell_type": "code",
 517 |    "execution_count": 9,
 518 |    "metadata": {},
 519 |    "outputs": [],
 520 |    "source": [
 521 |     "Xtr, Xt, ytr, yt = train_test_split(X,y, test_size = 0.4, train_size = 0.6)"
 522 |    ]
 523 |   },
 524 |   {
 525 |    "cell_type": "markdown",
 526 |    "metadata": {},
 527 |    "source": [
 528 |     "### SVM Classifier"
 529 |    ]
 530 |   },
 531 |   {
 532 |    "cell_type": "code",
 533 |    "execution_count": 10,
 534 |    "metadata": {},
 535 |    "outputs": [],
 536 |    "source": [
 537 |     "clf_SVM = SVC(gamma = 'auto')"
 538 |    ]
 539 |   },
 540 |   {
 541 |    "cell_type": "code",
 542 |    "execution_count": 11,
 543 |    "metadata": {},
 544 |    "outputs": [
 545 |     {
 546 |      "data": {
 547 |       "text/plain": [
 548 |        "SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
 549 |        "  decision_function_shape='ovr', degree=3, gamma='auto', kernel='rbf',\n",
 550 |        "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
 551 |        "  tol=0.001, verbose=False)"
 552 |       ]
 553 |      },
 554 |      "execution_count": 11,
 555 |      "metadata": {},
 556 |      "output_type": "execute_result"
 557 |     }
 558 |    ],
 559 |    "source": [
 560 |     "clf_SVM.fit(Xtr,ytr)"
 561 |    ]
 562 |   },
 563 |   {
 564 |    "cell_type": "code",
 565 |    "execution_count": 12,
 566 |    "metadata": {},
 567 |    "outputs": [],
 568 |    "source": [
 569 |     "yp_SVM = clf_SVM.predict(Xt)"
 570 |    ]
 571 |   },
 572 |   {
 573 |    "cell_type": "code",
 574 |    "execution_count": 13,
 575 |    "metadata": {},
 576 |    "outputs": [
 577 |     {
 578 |      "data": {
 579 |       "text/html": [
 580 |        "<div>\n",
 581 |        "<style scoped>\n",
 582 |        "    .dataframe tbody tr th:only-of-type {\n",
 583 |        "        vertical-align: middle;\n",
 584 |        "    }\n",
 585 |        "\n",
 586 |        "    .dataframe tbody tr th {\n",
 587 |        "        vertical-align: top;\n",
 588 |        "    }\n",
 589 |        "\n",
 590 |        "    .dataframe thead th {\n",
 591 |        "        text-align: right;\n",
 592 |        "    }\n",
 593 |        "</style>\n",
 594 |        "<table border=\"1\" class=\"dataframe\">\n",
 595 |        "  <thead>\n",
 596 |        "    <tr style=\"text-align: right;\">\n",
 597 |        "      <th></th>\n",
 598 |        "      <th>Precision</th>\n",
 599 |        "      <th>Recall</th>\n",
 600 |        "      <th>f1_score</th>\n",
 601 |        "    </tr>\n",
 602 |        "  </thead>\n",
 603 |        "  <tbody>\n",
 604 |        "    <tr>\n",
 605 |        "      <th>0</th>\n",
 606 |        "      <td>0.68</td>\n",
 607 |        "      <td>0.69</td>\n",
 608 |        "      <td>0.68</td>\n",
 609 |        "    </tr>\n",
 610 |        "    <tr>\n",
 611 |        "      <th>1</th>\n",
 612 |        "      <td>0.74</td>\n",
 613 |        "      <td>0.73</td>\n",
 614 |        "      <td>0.74</td>\n",
 615 |        "    </tr>\n",
 616 |        "  </tbody>\n",
 617 |        "</table>\n",
 618 |        "</div>"
 619 |       ],
 620 |       "text/plain": [
 621 |        "   Precision  Recall  f1_score\n",
 622 |        "0       0.68    0.69      0.68\n",
 623 |        "1       0.74    0.73      0.74"
 624 |       ]
 625 |      },
 626 |      "execution_count": 13,
 627 |      "metadata": {},
 628 |      "output_type": "execute_result"
 629 |     }
 630 |    ],
 631 |    "source": [
 632 |     "cr_SVM = CR(yp_SVM,yt)\n",
 633 |     "cr_SVM"
 634 |    ]
 635 |   },
 636 |   {
 637 |    "cell_type": "markdown",
 638 |    "metadata": {},
 639 |    "source": [
 640 |     "### Random Forest"
 641 |    ]
 642 |   },
 643 |   {
 644 |    "cell_type": "code",
 645 |    "execution_count": 14,
 646 |    "metadata": {},
 647 |    "outputs": [],
 648 |    "source": [
 649 |     "clf_RF = RandomForestClassifier(n_estimators=100, max_depth=40,random_state=0)"
 650 |    ]
 651 |   },
 652 |   {
 653 |    "cell_type": "code",
 654 |    "execution_count": 15,
 655 |    "metadata": {},
 656 |    "outputs": [
 657 |     {
 658 |      "data": {
 659 |       "text/plain": [
 660 |        "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
 661 |        "            max_depth=40, max_features='auto', max_leaf_nodes=None,\n",
 662 |        "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
 663 |        "            min_samples_leaf=1, min_samples_split=2,\n",
 664 |        "            min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=None,\n",
 665 |        "            oob_score=False, random_state=0, verbose=0, warm_start=False)"
 666 |       ]
 667 |      },
 668 |      "execution_count": 15,
 669 |      "metadata": {},
 670 |      "output_type": "execute_result"
 671 |     }
 672 |    ],
 673 |    "source": [
 674 |     "clf_RF.fit(Xtr,ytr)"
 675 |    ]
 676 |   },
 677 |   {
 678 |    "cell_type": "code",
 679 |    "execution_count": 16,
 680 |    "metadata": {},
 681 |    "outputs": [],
 682 |    "source": [
 683 |     "yp_RF = clf_RF.predict(Xt)"
 684 |    ]
 685 |   },
 686 |   {
 687 |    "cell_type": "code",
 688 |    "execution_count": 17,
 689 |    "metadata": {},
 690 |    "outputs": [
 691 |     {
 692 |      "data": {
 693 |       "text/html": [
 694 |        "<div>\n",
 695 |        "<style scoped>\n",
 696 |        "    .dataframe tbody tr th:only-of-type {\n",
 697 |        "        vertical-align: middle;\n",
 698 |        "    }\n",
 699 |        "\n",
 700 |        "    .dataframe tbody tr th {\n",
 701 |        "        vertical-align: top;\n",
 702 |        "    }\n",
 703 |        "\n",
 704 |        "    .dataframe thead th {\n",
 705 |        "        text-align: right;\n",
 706 |        "    }\n",
 707 |        "</style>\n",
 708 |        "<table border=\"1\" class=\"dataframe\">\n",
 709 |        "  <thead>\n",
 710 |        "    <tr style=\"text-align: right;\">\n",
 711 |        "      <th></th>\n",
 712 |        "      <th>Precision</th>\n",
 713 |        "      <th>Recall</th>\n",
 714 |        "      <th>f1_score</th>\n",
 715 |        "    </tr>\n",
 716 |        "  </thead>\n",
 717 |        "  <tbody>\n",
 718 |        "    <tr>\n",
 719 |        "      <th>0</th>\n",
 720 |        "      <td>0.83</td>\n",
 721 |        "      <td>0.69</td>\n",
 722 |        "      <td>0.76</td>\n",
 723 |        "    </tr>\n",
 724 |        "    <tr>\n",
 725 |        "      <th>1</th>\n",
 726 |        "      <td>0.78</td>\n",
 727 |        "      <td>0.89</td>\n",
 728 |        "      <td>0.83</td>\n",
 729 |        "    </tr>\n",
 730 |        "  </tbody>\n",
 731 |        "</table>\n",
 732 |        "</div>"
 733 |       ],
 734 |       "text/plain": [
 735 |        "   Precision  Recall  f1_score\n",
 736 |        "0       0.83    0.69      0.76\n",
 737 |        "1       0.78    0.89      0.83"
 738 |       ]
 739 |      },
 740 |      "execution_count": 17,
 741 |      "metadata": {},
 742 |      "output_type": "execute_result"
 743 |     }
 744 |    ],
 745 |    "source": [
 746 |     "cr_RF = CR(yp_RF,yt)\n",
 747 |     "cr_RF"
 748 |    ]
 749 |   },
 750 |   {
 751 |    "cell_type": "markdown",
 752 |    "metadata": {},
 753 |    "source": [
 754 |     "### Decision Tree"
 755 |    ]
 756 |   },
 757 |   {
 758 |    "cell_type": "code",
 759 |    "execution_count": 18,
 760 |    "metadata": {},
 761 |    "outputs": [],
 762 |    "source": [
 763 |     "clf_DT = DecisionTreeClassifier()"
 764 |    ]
 765 |   },
 766 |   {
 767 |    "cell_type": "code",
 768 |    "execution_count": 19,
 769 |    "metadata": {},
 770 |    "outputs": [
 771 |     {
 772 |      "data": {
 773 |       "text/plain": [
 774 |        "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=None,\n",
 775 |        "            max_features=None, max_leaf_nodes=None,\n",
 776 |        "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
 777 |        "            min_samples_leaf=1, min_samples_split=2,\n",
 778 |        "            min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n",
 779 |        "            splitter='best')"
 780 |       ]
 781 |      },
 782 |      "execution_count": 19,
 783 |      "metadata": {},
 784 |      "output_type": "execute_result"
 785 |     }
 786 |    ],
 787 |    "source": [
 788 |     "clf_DT.fit(Xtr,ytr)"
 789 |    ]
 790 |   },
 791 |   {
 792 |    "cell_type": "code",
 793 |    "execution_count": 20,
 794 |    "metadata": {},
 795 |    "outputs": [],
 796 |    "source": [
 797 |     "yp_DT = clf_DT.predict(Xt)"
 798 |    ]
 799 |   },
 800 |   {
 801 |    "cell_type": "code",
 802 |    "execution_count": 21,
 803 |    "metadata": {},
 804 |    "outputs": [
 805 |     {
 806 |      "data": {
 807 |       "text/html": [
 808 |        "<div>\n",
 809 |        "<style scoped>\n",
 810 |        "    .dataframe tbody tr th:only-of-type {\n",
 811 |        "        vertical-align: middle;\n",
 812 |        "    }\n",
 813 |        "\n",
 814 |        "    .dataframe tbody tr th {\n",
 815 |        "        vertical-align: top;\n",
 816 |        "    }\n",
 817 |        "\n",
 818 |        "    .dataframe thead th {\n",
 819 |        "        text-align: right;\n",
 820 |        "    }\n",
 821 |        "</style>\n",
 822 |        "<table border=\"1\" class=\"dataframe\">\n",
 823 |        "  <thead>\n",
 824 |        "    <tr style=\"text-align: right;\">\n",
 825 |        "      <th></th>\n",
 826 |        "      <th>Precision</th>\n",
 827 |        "      <th>Recall</th>\n",
 828 |        "      <th>f1_score</th>\n",
 829 |        "    </tr>\n",
 830 |        "  </thead>\n",
 831 |        "  <tbody>\n",
 832 |        "    <tr>\n",
 833 |        "      <th>0</th>\n",
 834 |        "      <td>0.70</td>\n",
 835 |        "      <td>0.64</td>\n",
 836 |        "      <td>0.67</td>\n",
 837 |        "    </tr>\n",
 838 |        "    <tr>\n",
 839 |        "      <th>1</th>\n",
 840 |        "      <td>0.72</td>\n",
 841 |        "      <td>0.77</td>\n",
 842 |        "      <td>0.75</td>\n",
 843 |        "    </tr>\n",
 844 |        "  </tbody>\n",
 845 |        "</table>\n",
 846 |        "</div>"
 847 |       ],
 848 |       "text/plain": [
 849 |        "   Precision  Recall  f1_score\n",
 850 |        "0       0.70    0.64      0.67\n",
 851 |        "1       0.72    0.77      0.75"
 852 |       ]
 853 |      },
 854 |      "execution_count": 21,
 855 |      "metadata": {},
 856 |      "output_type": "execute_result"
 857 |     }
 858 |    ],
 859 |    "source": [
 860 |     "cr_DT = CR(yp_DT,yt)\n",
 861 |     "cr_DT"
 862 |    ]
 863 |   },
 864 |   {
 865 |    "cell_type": "markdown",
 866 |    "metadata": {},
 867 |    "source": [
 868 |     "### K Nearest Neighbors"
 869 |    ]
 870 |   },
 871 |   {
 872 |    "cell_type": "code",
 873 |    "execution_count": 22,
 874 |    "metadata": {},
 875 |    "outputs": [],
 876 |    "source": [
 877 |     "clf_KNN = KNeighborsClassifier(n_neighbors=3)"
 878 |    ]
 879 |   },
 880 |   {
 881 |    "cell_type": "code",
 882 |    "execution_count": 23,
 883 |    "metadata": {},
 884 |    "outputs": [
 885 |     {
 886 |      "data": {
 887 |       "text/plain": [
 888 |        "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
 889 |        "           metric_params=None, n_jobs=None, n_neighbors=3, p=2,\n",
 890 |        "           weights='uniform')"
 891 |       ]
 892 |      },
 893 |      "execution_count": 23,
 894 |      "metadata": {},
 895 |      "output_type": "execute_result"
 896 |     }
 897 |    ],
 898 |    "source": [
 899 |     "clf_KNN.fit(Xtr,ytr)"
 900 |    ]
 901 |   },
 902 |   {
 903 |    "cell_type": "code",
 904 |    "execution_count": 24,
 905 |    "metadata": {},
 906 |    "outputs": [],
 907 |    "source": [
 908 |     "yp_KNN = clf_KNN.predict(Xt)"
 909 |    ]
 910 |   },
 911 |   {
 912 |    "cell_type": "code",
 913 |    "execution_count": 25,
 914 |    "metadata": {},
 915 |    "outputs": [
 916 |     {
 917 |      "data": {
 918 |       "text/html": [
 919 |        "<div>\n",
 920 |        "<style scoped>\n",
 921 |        "    .dataframe tbody tr th:only-of-type {\n",
 922 |        "        vertical-align: middle;\n",
 923 |        "    }\n",
 924 |        "\n",
 925 |        "    .dataframe tbody tr th {\n",
 926 |        "        vertical-align: top;\n",
 927 |        "    }\n",
 928 |        "\n",
 929 |        "    .dataframe thead th {\n",
 930 |        "        text-align: right;\n",
 931 |        "    }\n",
 932 |        "</style>\n",
 933 |        "<table border=\"1\" class=\"dataframe\">\n",
 934 |        "  <thead>\n",
 935 |        "    <tr style=\"text-align: right;\">\n",
 936 |        "      <th></th>\n",
 937 |        "      <th>Precision</th>\n",
 938 |        "      <th>Recall</th>\n",
 939 |        "      <th>f1_score</th>\n",
 940 |        "    </tr>\n",
 941 |        "  </thead>\n",
 942 |        "  <tbody>\n",
 943 |        "    <tr>\n",
 944 |        "      <th>0</th>\n",
 945 |        "      <td>0.64</td>\n",
 946 |        "      <td>0.83</td>\n",
 947 |        "      <td>0.72</td>\n",
 948 |        "    </tr>\n",
 949 |        "    <tr>\n",
 950 |        "      <th>1</th>\n",
 951 |        "      <td>0.82</td>\n",
 952 |        "      <td>0.61</td>\n",
 953 |        "      <td>0.70</td>\n",
 954 |        "    </tr>\n",
 955 |        "  </tbody>\n",
 956 |        "</table>\n",
 957 |        "</div>"
 958 |       ],
 959 |       "text/plain": [
 960 |        "   Precision  Recall  f1_score\n",
 961 |        "0       0.64    0.83      0.72\n",
 962 |        "1       0.82    0.61      0.70"
 963 |       ]
 964 |      },
 965 |      "execution_count": 25,
 966 |      "metadata": {},
 967 |      "output_type": "execute_result"
 968 |     }
 969 |    ],
 970 |    "source": [
 971 |     "cr_KNN = CR(yp_KNN,yt)\n",
 972 |     "cr_KNN"
 973 |    ]
 974 |   },
 975 |   {
 976 |    "cell_type": "markdown",
 977 |    "metadata": {},
 978 |    "source": [
 979 |     "### Store Model"
 980 |    ]
 981 |   },
 982 |   {
 983 |    "cell_type": "code",
 984 |    "execution_count": 26,
 985 |    "metadata": {},
 986 |    "outputs": [],
 987 |    "source": [
 988 |     "Models = {}"
 989 |    ]
 990 |   },
 991 |   {
 992 |    "cell_type": "code",
 993 |    "execution_count": 27,
 994 |    "metadata": {},
 995 |    "outputs": [],
 996 |    "source": [
 997 |     "Models['60-40'] = [clf_SVM, clf_RF, clf_DT, clf_KNN]"
 998 |    ]
 999 |   },
1000 |   {
1001 |    "cell_type": "markdown",
1002 |    "metadata": {},
1003 |    "source": [
1004 |     "## Train : Test = 70:30"
1005 |    ]
1006 |   },
1007 |   {
1008 |    "cell_type": "code",
1009 |    "execution_count": 28,
1010 |    "metadata": {},
1011 |    "outputs": [],
1012 |    "source": [
1013 |     "Xtr, Xt, ytr, yt = train_test_split(X,y, test_size = 0.3, train_size = 0.7)"
1014 |    ]
1015 |   },
1016 |   {
1017 |    "cell_type": "markdown",
1018 |    "metadata": {},
1019 |    "source": [
1020 |     "### SVM Classifier"
1021 |    ]
1022 |   },
1023 |   {
1024 |    "cell_type": "code",
1025 |    "execution_count": 29,
1026 |    "metadata": {},
1027 |    "outputs": [
1028 |     {
1029 |      "data": {
1030 |       "text/html": [
1031 |        "<div>\n",
1032 |        "<style scoped>\n",
1033 |        "    .dataframe tbody tr th:only-of-type {\n",
1034 |        "        vertical-align: middle;\n",
1035 |        "    }\n",
1036 |        "\n",
1037 |        "    .dataframe tbody tr th {\n",
1038 |        "        vertical-align: top;\n",
1039 |        "    }\n",
1040 |        "\n",
1041 |        "    .dataframe thead th {\n",
1042 |        "        text-align: right;\n",
1043 |        "    }\n",
1044 |        "</style>\n",
1045 |        "<table border=\"1\" class=\"dataframe\">\n",
1046 |        "  <thead>\n",
1047 |        "    <tr style=\"text-align: right;\">\n",
1048 |        "      <th></th>\n",
1049 |        "      <th>Precision</th>\n",
1050 |        "      <th>Recall</th>\n",
1051 |        "      <th>f1_score</th>\n",
1052 |        "    </tr>\n",
1053 |        "  </thead>\n",
1054 |        "  <tbody>\n",
1055 |        "    <tr>\n",
1056 |        "      <th>0</th>\n",
1057 |        "      <td>0.70</td>\n",
1058 |        "      <td>0.81</td>\n",
1059 |        "      <td>0.75</td>\n",
1060 |        "    </tr>\n",
1061 |        "    <tr>\n",
1062 |        "      <th>1</th>\n",
1063 |        "      <td>0.83</td>\n",
1064 |        "      <td>0.74</td>\n",
1065 |        "      <td>0.78</td>\n",
1066 |        "    </tr>\n",
1067 |        "  </tbody>\n",
1068 |        "</table>\n",
1069 |        "</div>"
1070 |       ],
1071 |       "text/plain": [
1072 |        "   Precision  Recall  f1_score\n",
1073 |        "0       0.70    0.81      0.75\n",
1074 |        "1       0.83    0.74      0.78"
1075 |       ]
1076 |      },
1077 |      "execution_count": 29,
1078 |      "metadata": {},
1079 |      "output_type": "execute_result"
1080 |     }
1081 |    ],
1082 |    "source": [
1083 |     "clf_SVM = SVC(gamma = 'auto')\n",
1084 |     "clf_SVM.fit(Xtr,ytr)\n",
1085 |     "yp_SVM = clf_SVM.predict(Xt)\n",
1086 |     "cr_SVM = CR(yp_SVM,yt)\n",
1087 |     "cr_SVM"
1088 |    ]
1089 |   },
1090 |   {
1091 |    "cell_type": "markdown",
1092 |    "metadata": {},
1093 |    "source": [
1094 |     "### Random Forest"
1095 |    ]
1096 |   },
1097 |   {
1098 |    "cell_type": "code",
1099 |    "execution_count": 30,
1100 |    "metadata": {},
1101 |    "outputs": [
1102 |     {
1103 |      "name": "stderr",
1104 |      "output_type": "stream",
1105 |      "text": [
1106 |       "C:\\Users\\Sumit\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\forest.py:248: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
1107 |       "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
1108 |      ]
1109 |     },
1110 |     {
1111 |      "data": {
1112 |       "text/html": [
1113 |        "<div>\n",
1114 |        "<style scoped>\n",
1115 |        "    .dataframe tbody tr th:only-of-type {\n",
1116 |        "        vertical-align: middle;\n",
1117 |        "    }\n",
1118 |        "\n",
1119 |        "    .dataframe tbody tr th {\n",
1120 |        "        vertical-align: top;\n",
1121 |        "    }\n",
1122 |        "\n",
1123 |        "    .dataframe thead th {\n",
1124 |        "        text-align: right;\n",
1125 |        "    }\n",
1126 |        "</style>\n",
1127 |        "<table border=\"1\" class=\"dataframe\">\n",
1128 |        "  <thead>\n",
1129 |        "    <tr style=\"text-align: right;\">\n",
1130 |        "      <th></th>\n",
1131 |        "      <th>Precision</th>\n",
1132 |        "      <th>Recall</th>\n",
1133 |        "      <th>f1_score</th>\n",
1134 |        "    </tr>\n",
1135 |        "  </thead>\n",
1136 |        "  <tbody>\n",
1137 |        "    <tr>\n",
1138 |        "      <th>0</th>\n",
1139 |        "      <td>0.71</td>\n",
1140 |        "      <td>0.92</td>\n",
1141 |        "      <td>0.8</td>\n",
1142 |        "    </tr>\n",
1143 |        "    <tr>\n",
1144 |        "      <th>1</th>\n",
1145 |        "      <td>0.92</td>\n",
1146 |        "      <td>0.71</td>\n",
1147 |        "      <td>0.8</td>\n",
1148 |        "    </tr>\n",
1149 |        "  </tbody>\n",
1150 |        "</table>\n",
1151 |        "</div>"
1152 |       ],
1153 |       "text/plain": [
1154 |        "   Precision  Recall  f1_score\n",
1155 |        "0       0.71    0.92       0.8\n",
1156 |        "1       0.92    0.71       0.8"
1157 |       ]
1158 |      },
1159 |      "execution_count": 30,
1160 |      "metadata": {},
1161 |      "output_type": "execute_result"
1162 |     }
1163 |    ],
1164 |    "source": [
1165 |     "clf_RF = RandomForestClassifier()\n",
1166 |     "clf_RF.fit(Xtr,ytr)\n",
1167 |     "yp_RF = clf_RF.predict(Xt)\n",
1168 |     "cr_RF = CR(yp_RF,yt)\n",
1169 |     "cr_RF"
1170 |    ]
1171 |   },
1172 |   {
1173 |    "cell_type": "markdown",
1174 |    "metadata": {},
1175 |    "source": [
1176 |     "### Decision Tree"
1177 |    ]
1178 |   },
1179 |   {
1180 |    "cell_type": "code",
1181 |    "execution_count": 31,
1182 |    "metadata": {},
1183 |    "outputs": [
1184 |     {
1185 |      "data": {
1186 |       "text/html": [
1187 |        "<div>\n",
1188 |        "<style scoped>\n",
1189 |        "    .dataframe tbody tr th:only-of-type {\n",
1190 |        "        vertical-align: middle;\n",
1191 |        "    }\n",
1192 |        "\n",
1193 |        "    .dataframe tbody tr th {\n",
1194 |        "        vertical-align: top;\n",
1195 |        "    }\n",
1196 |        "\n",
1197 |        "    .dataframe thead th {\n",
1198 |        "        text-align: right;\n",
1199 |        "    }\n",
1200 |        "</style>\n",
1201 |        "<table border=\"1\" class=\"dataframe\">\n",
1202 |        "  <thead>\n",
1203 |        "    <tr style=\"text-align: right;\">\n",
1204 |        "      <th></th>\n",
1205 |        "      <th>Precision</th>\n",
1206 |        "      <th>Recall</th>\n",
1207 |        "      <th>f1_score</th>\n",
1208 |        "    </tr>\n",
1209 |        "  </thead>\n",
1210 |        "  <tbody>\n",
1211 |        "    <tr>\n",
1212 |        "      <th>0</th>\n",
1213 |        "      <td>0.53</td>\n",
1214 |        "      <td>0.62</td>\n",
1215 |        "      <td>0.57</td>\n",
1216 |        "    </tr>\n",
1217 |        "    <tr>\n",
1218 |        "      <th>1</th>\n",
1219 |        "      <td>0.67</td>\n",
1220 |        "      <td>0.59</td>\n",
1221 |        "      <td>0.62</td>\n",
1222 |        "    </tr>\n",
1223 |        "  </tbody>\n",
1224 |        "</table>\n",
1225 |        "</div>"
1226 |       ],
1227 |       "text/plain": [
1228 |        "   Precision  Recall  f1_score\n",
1229 |        "0       0.53    0.62      0.57\n",
1230 |        "1       0.67    0.59      0.62"
1231 |       ]
1232 |      },
1233 |      "execution_count": 31,
1234 |      "metadata": {},
1235 |      "output_type": "execute_result"
1236 |     }
1237 |    ],
1238 |    "source": [
1239 |     "clf_DT = DecisionTreeClassifier()\n",
1240 |     "clf_DT.fit(Xtr,ytr)\n",
1241 |     "yp_DT = clf_DT.predict(Xt)\n",
1242 |     "cr_DT = CR(yp_DT,yt)\n",
1243 |     "cr_DT"
1244 |    ]
1245 |   },
1246 |   {
1247 |    "cell_type": "markdown",
1248 |    "metadata": {},
1249 |    "source": [
1250 |     "### K Nearest Neighbors"
1251 |    ]
1252 |   },
1253 |   {
1254 |    "cell_type": "code",
1255 |    "execution_count": 32,
1256 |    "metadata": {},
1257 |    "outputs": [
1258 |     {
1259 |      "data": {
1260 |       "text/html": [
1261 |        "<div>\n",
1262 |        "<style scoped>\n",
1263 |        "    .dataframe tbody tr th:only-of-type {\n",
1264 |        "        vertical-align: middle;\n",
1265 |        "    }\n",
1266 |        "\n",
1267 |        "    .dataframe tbody tr th {\n",
1268 |        "        vertical-align: top;\n",
1269 |        "    }\n",
1270 |        "\n",
1271 |        "    .dataframe thead th {\n",
1272 |        "        text-align: right;\n",
1273 |        "    }\n",
1274 |        "</style>\n",
1275 |        "<table border=\"1\" class=\"dataframe\">\n",
1276 |        "  <thead>\n",
1277 |        "    <tr style=\"text-align: right;\">\n",
1278 |        "      <th></th>\n",
1279 |        "      <th>Precision</th>\n",
1280 |        "      <th>Recall</th>\n",
1281 |        "      <th>f1_score</th>\n",
1282 |        "    </tr>\n",
1283 |        "  </thead>\n",
1284 |        "  <tbody>\n",
1285 |        "    <tr>\n",
1286 |        "      <th>0</th>\n",
1287 |        "      <td>0.68</td>\n",
1288 |        "      <td>0.88</td>\n",
1289 |        "      <td>0.77</td>\n",
1290 |        "    </tr>\n",
1291 |        "    <tr>\n",
1292 |        "      <th>1</th>\n",
1293 |        "      <td>0.88</td>\n",
1294 |        "      <td>0.68</td>\n",
1295 |        "      <td>0.77</td>\n",
1296 |        "    </tr>\n",
1297 |        "  </tbody>\n",
1298 |        "</table>\n",
1299 |        "</div>"
1300 |       ],
1301 |       "text/plain": [
1302 |        "   Precision  Recall  f1_score\n",
1303 |        "0       0.68    0.88      0.77\n",
1304 |        "1       0.88    0.68      0.77"
1305 |       ]
1306 |      },
1307 |      "execution_count": 32,
1308 |      "metadata": {},
1309 |      "output_type": "execute_result"
1310 |     }
1311 |    ],
1312 |    "source": [
1313 |     "clf_KNN = KNeighborsClassifier(n_neighbors=3)\n",
1314 |     "clf_KNN.fit(Xtr,ytr)\n",
1315 |     "yp_KNN = clf_KNN.predict(Xt)\n",
1316 |     "cr_KNN = CR(yp_KNN,yt)\n",
1317 |     "cr_KNN"
1318 |    ]
1319 |   },
1320 |   {
1321 |    "cell_type": "markdown",
1322 |    "metadata": {},
1323 |    "source": [
1324 |     "### Store Model"
1325 |    ]
1326 |   },
1327 |   {
1328 |    "cell_type": "code",
1329 |    "execution_count": 33,
1330 |    "metadata": {},
1331 |    "outputs": [],
1332 |    "source": [
1333 |     "Models['70-30'] = [clf_SVM, clf_RF, clf_DT, clf_KNN]"
1334 |    ]
1335 |   },
1336 |   {
1337 |    "cell_type": "markdown",
1338 |    "metadata": {},
1339 |    "source": [
1340 |     "## Train : Test = 80:20"
1341 |    ]
1342 |   },
1343 |   {
1344 |    "cell_type": "code",
1345 |    "execution_count": 34,
1346 |    "metadata": {},
1347 |    "outputs": [],
1348 |    "source": [
1349 |     "Xtr, Xt, ytr, yt = train_test_split(X,y, test_size = 0.2, train_size = 0.8)"
1350 |    ]
1351 |   },
1352 |   {
1353 |    "cell_type": "markdown",
1354 |    "metadata": {},
1355 |    "source": [
1356 |     "### SVM Classifier"
1357 |    ]
1358 |   },
1359 |   {
1360 |    "cell_type": "code",
1361 |    "execution_count": 35,
1362 |    "metadata": {},
1363 |    "outputs": [
1364 |     {
1365 |      "data": {
1366 |       "text/html": [
1367 |        "<div>\n",
1368 |        "<style scoped>\n",
1369 |        "    .dataframe tbody tr th:only-of-type {\n",
1370 |        "        vertical-align: middle;\n",
1371 |        "    }\n",
1372 |        "\n",
1373 |        "    .dataframe tbody tr th {\n",
1374 |        "        vertical-align: top;\n",
1375 |        "    }\n",
1376 |        "\n",
1377 |        "    .dataframe thead th {\n",
1378 |        "        text-align: right;\n",
1379 |        "    }\n",
1380 |        "</style>\n",
1381 |        "<table border=\"1\" class=\"dataframe\">\n",
1382 |        "  <thead>\n",
1383 |        "    <tr style=\"text-align: right;\">\n",
1384 |        "      <th></th>\n",
1385 |        "      <th>Precision</th>\n",
1386 |        "      <th>Recall</th>\n",
1387 |        "      <th>f1_score</th>\n",
1388 |        "    </tr>\n",
1389 |        "  </thead>\n",
1390 |        "  <tbody>\n",
1391 |        "    <tr>\n",
1392 |        "      <th>0</th>\n",
1393 |        "      <td>0.81</td>\n",
1394 |        "      <td>0.85</td>\n",
1395 |        "      <td>0.83</td>\n",
1396 |        "    </tr>\n",
1397 |        "    <tr>\n",
1398 |        "      <th>1</th>\n",
1399 |        "      <td>0.84</td>\n",
1400 |        "      <td>0.80</td>\n",
1401 |        "      <td>0.82</td>\n",
1402 |        "    </tr>\n",
1403 |        "  </tbody>\n",
1404 |        "</table>\n",
1405 |        "</div>"
1406 |       ],
1407 |       "text/plain": [
1408 |        "   Precision  Recall  f1_score\n",
1409 |        "0       0.81    0.85      0.83\n",
1410 |        "1       0.84    0.80      0.82"
1411 |       ]
1412 |      },
1413 |      "execution_count": 35,
1414 |      "metadata": {},
1415 |      "output_type": "execute_result"
1416 |     }
1417 |    ],
1418 |    "source": [
1419 |     "clf_SVM = SVC(gamma = 'auto')\n",
1420 |     "clf_SVM.fit(Xtr,ytr)\n",
1421 |     "yp_SVM = clf_SVM.predict(Xt)\n",
1422 |     "cr_SVM = CR(yp_SVM,yt)\n",
1423 |     "cr_SVM"
1424 |    ]
1425 |   },
1426 |   {
1427 |    "cell_type": "markdown",
1428 |    "metadata": {},
1429 |    "source": [
1430 |     "### Random Forest"
1431 |    ]
1432 |   },
1433 |   {
1434 |    "cell_type": "code",
1435 |    "execution_count": 36,
1436 |    "metadata": {},
1437 |    "outputs": [
1438 |     {
1439 |      "data": {
1440 |       "text/html": [
1441 |        "<div>\n",
1442 |        "<style scoped>\n",
1443 |        "    .dataframe tbody tr th:only-of-type {\n",
1444 |        "        vertical-align: middle;\n",
1445 |        "    }\n",
1446 |        "\n",
1447 |        "    .dataframe tbody tr th {\n",
1448 |        "        vertical-align: top;\n",
1449 |        "    }\n",
1450 |        "\n",
1451 |        "    .dataframe thead th {\n",
1452 |        "        text-align: right;\n",
1453 |        "    }\n",
1454 |        "</style>\n",
1455 |        "<table border=\"1\" class=\"dataframe\">\n",
1456 |        "  <thead>\n",
1457 |        "    <tr style=\"text-align: right;\">\n",
1458 |        "      <th></th>\n",
1459 |        "      <th>Precision</th>\n",
1460 |        "      <th>Recall</th>\n",
1461 |        "      <th>f1_score</th>\n",
1462 |        "    </tr>\n",
1463 |        "  </thead>\n",
1464 |        "  <tbody>\n",
1465 |        "    <tr>\n",
1466 |        "      <th>0</th>\n",
1467 |        "      <td>0.84</td>\n",
1468 |        "      <td>0.80</td>\n",
1469 |        "      <td>0.82</td>\n",
1470 |        "    </tr>\n",
1471 |        "    <tr>\n",
1472 |        "      <th>1</th>\n",
1473 |        "      <td>0.81</td>\n",
1474 |        "      <td>0.85</td>\n",
1475 |        "      <td>0.83</td>\n",
1476 |        "    </tr>\n",
1477 |        "  </tbody>\n",
1478 |        "</table>\n",
1479 |        "</div>"
1480 |       ],
1481 |       "text/plain": [
1482 |        "   Precision  Recall  f1_score\n",
1483 |        "0       0.84    0.80      0.82\n",
1484 |        "1       0.81    0.85      0.83"
1485 |       ]
1486 |      },
1487 |      "execution_count": 36,
1488 |      "metadata": {},
1489 |      "output_type": "execute_result"
1490 |     }
1491 |    ],
1492 |    "source": [
1493 |     "clf_RF = RandomForestClassifier(n_estimators=100, max_depth=40,random_state=0)\n",
1494 |     "clf_RF.fit(Xtr,ytr)\n",
1495 |     "yp_RF = clf_RF.predict(Xt)\n",
1496 |     "cr_RF = CR(yp_RF,yt)\n",
1497 |     "cr_RF"
1498 |    ]
1499 |   },
1500 |   {
1501 |    "cell_type": "markdown",
1502 |    "metadata": {},
1503 |    "source": [
1504 |     "### Decision Tree"
1505 |    ]
1506 |   },
1507 |   {
1508 |    "cell_type": "code",
1509 |    "execution_count": 37,
1510 |    "metadata": {},
1511 |    "outputs": [
1512 |     {
1513 |      "data": {
1514 |       "text/html": [
1515 |        "<div>\n",
1516 |        "<style scoped>\n",
1517 |        "    .dataframe tbody tr th:only-of-type {\n",
1518 |        "        vertical-align: middle;\n",
1519 |        "    }\n",
1520 |        "\n",
1521 |        "    .dataframe tbody tr th {\n",
1522 |        "        vertical-align: top;\n",
1523 |        "    }\n",
1524 |        "\n",
1525 |        "    .dataframe thead th {\n",
1526 |        "        text-align: right;\n",
1527 |        "    }\n",
1528 |        "</style>\n",
1529 |        "<table border=\"1\" class=\"dataframe\">\n",
1530 |        "  <thead>\n",
1531 |        "    <tr style=\"text-align: right;\">\n",
1532 |        "      <th></th>\n",
1533 |        "      <th>Precision</th>\n",
1534 |        "      <th>Recall</th>\n",
1535 |        "      <th>f1_score</th>\n",
1536 |        "    </tr>\n",
1537 |        "  </thead>\n",
1538 |        "  <tbody>\n",
1539 |        "    <tr>\n",
1540 |        "      <th>0</th>\n",
1541 |        "      <td>0.68</td>\n",
1542 |        "      <td>0.75</td>\n",
1543 |        "      <td>0.71</td>\n",
1544 |        "    </tr>\n",
1545 |        "    <tr>\n",
1546 |        "      <th>1</th>\n",
1547 |        "      <td>0.72</td>\n",
1548 |        "      <td>0.65</td>\n",
1549 |        "      <td>0.68</td>\n",
1550 |        "    </tr>\n",
1551 |        "  </tbody>\n",
1552 |        "</table>\n",
1553 |        "</div>"
1554 |       ],
1555 |       "text/plain": [
1556 |        "   Precision  Recall  f1_score\n",
1557 |        "0       0.68    0.75      0.71\n",
1558 |        "1       0.72    0.65      0.68"
1559 |       ]
1560 |      },
1561 |      "execution_count": 37,
1562 |      "metadata": {},
1563 |      "output_type": "execute_result"
1564 |     }
1565 |    ],
1566 |    "source": [
1567 |     "clf_DT = DecisionTreeClassifier()\n",
1568 |     "clf_DT.fit(Xtr,ytr)\n",
1569 |     "yp_DT = clf_DT.predict(Xt)\n",
1570 |     "cr_DT = CR(yp_DT,yt)\n",
1571 |     "cr_DT"
1572 |    ]
1573 |   },
1574 |   {
1575 |    "cell_type": "markdown",
1576 |    "metadata": {},
1577 |    "source": [
1578 |     "### K Nearest Neighbors"
1579 |    ]
1580 |   },
1581 |   {
1582 |    "cell_type": "code",
1583 |    "execution_count": 38,
1584 |    "metadata": {},
1585 |    "outputs": [
1586 |     {
1587 |      "data": {
1588 |       "text/html": [
1589 |        "<div>\n",
1590 |        "<style scoped>\n",
1591 |        "    .dataframe tbody tr th:only-of-type {\n",
1592 |        "        vertical-align: middle;\n",
1593 |        "    }\n",
1594 |        "\n",
1595 |        "    .dataframe tbody tr th {\n",
1596 |        "        vertical-align: top;\n",
1597 |        "    }\n",
1598 |        "\n",
1599 |        "    .dataframe thead th {\n",
1600 |        "        text-align: right;\n",
1601 |        "    }\n",
1602 |        "</style>\n",
1603 |        "<table border=\"1\" class=\"dataframe\">\n",
1604 |        "  <thead>\n",
1605 |        "    <tr style=\"text-align: right;\">\n",
1606 |        "      <th></th>\n",
1607 |        "      <th>Precision</th>\n",
1608 |        "      <th>Recall</th>\n",
1609 |        "      <th>f1_score</th>\n",
1610 |        "    </tr>\n",
1611 |        "  </thead>\n",
1612 |        "  <tbody>\n",
1613 |        "    <tr>\n",
1614 |        "      <th>0</th>\n",
1615 |        "      <td>0.70</td>\n",
1616 |        "      <td>0.95</td>\n",
1617 |        "      <td>0.81</td>\n",
1618 |        "    </tr>\n",
1619 |        "    <tr>\n",
1620 |        "      <th>1</th>\n",
1621 |        "      <td>0.92</td>\n",
1622 |        "      <td>0.60</td>\n",
1623 |        "      <td>0.73</td>\n",
1624 |        "    </tr>\n",
1625 |        "  </tbody>\n",
1626 |        "</table>\n",
1627 |        "</div>"
1628 |       ],
1629 |       "text/plain": [
1630 |        "   Precision  Recall  f1_score\n",
1631 |        "0       0.70    0.95      0.81\n",
1632 |        "1       0.92    0.60      0.73"
1633 |       ]
1634 |      },
1635 |      "execution_count": 38,
1636 |      "metadata": {},
1637 |      "output_type": "execute_result"
1638 |     }
1639 |    ],
1640 |    "source": [
1641 |     "clf_KNN = KNeighborsClassifier(n_neighbors=3)\n",
1642 |     "clf_KNN.fit(Xtr,ytr)\n",
1643 |     "yp_KNN = clf_KNN.predict(Xt)\n",
1644 |     "cr_KNN = CR(yp_KNN,yt)\n",
1645 |     "cr_KNN"
1646 |    ]
1647 |   },
1648 |   {
1649 |    "cell_type": "markdown",
1650 |    "metadata": {},
1651 |    "source": [
1652 |     "### Store Model"
1653 |    ]
1654 |   },
1655 |   {
1656 |    "cell_type": "code",
1657 |    "execution_count": 39,
1658 |    "metadata": {},
1659 |    "outputs": [],
1660 |    "source": [
1661 |     "Models['80-20'] = [clf_SVM, clf_RF, clf_DT, clf_KNN]"
1662 |    ]
1663 |   },
1664 |   {
1665 |    "cell_type": "markdown",
1666 |    "metadata": {},
1667 |    "source": [
1668 |     "## Function to predict for a single Image"
1669 |    ]
1670 |   },
1671 |   {
1672 |    "cell_type": "code",
1673 |    "execution_count": 40,
1674 |    "metadata": {},
1675 |    "outputs": [],
1676 |    "source": [
1677 |     "def read_preprocess_path(path):\n",
1678 |     "    \n",
1679 |     "    frame = cv2.imread(path)\n",
1680 |     "    imG = cv2.equalizeHist(frame[:,:,1])\n",
1681 |     "    blG = gaussian_filter(np.max(imG) - imG, sigma=50)  \n",
1682 |     "    fG = ( blG < np.mean(blG) )* (np.max(imG) - imG)\n",
1683 |     "    pifG = (fG/np.mean(fG))**2\n",
1684 |     "    pifG = (pifG * (255/np.max(pifG))).astype(int)\n",
1685 |     "    frame[:,:,1] = pifG\n",
1686 |     "    fr = cv2.resize(frame,(256,256), interpolation = cv2.INTER_AREA)\n",
1687 |     "    fg = fr[:,:,1]\n",
1688 |     "    fg1 = np.max(fg**2) - fg**2\n",
1689 |     "    fg2 = (fg1>245) * fg1\n",
1690 |     "    fg3 = (fg2>=248) * 255\n",
1691 |     "    \n",
1692 |     "    return (fg3.reshape(256*256))/255  "
1693 |    ]
1694 |   },
1695 |   {
1696 |    "cell_type": "code",
1697 |    "execution_count": 41,
1698 |    "metadata": {},
1699 |    "outputs": [],
1700 |    "source": [
1701 |     "def Predict(path,ratio,model_index):\n",
1702 |     "    X1 = read_preprocess_path(path)\n",
1703 |     "    X1 = X1.reshape((1,256*256))\n",
1704 |     "    pred = Models[str(ratio)][model_index].predict(X1)\n",
1705 |     "    return pred"
1706 |    ]
1707 |   },
1708 |   {
1709 |    "cell_type": "code",
1710 |    "execution_count": 42,
1711 |    "metadata": {},
1712 |    "outputs": [
1713 |     {
1714 |      "data": {
1715 |       "text/plain": [
1716 |        "array([1], dtype=int64)"
1717 |       ]
1718 |      },
1719 |      "execution_count": 42,
1720 |      "metadata": {},
1721 |      "output_type": "execute_result"
1722 |     }
1723 |    ],
1724 |    "source": [
1725 |     "Predict('dataset/trainimages/0ceb222f6629.png','70-30',3)"
1726 |    ]
1727 |   },
1728 |   {
1729 |    "cell_type": "markdown",
1730 |    "metadata": {},
1731 |    "source": [
1732 |     "## Confusion Matrix"
1733 |    ]
1734 |   },
1735 |   {
1736 |    "cell_type": "code",
1737 |    "execution_count": 43,
1738 |    "metadata": {},
1739 |    "outputs": [
1740 |     {
1741 |      "data": {
1742 |       "text/plain": [
1743 |        "<matplotlib.axes._subplots.AxesSubplot at 0x29705744400>"
1744 |       ]
1745 |      },
1746 |      "execution_count": 43,
1747 |      "metadata": {},
1748 |      "output_type": "execute_result"
1749 |     },
1750 |     {
1751 |      "data": {
1752 |       "image/png": "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\n",
1753 |       "text/plain": [
1754 |        "<Figure size 648x576 with 2 Axes>"
1755 |       ]
1756 |      },
1757 |      "metadata": {
1758 |       "needs_background": "light"
1759 |      },
1760 |      "output_type": "display_data"
1761 |     }
1762 |    ],
1763 |    "source": [
1764 |     "cm = ConfusionMatrix(Models['70-30'][1].predict(Xt),yt)\n",
1765 |     "cm.plot() "
1766 |    ]
1767 |   },
1768 |   {
1769 |    "cell_type": "code",
1770 |    "execution_count": 44,
1771 |    "metadata": {},
1772 |    "outputs": [
1773 |     {
1774 |      "data": {
1775 |       "text/plain": [
1776 |        "Predicted  False  True  __all__\n",
1777 |        "Actual                         \n",
1778 |        "False         20     2       22\n",
1779 |        "True           0    18       18\n",
1780 |        "__all__       20    20       40"
1781 |       ]
1782 |      },
1783 |      "execution_count": 44,
1784 |      "metadata": {},
1785 |      "output_type": "execute_result"
1786 |     }
1787 |    ],
1788 |    "source": [
1789 |     "cm"
1790 |    ]
1791 |   },
1792 |   {
1793 |    "cell_type": "code",
1794 |    "execution_count": null,
1795 |    "metadata": {},
1796 |    "outputs": [],
1797 |    "source": []
1798 |   }
1799 |  ],
1800 |  "metadata": {
1801 |   "kernelspec": {
1802 |    "display_name": "Python 3",
1803 |    "language": "python",
1804 |    "name": "python3"
1805 |   },
1806 |   "language_info": {
1807 |    "codemirror_mode": {
1808 |     "name": "ipython",
1809 |     "version": 3
1810 |    },
1811 |    "file_extension": ".py",
1812 |    "mimetype": "text/x-python",
1813 |    "name": "python",
1814 |    "nbconvert_exporter": "python",
1815 |    "pygments_lexer": "ipython3",
1816 |    "version": "3.7.0"
1817 |   }
1818 |  },
1819 |  "nbformat": 4,
1820 |  "nbformat_minor": 2
1821 | }
1822 | 


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