vgg_tools.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 基本信息\n",
    "在训练集或者验证集中标注了目标信息，可以计算\n",
    "\n",
    "1、目标的类别\n",
    "\n",
    "2、目标的尺寸位置(y1,x1,y2,x2)\n",
    "\n",
    "3、目标的面积\n",
    "\n",
    "4、宽高比\n",
    "\n",
    "5、目标的多边形面积\n",
    "\n",
    "6、面积比"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:528: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:529: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:530: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:535: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n",
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 58.  59.  60.  61.  62.  63.  64.  65.  67.  68.  70.  73.  74.  75.\n",
      "  76.  78.  80.  82.  83.  84.  85.  86.  87.  88.  89.  92.  95.  96.\n",
      "  97.  98.  99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110.\n",
      " 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 125. 127.\n",
      " 128. 129. 130. 132. 134. 135. 136. 137. 138. 139. 142. 170. 171. 172.\n",
      " 173. 174. 175. 176. 177. 178. 179. 180. 182. 266. 267. 268. 269. 270.\n",
      " 271. 272. 273. 274. 275. 276. 277. 278. 279. 280. 281. 282. 283. 284.\n",
      " 285. 286. 349. 350. 351. 352. 353. 354. 355. 356. 358. 363. 365. 367.\n",
      " 368. 369. 370. 371. 372. 373. 375. 376. 377. 378. 379. 380. 381. 382.\n",
      " 383. 384. 387. 402. 404. 405. 406. 407. 408. 409. 410.]\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00028.jpeg\n",
      "7 [ 321  803  740 1022] 91761 0.522673031026253 62338 0.679351794335284\n",
      "7 [329   1 757 186] 79180 0.4322429906542056 56305 0.7111012882040919\n",
      "2 [321 702 465 760] 8352 0.4027777777777778 4294 0.5141283524904214\n",
      "2 [390 812 600 902] 18900 0.42857142857142855 8716 0.46116402116402117\n",
      "2 [326 226 465 274] 6672 0.34532374100719426 3120 0.4676258992805755\n",
      "2 [398  90 592 186] 18624 0.4948453608247423 7705 0.41371348797250856\n",
      "3 [318 181 674 320] 49484 0.3904494382022472 5989 0.12102901948104437\n",
      "3 [269 332 498 358] 5954 0.11353711790393013 2280 0.3829358414511253\n",
      "3 [241 387 535 425] 11172 0.1292517006802721 3385 0.30298961689939136\n",
      "3 [237 562 455 590] 6104 0.12844036697247707 2791 0.45724115334207077\n",
      "3 [240 480 619 502] 8338 0.05804749340369393 5445 0.6530343007915568\n",
      "3 [307 670 668 798] 46208 0.3545706371191136 5563 0.12039040858725762\n",
      "4 [266   5 543 308] 83931 1.0938628158844765 12823 0.15278025997545602\n",
      "4 [241 386 283 444] 2436 1.380952380952381 1230 0.5049261083743842\n",
      "4 [236 529 275 591] 2418 1.5897435897435896 1166 0.48221670802315963\n",
      "4 [ 256  670  550 1023] 103782 1.2006802721088434 13409 0.12920352276888092\n",
      "5 [265 321 362 366] 4365 0.4639175257731959 3364 0.7706758304696449\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00029.jpeg\n",
      "7 [ 322  799  738 1023] 93184 0.5384615384615384 63565 0.6821450034340659\n",
      "7 [326   1 761 186] 80475 0.42528735632183906 57073 0.709201615408512\n",
      "3 [310 671 665 798] 45085 0.35774647887323946 5994 0.13294887434845293\n",
      "3 [241 476 613 504] 10416 0.07526881720430108 6936 0.6658986175115207\n",
      "3 [240 386 534 427] 12054 0.13945578231292516 4029 0.33424589347934297\n",
      "3 [270 332 498 359] 6156 0.11842105263157894 2780 0.45159194282001297\n",
      "3 [317 181 674 320] 49623 0.38935574229691877 6267 0.12629224351611149\n",
      "4 [255   2 542 311] 88683 1.0766550522648084 14036 0.15827159658559137\n",
      "4 [241 384 282 445] 2501 1.4878048780487805 1223 0.48900439824070374\n",
      "4 [237 530 274 590] 2220 1.6216216216216217 990 0.44594594594594594\n",
      "4 [ 254  669  552 1022] 105194 1.1845637583892616 12974 0.12333403045801092\n",
      "5 [264 318 362 366] 4704 0.4897959183673469 3584 0.7619047619047619\n",
      "2 [390 811 602 903] 19504 0.4339622641509434 8748 0.4485233798195242\n",
      "2 [321 699 466 762] 9135 0.43448275862068964 4766 0.5217296113847838\n",
      "2 [324 227 462 277] 6900 0.36231884057971014 3641 0.5276811594202898\n",
      "2 [394  88 593 189] 20099 0.507537688442211 8484 0.4221105527638191\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00030.jpeg\n",
      "7 [ 319  801  738 1022] 92599 0.5274463007159904 63430 0.6849965982354022\n",
      "7 [328   2 756 186] 78752 0.42990654205607476 55900 0.7098232425843153\n",
      "2 [323 699 466 758] 8437 0.4125874125874126 4573 0.5420173047291691\n",
      "2 [389 811 601 904] 19716 0.4386792452830189 8850 0.44887401095556906\n",
      "2 [325 230 465 276] 6440 0.32857142857142857 3385 0.5256211180124224\n",
      "2 [398  87 592 186] 19206 0.5103092783505154 7915 0.41211079870873685\n",
      "3 [318 182 675 322] 49980 0.39215686274509803 6077 0.12158863545418168\n",
      "3 [270 332 500 357] 5750 0.10869565217391304 2312 0.40208695652173915\n",
      "3 [243 387 537 427] 11760 0.1360544217687075 3967 0.3373299319727891\n",
      "3 [241 479 617 503] 9024 0.06382978723404255 5828 0.6458333333333334\n",
      "3 [238 557 497 591] 8806 0.13127413127413126 4338 0.49261866908925733\n",
      "3 [308 668 667 799] 47029 0.3649025069637883 6803 0.14465542537583193\n",
      "4 [265   2 542 309] 85039 1.108303249097473 12612 0.1483084231940639\n",
      "4 [241 386 282 446] 2460 1.4634146341463414 1122 0.4560975609756098\n",
      "4 [238 529 275 590] 2257 1.6486486486486487 1104 0.48914488258750555\n",
      "4 [ 259  669  552 1022] 103429 1.204778156996587 14269 0.13795937309652032\n",
      "5 [265 320 361 362] 4032 0.4375 3178 0.7881944444444444\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00031.jpeg\n",
      "7 [ 321  803  738 1022] 91323 0.5251798561151079 62427 0.6835846391380047\n",
      "7 [328   1 754 185] 78384 0.431924882629108 55653 0.7100045927740355\n",
      "2 [325 226 462 278] 7124 0.3795620437956204 3661 0.5138966872543514\n",
      "2 [398  90 593 186] 18720 0.49230769230769234 7967 0.4255876068376068\n",
      "2 [322 699 466 759] 8640 0.4166666666666667 4696 0.5435185185185185\n",
      "2 [390 814 600 906] 19320 0.4380952380952381 8645 0.447463768115942\n",
      "3 [306 666 671 796] 47450 0.3561643835616438 5954 0.12547945205479452\n",
      "3 [238 558 497 590] 8288 0.12355212355212356 3763 0.45402992277992277\n",
      "3 [238 477 606 504] 9936 0.07336956521739131 7503 0.7551328502415459\n",
      "3 [241 386 534 427] 12013 0.13993174061433447 4705 0.39165903604428537\n",
      "3 [270 332 500 358] 5980 0.11304347826086956 2409 0.40284280936454847\n",
      "3 [318 181 674 320] 49484 0.3904494382022472 6567 0.1327095626869291\n",
      "4 [259   4 543 307] 86052 1.0669014084507042 13096 0.1521870496908846\n",
      "4 [240 385 281 446] 2501 1.4878048780487805 1194 0.4774090363854458\n",
      "4 [238 529 277 589] 2340 1.5384615384615385 1134 0.4846153846153846\n",
      "4 [ 258  665  552 1023] 105252 1.217687074829932 13690 0.13006878729145288\n",
      "5 [259 319 362 364] 4635 0.4368932038834951 3407 0.735059331175836\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00032.jpeg\n",
      "7 [ 321  802  734 1022] 90860 0.5326876513317191 62002 0.6823904908650671\n",
      "7 [325   2 760 186] 80040 0.42298850574712643 56676 0.708095952023988\n",
      "2 [326 228 461 276] 6480 0.35555555555555557 3209 0.4952160493827161\n",
      "2 [398  90 593 189] 19305 0.5076923076923077 8258 0.42776482776482777\n",
      "2 [319 700 466 760] 8820 0.40816326530612246 4919 0.5577097505668934\n",
      "2 [387 813 601 904] 19474 0.4252336448598131 9083 0.4664167608092842\n",
      "3 [307 667 669 796] 46698 0.356353591160221 6557 0.14041286564735106\n",
      "3 [238 555 509 590] 9485 0.12915129151291513 4274 0.45060622034791775\n",
      "3 [230 481 571 502] 7161 0.06158357771260997 5023 0.7014383465996369\n",
      "3 [238 384 534 427] 12728 0.14527027027027026 4974 0.3907919547454431\n",
      "3 [270 331 502 358] 6264 0.11637931034482758 2703 0.4315134099616858\n",
      "3 [315 178 674 323] 52055 0.403899721448468 7504 0.14415522043991932\n",
      "4 [254   4 542 310] 88128 1.0625 12499 0.1418277959331881\n",
      "4 [238 384 282 445] 2684 1.3863636363636365 1179 0.4392697466467958\n",
      "4 [238 529 274 589] 2160 1.6666666666666667 1193 0.5523148148148148\n",
      "4 [ 254  665  550 1022] 105672 1.2060810810810811 14391 0.13618555530320237\n",
      "5 [262 321 357 366] 4275 0.47368421052631576 3116 0.7288888888888889\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00033.jpeg\n",
      "7 [ 323  800  743 1022] 93240 0.5285714285714286 63290 0.6787859287859288\n",
      "7 [328   2 763 186] 80040 0.42298850574712643 56315 0.7035857071464268\n",
      "2 [324 227 461 278] 6987 0.3722627737226277 3593 0.5142407327894661\n",
      "2 [397  90 594 187] 19109 0.49238578680203043 8333 0.4360772410905856\n",
      "2 [321 700 466 758] 8410 0.4 4605 0.5475624256837098\n",
      "2 [389 813 601 903] 19080 0.42452830188679247 8550 0.4481132075471698\n",
      "3 [307 670 670 798] 46464 0.3526170798898072 6474 0.13933367768595042\n",
      "3 [238 557 502 590] 8712 0.125 4044 0.4641873278236915\n",
      "3 [238 478 550 503] 7800 0.08012820512820513 6158 0.7894871794871795\n",
      "3 [241 386 524 428] 11886 0.14840989399293286 4016 0.33787649335352515\n",
      "3 [272 334 498 359] 5650 0.11061946902654868 2538 0.44920353982300887\n",
      "3 [317 181 673 322] 50196 0.3960674157303371 6805 0.1355685712008925\n",
      "4 [254   5 540 306] 86086 1.0524475524475525 12490 0.14508747066886601\n",
      "4 [240 387 282 444] 2394 1.3571428571428572 1137 0.474937343358396\n",
      "4 [238 528 275 594] 2442 1.7837837837837838 1367 0.5597870597870598\n",
      "4 [ 254  667  550 1022] 105080 1.1993243243243243 13169 0.12532356299961933\n",
      "5 [262 321 362 365] 4400 0.44 3333 0.7575\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00034.jpeg\n",
      "7 [ 322  802  738 1022] 91520 0.5288461538461539 62423 0.682069493006993\n",
      "7 [328   1 756 184] 78324 0.42757009345794394 55781 0.712182728154844\n",
      "2 [325 227 467 278] 7242 0.3591549295774648 3714 0.5128417564208783\n",
      "2 [397  91 593 189] 19208 0.5 8397 0.437161599333611\n",
      "2 [323 702 463 759] 7980 0.40714285714285714 4341 0.543984962406015\n",
      "2 [389 812 598 902] 18810 0.430622009569378 8462 0.44986709197235514\n",
      "3 [306 666 670 798] 48048 0.3626373626373626 7244 0.15076590076590077\n",
      "3 [237 558 502 588] 7950 0.11320754716981132 3951 0.4969811320754717\n",
      "3 [233 479 530 499] 5940 0.06734006734006734 3780 0.6363636363636364\n",
      "3 [240 386 506 424] 10108 0.14285714285714285 3886 0.38444796201028886\n",
      "3 [270 334 498 358] 5472 0.10526315789473684 2748 0.5021929824561403\n",
      "3 [313 182 674 325] 51623 0.3961218836565097 7730 0.1497394572186816\n",
      "4 [259   3 542 309] 86598 1.0812720848056536 13941 0.16098524215339846\n",
      "4 [240 384 282 444] 2520 1.4285714285714286 1324 0.5253968253968254\n",
      "4 [237 529 275 589] 2280 1.5789473684210527 1164 0.5105263157894737\n",
      "4 [ 256  666  552 1022] 105376 1.2027027027027026 15051 0.14283138475554205\n",
      "5 [258 318 363 365] 4935 0.44761904761904764 3700 0.7497467071935157\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00035.jpeg\n",
      "7 [ 316  801  738 1022] 93262 0.523696682464455 63379 0.6795801076537068\n",
      "3 [305 666 667 798] 47784 0.36464088397790057 8246 0.17256822367319605\n",
      "7 [327   2 762 187] 80475 0.42528735632183906 57203 0.7108170239204722\n",
      "3 [238 557 498 589] 8320 0.12307692307692308 3760 0.4519230769230769\n",
      "3 [232 479 512 498] 5320 0.06785714285714285 4107 0.7719924812030076\n",
      "3 [241 384 491 424] 10000 0.16 3873 0.3873\n",
      "3 [272 331 500 359] 6384 0.12280701754385964 2890 0.4526942355889724\n",
      "3 [318 182 674 323] 50196 0.3960674157303371 6994 0.13933381145908041\n",
      "2 [324 226 463 276] 6950 0.3597122302158273 3612 0.5197122302158274\n",
      "2 [396  86 593 189] 20291 0.5228426395939086 8348 0.4114139273569563\n",
      "2 [321 696 465 765] 9936 0.4791666666666667 4965 0.49969806763285024\n",
      "2 [389 812 601 903] 19292 0.42924528301886794 8504 0.4408044785403276\n",
      "4 [261   3 542 309] 85986 1.0889679715302492 12913 0.15017560998302049\n",
      "4 [240 384 281 446] 2542 1.5121951219512195 1157 0.45515342250196694\n",
      "4 [238 529 275 586] 2109 1.5405405405405406 1145 0.5429113323850165\n",
      "4 [ 256  667  553 1022] 105435 1.1952861952861953 13986 0.1326504481434059\n",
      "5 [264 319 359 363] 4180 0.4631578947368421 3103 0.7423444976076555\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00036.jpeg\n",
      "7 [ 319  798  744 1023] 95625 0.5294117647058824 64487 0.6743738562091504\n",
      "7 [328   2 762 186] 79856 0.423963133640553 56813 0.7114430975756362\n",
      "2 [325 222 466 276] 7614 0.3829787234042553 3673 0.48240084055686894\n",
      "2 [397  91 594 189] 19306 0.49746192893401014 8221 0.42582616803066403\n",
      "2 [321 698 466 759] 8845 0.4206896551724138 4926 0.5569248162803844\n",
      "2 [390 814 602 905] 19292 0.42924528301886794 8975 0.4652187435206303\n",
      "3 [306 670 670 798] 46592 0.3516483516483517 6327 0.13579584478021978\n",
      "3 [243 480 498 499] 4845 0.07450980392156863 3755 0.7750257997936016\n",
      "3 [242 384 486 422] 9272 0.1557377049180328 3017 0.32538826574633306\n",
      "3 [269 331 498 359] 6412 0.1222707423580786 2720 0.42420461634435436\n",
      "3 [318 180 673 318] 48990 0.38873239436619716 6001 0.12249438660951215\n",
      "4 [258   4 540 309] 86010 1.0815602836879432 13657 0.1587838623415882\n",
      "4 [240 384 282 445] 2562 1.4523809523809523 1221 0.4765807962529274\n",
      "4 [238 530 274 589] 2124 1.6388888888888888 988 0.4651600753295669\n",
      "4 [ 256  666  550 1023] 104958 1.2142857142857142 14475 0.13791230778025496\n",
      "5 [258 318 362 366] 4992 0.46153846153846156 3525 0.7061298076923077\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00037.jpeg\n",
      "7 [ 322  803  738 1019] 89856 0.5192307692307693 61227 0.6813902243589743\n",
      "7 [328   2 756 184] 77896 0.4252336448598131 56313 0.7229254390469344\n",
      "2 [323 226 465 276] 7100 0.352112676056338 3877 0.546056338028169\n",
      "2 [394  91 590 190] 19404 0.5051020408163265 8970 0.46227581941867657\n",
      "2 [321 698 466 760] 8990 0.42758620689655175 4688 0.52146829810901\n",
      "2 [387 814 602 903] 19135 0.413953488372093 8961 0.46830415469035797\n",
      "3 [306 670 666 796] 45360 0.35 6215 0.13701499118165786\n",
      "3 [238 479 486 498] 4712 0.07661290322580645 3925 0.8329796264855688\n",
      "3 [241 385 477 422] 8732 0.15677966101694915 3608 0.41319285387082\n",
      "3 [270 332 501 358] 6006 0.11255411255411256 2900 0.4828504828504829\n",
      "3 [319 181 671 320] 48928 0.39488636363636365 6269 0.12812704381948986\n",
      "4 [259   4 542 307] 85749 1.0706713780918728 14401 0.16794364948862378\n",
      "4 [240 385 282 446] 2562 1.4523809523809523 1262 0.492583918813427\n",
      "4 [237 527 275 590] 2394 1.6578947368421053 1240 0.5179615705931495\n",
      "4 [ 259  667  552 1022] 104015 1.21160409556314 14271 0.1372013651877133\n",
      "5 [260 319 361 365] 4646 0.45544554455445546 3523 0.7582866982350409\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00038.jpeg\n",
      "7 [ 321  802  738 1022] 91740 0.5275779376498801 62971 0.6864072378460868\n",
      "7 [330   2 756 186] 78384 0.431924882629108 55232 0.704633598693611\n",
      "2 [325 228 462 276] 6576 0.35036496350364965 3426 0.5209854014598541\n",
      "2 [398  91 593 186] 18525 0.48717948717948717 8022 0.43303643724696356\n",
      "2 [321 703 466 762] 8555 0.4068965517241379 4629 0.5410870835768556\n",
      "2 [390 812 595 905] 19065 0.45365853658536587 8631 0.45271439811172304\n",
      "3 [306 669 668 794] 45250 0.3453038674033149 5946 0.13140331491712706\n",
      "3 [243 479 472 498] 4351 0.08296943231441048 3904 0.8972649965525167\n",
      "3 [242 386 474 422] 8352 0.15517241379310345 3352 0.4013409961685824\n",
      "3 [270 331 498 358] 6156 0.11842105263157894 2804 0.4554905782975958\n",
      "3 [321 178 673 320] 49984 0.4034090909090909 5858 0.11719750320102433\n",
      "4 [255   4 542 309] 87535 1.0627177700348431 13887 0.15864511338321813\n",
      "4 [241 387 281 446] 2360 1.475 1094 0.4635593220338983\n",
      "4 [238 528 274 589] 2196 1.6944444444444444 1060 0.48269581056466304\n",
      "4 [ 253  666  552 1022] 106444 1.1906354515050168 14004 0.13156213595881402\n",
      "5 [258 319 365 366] 5029 0.4392523364485981 3452 0.6864187711274607\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00039.jpeg\n",
      "7 [ 321  801  738 1024] 92991 0.5347721822541966 63302 0.6807325440096353\n",
      "7 [329   2 757 186] 78752 0.42990654205607476 55989 0.710953372612759\n",
      "2 [323 227 462 276] 6811 0.35251798561151076 3579 0.5254734987520188\n",
      "2 [397  93 593 187] 18424 0.47959183673469385 7767 0.4215696917064698\n",
      "2 [322 699 465 762] 9009 0.4405594405594406 4886 0.5423465423465423\n",
      "2 [389 814 599 902] 18480 0.41904761904761906 8715 0.4715909090909091\n",
      "3 [306 669 664 797] 45824 0.3575418994413408 7321 0.15976344273743018\n",
      "3 [241 478 456 498] 4300 0.09302325581395349 3539 0.8230232558139535\n",
      "3 [241 386 477 420] 8024 0.1440677966101695 2906 0.36216350947158527\n",
      "3 [270 332 498 358] 5928 0.11403508771929824 2503 0.42223346828609987\n",
      "3 [318 179 673 319] 49700 0.39436619718309857 6150 0.12374245472837023\n",
      "4 [254   2 540 309] 87802 1.0734265734265733 13892 0.15821963053233412\n",
      "4 [240 382 280 446] 2560 1.6 1181 0.461328125\n",
      "4 [239 529 275 588] 2124 1.6388888888888888 1132 0.5329566854990584\n",
      "4 [ 256  666  553 1023] 106029 1.202020202020202 15315 0.1444416150298503\n",
      "5 [260 318 358 368] 4900 0.5102040816326531 3412 0.6963265306122449\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00040.jpeg\n",
      "7 [ 318  799  741 1022] 94329 0.5271867612293144 64274 0.6813811235145077\n",
      "7 [325   3 756 186] 78873 0.4245939675174014 56180 0.7122843051487835\n",
      "2 [323 226 462 277] 7089 0.3669064748201439 3806 0.5368881365495839\n",
      "2 [397  88 590 191] 19879 0.533678756476684 8485 0.42683233563056494\n",
      "2 [389 809 601 902] 19716 0.4386792452830189 8776 0.44512071414079935\n",
      "2 [321 700 466 759] 8555 0.4068965517241379 4830 0.5645821157218002\n",
      "3 [241 480 442 498] 3618 0.08955223880597014 3175 0.8775566611387506\n",
      "3 [242 390 486 422] 7808 0.13114754098360656 2927 0.3748719262295082\n",
      "3 [270 332 498 361] 6612 0.12719298245614036 3021 0.45689655172413796\n",
      "3 [317 181 673 320] 49484 0.3904494382022472 6555 0.13246706005981732\n",
      "4 [258   5 542 312] 87188 1.0809859154929577 13798 0.15825572326466944\n",
      "4 [240 384 282 446] 2604 1.4761904761904763 1217 0.467357910906298\n",
      "4 [237 528 274 589] 2257 1.6486486486486487 1029 0.4559149313247674\n",
      "4 [ 258  666  551 1022] 104308 1.2150170648464165 14496 0.13897304137745906\n",
      "5 [262 321 358 366] 4320 0.46875 3367 0.7793981481481481\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00041.jpeg\n",
      "7 [ 318  801  741 1023] 93906 0.524822695035461 64480 0.68664409089941\n",
      "7 [329   2 761 186] 79488 0.42592592592592593 56484 0.7105978260869565\n",
      "2 [325 226 466 276] 7050 0.3546099290780142 3788 0.5373049645390071\n",
      "2 [398  91 593 189] 19110 0.5025641025641026 7967 0.41690214547357407\n",
      "2 [323 702 466 758] 8008 0.3916083916083916 4581 0.5720529470529471\n",
      "2 [391 814 601 904] 18900 0.42857142857142855 8971 0.47465608465608466\n",
      "3 [308 670 665 796] 44982 0.35294117647058826 5367 0.11931439242363612\n",
      "3 [244 478 428 497] 3496 0.10326086956521739 2853 0.8160755148741419\n",
      "3 [241 385 500 424] 10101 0.15057915057915058 3658 0.3621423621423621\n",
      "3 [270 333 500 359] 5980 0.11304347826086956 3117 0.5212374581939799\n",
      "3 [318 178 671 319] 49773 0.39943342776203966 6716 0.134932593976654\n",
      "4 [256   4 542 309] 87230 1.0664335664335665 14340 0.1643929840651152\n",
      "4 [241 386 280 445] 2301 1.5128205128205128 1163 0.505432420686658\n",
      "4 [238 528 274 588] 2160 1.6666666666666667 1064 0.4925925925925926\n",
      "4 [ 255  664  552 1022] 106326 1.2053872053872055 15364 0.14449899366100483\n",
      "5 [258 318 357 363] 4455 0.45454545454545453 3268 0.7335578002244669\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00042.jpeg\n",
      "7 [ 321  800  740 1022] 93018 0.5298329355608592 63445 0.6820722870842203\n",
      "7 [327   2 761 186] 79856 0.423963133640553 56691 0.7099153476257263\n",
      "2 [322 225 463 276] 7191 0.3617021276595745 3850 0.5353914615491586\n",
      "2 [396  89 593 190] 19897 0.5126903553299492 8439 0.424134291601749\n",
      "2 [319 700 474 762] 9610 0.4 5181 0.5391259105098856\n",
      "2 [389 812 604 905] 19995 0.4325581395348837 9019 0.45106276569142284\n",
      "3 [310 666 666 798] 46992 0.3707865168539326 7482 0.15921859039836567\n",
      "3 [238 478 413 497] 3325 0.10857142857142857 3034 0.9124812030075188\n",
      "3 [242 386 518 425] 10764 0.14130434782608695 3945 0.3664994425863991\n",
      "3 [270 331 500 358] 6210 0.11739130434782609 2777 0.4471819645732689\n",
      "3 [317 179 673 322] 50908 0.40168539325842695 7148 0.1404101516461067\n",
      "4 [255   2 544 310] 89012 1.065743944636678 14165 0.15913584685210982\n",
      "4 [240 383 281 446] 2583 1.5365853658536586 1188 0.45993031358885017\n",
      "4 [237 528 274 591] 2331 1.7027027027027026 1143 0.49034749034749037\n",
      "4 [ 255  662  551 1022] 106560 1.2162162162162162 14950 0.14029654654654655\n",
      "5 [261 318 359 364] 4508 0.46938775510204084 3547 0.7868234250221828\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00043.jpeg\n",
      "7 [ 322  803  738 1022] 91104 0.5264423076923077 62647 0.687642694063927\n",
      "7 [327   2 755 186] 78752 0.42990654205607476 56338 0.7153850060950833\n",
      "2 [325 228 465 274] 6440 0.32857142857142857 3491 0.5420807453416149\n",
      "2 [398  90 595 190] 19700 0.5076142131979695 8500 0.43147208121827413\n",
      "2 [322 700 466 760] 8640 0.4166666666666667 4670 0.5405092592592593\n",
      "2 [388 811 598 907] 20160 0.45714285714285713 8921 0.4425099206349206\n",
      "3 [310 663 670 797] 48240 0.37222222222222223 7864 0.16301824212271973\n",
      "3 [263 618 428 644] 4290 0.15757575757575756 2102 0.48997668997668997\n",
      "3 [241 478 402 496] 2898 0.11180124223602485 2718 0.937888198757764\n",
      "3 [241 386 520 424] 10602 0.13620071684587814 3842 0.36238445576306355\n",
      "3 [272 332 500 359] 6156 0.11842105263157894 2908 0.4723846653671215\n",
      "3 [318 179 674 322] 50908 0.40168539325842695 7227 0.141961970613656\n",
      "4 [259   6 542 313] 86881 1.0848056537102473 15077 0.17353621620377296\n",
      "4 [240 386 278 446] 2280 1.5789473684210527 1243 0.5451754385964912\n",
      "4 [237 529 274 590] 2257 1.6486486486486487 1235 0.5471865307930882\n",
      "4 [ 256  666  550 1023] 104958 1.2142857142857142 15510 0.14777339507231463\n",
      "5 [260 319 357 369] 4850 0.5154639175257731 3542 0.7303092783505155\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00044.jpeg\n",
      "7 [ 318  801  738 1023] 93240 0.5285714285714286 63990 0.6862934362934363\n",
      "7 [328   1 760 187] 80352 0.4305555555555556 56697 0.705607825567503\n",
      "2 [325 228 465 276] 6720 0.34285714285714286 3626 0.5395833333333333\n",
      "2 [398  88 592 190] 19788 0.5257731958762887 8840 0.44673539518900346\n",
      "2 [320 700 466 762] 9052 0.4246575342465753 4846 0.5353513035793195\n",
      "2 [389 811 601 904] 19716 0.4386792452830189 8663 0.43938932846419154\n",
      "3 [302 671 668 796] 45750 0.34153005464480873 6032 0.13184699453551912\n",
      "3 [264 618 431 644] 4342 0.15568862275449102 1916 0.44127130354675265\n",
      "3 [240 478 391 496] 2718 0.11920529801324503 2324 0.8550404709345106\n",
      "3 [240 384 507 424] 10680 0.149812734082397 4007 0.375187265917603\n",
      "3 [270 332 496 358] 5876 0.11504424778761062 2479 0.4218856364874064\n",
      "3 [317 181 673 322] 50196 0.3960674157303371 6344 0.1263845724758945\n",
      "4 [258   3 544 307] 86944 1.062937062937063 13200 0.15182186234817813\n",
      "4 [240 384 281 446] 2542 1.5121951219512195 1220 0.47993705743509046\n",
      "4 [236 528 274 589] 2318 1.605263157894737 1183 0.5103537532355479\n",
      "4 [ 258  666  554 1023] 105672 1.2060810810810811 14976 0.14172155348625937\n",
      "5 [260 319 346 365] 3956 0.5348837209302325 3172 0.8018200202224469\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00045.jpeg\n",
      "7 [ 321  801  737 1022] 91936 0.53125 64045 0.6966259136790811\n",
      "7 [307 667 669 797] 47060 0.35911602209944754 6836 0.14526136846578835\n",
      "3 [262 618 483 644] 5746 0.11764705882352941 2916 0.5074834667594849\n",
      "3 [238 479 384 496] 2482 0.11643835616438356 2296 0.9250604351329573\n",
      "3 [240 385 498 424] 10062 0.1511627906976744 4112 0.40866626913138543\n",
      "3 [271 331 502 359] 6468 0.12121212121212122 2717 0.4200680272108844\n",
      "3 [316 178 673 320] 50694 0.39775910364145656 6332 0.12490630054838837\n",
      "2 [394  85 592 187] 20196 0.5151515151515151 7896 0.3909685086155674\n",
      "2 [325 226 462 278] 7124 0.3795620437956204 3465 0.4863840539023021\n",
      "2 [321 698 469 759] 9028 0.41216216216216217 4931 0.546189632255206\n",
      "2 [389 812 602 902] 19170 0.4225352112676056 9090 0.47417840375586856\n",
      "4 [240 384 281 446] 2542 1.5121951219512195 1264 0.4972462627852085\n",
      "4 [259   2 542 307] 86315 1.0777385159010602 13135 0.1521751723338933\n",
      "4 [237 530 275 589] 2242 1.5526315789473684 1028 0.45851917930419267\n",
      "4 [ 258  663  550 1022] 104828 1.2294520547945205 15097 0.14401686572289846\n",
      "5 [262 320 357 364] 4180 0.4631578947368421 3395 0.812200956937799\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00046.jpeg\n",
      "7 [ 322  801  738 1022] 91936 0.53125 63335 0.6889031500174034\n",
      "7 [325   2 757 185] 79056 0.4236111111111111 56204 0.7109390811576604\n",
      "2 [326 226 462 276] 6800 0.36764705882352944 3503 0.5151470588235294\n",
      "2 [398  87 592 190] 19982 0.5309278350515464 8202 0.4104694224802322\n",
      "2 [319 700 466 762] 9114 0.4217687074829932 4910 0.5387316216809305\n",
      "2 [390 810 598 904] 19552 0.4519230769230769 8975 0.4590323240589198\n",
      "3 [308 670 666 798] 45824 0.3575418994413408 6787 0.14811016061452514\n",
      "3 [265 618 492 642] 5448 0.10572687224669604 2618 0.48054331864904554\n",
      "3 [242 477 378 496] 2584 0.13970588235294118 2389 0.9245356037151703\n",
      "3 [242 386 494 422] 9072 0.14285714285714285 3640 0.4012345679012346\n",
      "3 [270 331 502 358] 6264 0.11637931034482758 2828 0.45146871008939976\n",
      "3 [317 180 673 319] 49484 0.3904494382022472 6096 0.1231913345727912\n",
      "4 [260   2 543 306] 86032 1.0742049469964665 13248 0.15398921331597545\n",
      "4 [239 385 282 446] 2623 1.4186046511627908 1270 0.48417842165459396\n",
      "4 [237 530 274 588] 2146 1.5675675675675675 1021 0.47576887232059645\n",
      "4 [ 252  665  550 1023] 106684 1.2013422818791946 15672 0.14690112856660792\n",
      "5 [260 320 356 365] 4320 0.46875 3177 0.7354166666666667\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00047.jpeg\n",
      "7 [ 319  801  741 1022] 93262 0.523696682464455 63238 0.678068237867513\n",
      "7 [328   2 757 186] 78936 0.4289044289044289 55696 0.7055842708016621\n",
      "2 [323 224 463 276] 7280 0.37142857142857144 3809 0.5232142857142857\n",
      "2 [396  92 594 190] 19404 0.494949494949495 8341 0.4298598227169656\n",
      "2 [321 702 466 759] 8265 0.3931034482758621 5091 0.6159709618874774\n",
      "2 [385 811 602 905] 20398 0.43317972350230416 9461 0.4638199823512109\n",
      "3 [306 666 668 797] 47422 0.36187845303867405 6615 0.1394922188014002\n",
      "3 [265 615 495 644] 6670 0.12608695652173912 3357 0.5032983508245877\n",
      "3 [236 479 375 496] 2363 0.1223021582733813 2113 0.894202285230639\n",
      "3 [241 386 491 423] 9250 0.148 3921 0.4238918918918919\n",
      "3 [270 332 502 360] 6496 0.1206896551724138 3184 0.49014778325123154\n",
      "3 [318 185 665 318] 46151 0.38328530259365995 2171 0.047041234209442916\n",
      "4 [256   3 543 310] 88109 1.0696864111498259 14603 0.1657378928372811\n",
      "4 [240 386 282 446] 2520 1.4285714285714286 1160 0.4603174603174603\n",
      "4 [237 529 274 590] 2257 1.6486486486486487 1091 0.4833850243686309\n",
      "4 [ 256  667  554 1023] 106088 1.1946308724832215 14985 0.14125065982957544\n",
      "5 [261 318 358 363] 4365 0.4639175257731959 3378 0.7738831615120275\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00048.jpeg\n",
      "7 [ 321  801  740 1020] 91761 0.522673031026253 62891 0.6853783197654777\n",
      "7 [329   2 756 186] 78568 0.4309133489461358 55671 0.7085709194583036\n",
      "2 [322 226 462 277] 7140 0.36428571428571427 3591 0.5029411764705882\n",
      "2 [398  91 593 186] 18525 0.48717948717948717 7796 0.4208367071524966\n",
      "2 [321 698 467 762] 9344 0.4383561643835616 5012 0.5363869863013698\n",
      "2 [389 811 602 904] 19809 0.43661971830985913 9241 0.4665051239335655\n",
      "3 [310 666 671 798] 47652 0.3656509695290859 7348 0.15420129270544783\n",
      "3 [262 618 497 644] 6110 0.11063829787234042 2864 0.46873977086743046\n",
      "3 [240 478 388 498] 2960 0.13513513513513514 2875 0.9712837837837838\n",
      "3 [238 383 500 426] 11266 0.16412213740458015 4577 0.406266642996627\n",
      "3 [269 332 502 359] 6291 0.11587982832618025 2786 0.4428548720394214\n",
      "3 [318 178 674 322] 51264 0.4044943820224719 8238 0.16069756554307116\n",
      "4 [254   5 544 307] 87580 1.0413793103448277 12837 0.14657456040191824\n",
      "4 [240 383 282 446] 2646 1.5 1335 0.5045351473922902\n",
      "4 [238 528 274 589] 2196 1.6944444444444444 1056 0.4808743169398907\n",
      "4 [ 256  666  553 1022] 105732 1.1986531986531987 15944 0.15079635304354405\n",
      "5 [258 318 357 366] 4752 0.48484848484848486 3707 0.7800925925925926\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00049.jpeg\n",
      "7 [ 318  799  741 1022] 94329 0.5271867612293144 64199 0.680586033987427\n",
      "7 [327   2 755 189] 80036 0.4369158878504673 57471 0.7180643710330351\n",
      "2 [323 222 462 277] 7645 0.39568345323741005 4010 0.5245258338783518\n",
      "2 [398  88 592 187] 19206 0.5103092783505154 8157 0.4247110278038113\n",
      "2 [322 701 466 759] 8352 0.4027777777777778 4687 0.5611829501915708\n",
      "2 [389 811 602 905] 20022 0.4413145539906103 9335 0.4662371391469384\n",
      "3 [302 666 666 799] 48412 0.36538461538461536 7220 0.14913657770800628\n",
      "3 [264 618 495 643] 5775 0.10822510822510822 3033 0.5251948051948052\n",
      "3 [238 477 393 496] 2945 0.12258064516129032 2763 0.9382003395585738\n",
      "3 [241 384 502 424] 10440 0.1532567049808429 3552 0.34022988505747126\n",
      "3 [267 330 501 360] 7020 0.1282051282051282 3177 0.45256410256410257\n",
      "3 [317 180 678 324] 51984 0.3988919667590028 7379 0.14194752231455832\n",
      "4 [258   3 541 307] 86032 1.0742049469964665 13424 0.1560349637344244\n",
      "4 [240 383 282 446] 2646 1.5 1297 0.49017384731670444\n",
      "4 [235 528 274 591] 2457 1.6153846153846154 1175 0.47822547822547823\n",
      "4 [ 256  666  552 1022] 105376 1.2027027027027026 14658 0.13910188278165805\n",
      "5 [260 319 362 364] 4590 0.4411764705882353 3637 0.7923747276688453\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00050.jpeg\n",
      "7 [ 321  801  741 1023] 93240 0.5285714285714286 62830 0.6738524238524238\n",
      "7 [326   2 766 187] 81400 0.42045454545454547 57555 0.707063882063882\n",
      "2 [326 226 462 276] 6800 0.36764705882352944 3528 0.5188235294117647\n",
      "2 [398  88 590 190] 19584 0.53125 8606 0.4394403594771242\n",
      "2 [321 701 466 758] 8265 0.3931034482758621 4765 0.5765275257108288\n",
      "2 [389 811 602 907] 20448 0.4507042253521127 9545 0.4667938184663537\n",
      "3 [308 668 671 796] 46464 0.3526170798898072 7182 0.15457128099173553\n",
      "3 [264 615 496 647] 7424 0.13793103448275862 3740 0.5037715517241379\n",
      "3 [242 477 392 498] 3150 0.14 3000 0.9523809523809523\n",
      "3 [241 384 514 425] 11193 0.15018315018315018 4113 0.36746180648619675\n",
      "3 [270 331 500 360] 6670 0.12608695652173912 2881 0.43193403298350824\n",
      "3 [317 179 674 320] 50337 0.3949579831932773 6636 0.13183145598664997\n",
      "4 [259   2 541 307] 86010 1.0815602836879432 14644 0.17025927217765377\n",
      "4 [240 384 282 446] 2604 1.4761904761904763 1219 0.4681259600614439\n",
      "4 [237 527 275 591] 2432 1.6842105263157894 1317 0.5415296052631579\n",
      "4 [ 259  664  552 1022] 104894 1.2218430034129693 15926 0.1518294659370412\n",
      "5 [261 319 362 366] 4747 0.46534653465346537 3805 0.8015588792921845\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00051.jpeg\n",
      "7 [ 321  799  741 1022] 93660 0.530952380952381 64915 0.6930920350202862\n",
      "7 [329   1 762 186] 80105 0.42725173210161665 57052 0.7122152175269958\n",
      "2 [321 227 466 276] 7105 0.33793103448275863 4026 0.566643209007741\n",
      "2 [394  89 595 187] 19698 0.48756218905472637 8619 0.4375571123971977\n",
      "2 [321 698 466 762] 9280 0.4413793103448276 5317 0.5729525862068966\n",
      "2 [389 811 602 905] 20022 0.4413145539906103 9528 0.47587653581060835\n",
      "3 [308 669 669 794] 45125 0.3462603878116344 6255 0.13861495844875346\n",
      "3 [262 618 497 644] 6110 0.11063829787234042 2950 0.4828150572831424\n",
      "3 [230 478 430 496] 3600 0.09 3044 0.8455555555555555\n",
      "3 [240 383 523 426] 12169 0.1519434628975265 3888 0.3195003697920947\n",
      "3 [270 331 502 359] 6496 0.1206896551724138 2728 0.41995073891625617\n",
      "3 [318 178 673 322] 51120 0.4056338028169014 7223 0.14129499217527386\n",
      "4 [258   4 542 308] 86336 1.0704225352112675 14880 0.17234988880652335\n",
      "4 [238 384 280 446] 2604 1.4761904761904763 1176 0.45161290322580644\n",
      "4 [238 529 277 589] 2340 1.5384615384615385 1268 0.5418803418803418\n",
      "4 [ 256  666  553 1022] 105732 1.1986531986531987 14844 0.1403926909544887\n",
      "5 [259 317 358 365] 4752 0.48484848484848486 3603 0.7582070707070707\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00052.jpeg\n",
      "7 [ 321  801  744 1022] 93483 0.5224586288416075 63923 0.6837927751569804\n",
      "7 [327   2 757 186] 79120 0.42790697674418604 56382 0.7126137512639029\n",
      "2 [327 226 462 276] 6750 0.37037037037037035 3549 0.5257777777777778\n",
      "2 [394  83 594 189] 21200 0.53 8362 0.39443396226415095\n",
      "2 [321 700 466 760] 8700 0.41379310344827586 4808 0.5526436781609195\n",
      "2 [389 809 602 905] 20448 0.4507042253521127 9630 0.4709507042253521\n",
      "3 [306 667 671 798] 47815 0.3589041095890411 7956 0.1663912998013176\n",
      "3 [264 618 497 646] 6524 0.12017167381974249 3396 0.5205395462906193\n",
      "3 [238 479 439 497] 3618 0.08955223880597014 3571 0.9870093974571587\n",
      "3 [241 382 530 427] 13005 0.15570934256055363 4680 0.35986159169550175\n",
      "3 [272 330 502 358] 6440 0.12173913043478261 2854 0.443167701863354\n",
      "3 [318 179 676 322] 51194 0.3994413407821229 7147 0.13960620385201392\n",
      "4 [252   4 542 309] 88450 1.0517241379310345 13366 0.1511136235161108\n",
      "4 [239 382 281 444] 2604 1.4761904761904763 1138 0.43701996927803377\n",
      "4 [238 530 274 590] 2160 1.6666666666666667 1105 0.5115740740740741\n",
      "4 [ 260  666  550 1022] 103240 1.2275862068965517 14386 0.13934521503293298\n",
      "5 [262 318 362 365] 4700 0.47 3320 0.7063829787234043\n",
      "[{'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00028.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  803,  740, 1022]), 'bbox_area': 91761, 'wh_ratio': 0.522673031026253, 'mask_area': 62338, 'mask_ratio': 0.679351794335284, 'size': 302.92078172353905}, {'class_id': 7, 'bbox': array([329,   1, 757, 186]), 'bbox_area': 79180, 'wh_ratio': 0.4322429906542056, 'mask_area': 56305, 'mask_ratio': 0.7111012882040919, 'size': 281.389409893123}, {'class_id': 2, 'bbox': array([321, 702, 465, 760]), 'bbox_area': 8352, 'wh_ratio': 0.4027777777777778, 'mask_area': 4294, 'mask_ratio': 0.5141283524904214, 'size': 91.3892772703669}, {'class_id': 2, 'bbox': array([390, 812, 600, 902]), 'bbox_area': 18900, 'wh_ratio': 0.42857142857142855, 'mask_area': 8716, 'mask_ratio': 0.46116402116402117, 'size': 137.4772708486752}, {'class_id': 2, 'bbox': array([326, 226, 465, 274]), 'bbox_area': 6672, 'wh_ratio': 0.34532374100719426, 'mask_area': 3120, 'mask_ratio': 0.4676258992805755, 'size': 81.68231142664855}, {'class_id': 2, 'bbox': array([398,  90, 592, 186]), 'bbox_area': 18624, 'wh_ratio': 0.4948453608247423, 'mask_area': 7705, 'mask_ratio': 0.41371348797250856, 'size': 136.46977687385584}, {'class_id': 3, 'bbox': array([318, 181, 674, 320]), 'bbox_area': 49484, 'wh_ratio': 0.3904494382022472, 'mask_area': 5989, 'mask_ratio': 0.12102901948104437, 'size': 222.44999438075965}, {'class_id': 3, 'bbox': array([269, 332, 498, 358]), 'bbox_area': 5954, 'wh_ratio': 0.11353711790393013, 'mask_area': 2280, 'mask_ratio': 0.3829358414511253, 'size': 77.1621668954417}, {'class_id': 3, 'bbox': array([241, 387, 535, 425]), 'bbox_area': 11172, 'wh_ratio': 0.1292517006802721, 'mask_area': 3385, 'mask_ratio': 0.30298961689939136, 'size': 105.69768209379049}, {'class_id': 3, 'bbox': array([237, 562, 455, 590]), 'bbox_area': 6104, 'wh_ratio': 0.12844036697247707, 'mask_area': 2791, 'mask_ratio': 0.45724115334207077, 'size': 78.12809993849844}, {'class_id': 3, 'bbox': array([240, 480, 619, 502]), 'bbox_area': 8338, 'wh_ratio': 0.05804749340369393, 'mask_area': 5445, 'mask_ratio': 0.6530343007915568, 'size': 91.31264972609216}, {'class_id': 3, 'bbox': array([307, 670, 668, 798]), 'bbox_area': 46208, 'wh_ratio': 0.3545706371191136, 'mask_area': 5563, 'mask_ratio': 0.12039040858725762, 'size': 214.96046148071045}, {'class_id': 4, 'bbox': array([266,   5, 543, 308]), 'bbox_area': 83931, 'wh_ratio': 1.0938628158844765, 'mask_area': 12823, 'mask_ratio': 0.15278025997545602, 'size': 289.70847415980086}, {'class_id': 4, 'bbox': array([241, 386, 283, 444]), 'bbox_area': 2436, 'wh_ratio': 1.380952380952381, 'mask_area': 1230, 'mask_ratio': 0.5049261083743842, 'size': 49.35585071701227}, {'class_id': 4, 'bbox': array([236, 529, 275, 591]), 'bbox_area': 2418, 'wh_ratio': 1.5897435897435896, 'mask_area': 1166, 'mask_ratio': 0.48221670802315963, 'size': 49.17316341257698}, {'class_id': 4, 'bbox': array([ 256,  670,  550, 1023]), 'bbox_area': 103782, 'wh_ratio': 1.2006802721088434, 'mask_area': 13409, 'mask_ratio': 0.12920352276888092, 'size': 322.1521379721078}, {'class_id': 5, 'bbox': array([265, 321, 362, 366]), 'bbox_area': 4365, 'wh_ratio': 0.4639175257731959, 'mask_area': 3364, 'mask_ratio': 0.7706758304696449, 'size': 66.06814663663572}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00029.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 322,  799,  738, 1023]), 'bbox_area': 93184, 'wh_ratio': 0.5384615384615384, 'mask_area': 63565, 'mask_ratio': 0.6821450034340659, 'size': 305.26054445342265}, {'class_id': 7, 'bbox': array([326,   1, 761, 186]), 'bbox_area': 80475, 'wh_ratio': 0.42528735632183906, 'mask_area': 57073, 'mask_ratio': 0.709201615408512, 'size': 283.6811590500857}, {'class_id': 3, 'bbox': array([310, 671, 665, 798]), 'bbox_area': 45085, 'wh_ratio': 0.35774647887323946, 'mask_area': 5994, 'mask_ratio': 0.13294887434845293, 'size': 212.33228675827894}, {'class_id': 3, 'bbox': array([241, 476, 613, 504]), 'bbox_area': 10416, 'wh_ratio': 0.07526881720430108, 'mask_area': 6936, 'mask_ratio': 0.6658986175115207, 'size': 102.05880657738459}, {'class_id': 3, 'bbox': array([240, 386, 534, 427]), 'bbox_area': 12054, 'wh_ratio': 0.13945578231292516, 'mask_area': 4029, 'mask_ratio': 0.33424589347934297, 'size': 109.79070998950685}, {'class_id': 3, 'bbox': array([270, 332, 498, 359]), 'bbox_area': 6156, 'wh_ratio': 0.11842105263157894, 'mask_area': 2780, 'mask_ratio': 0.45159194282001297, 'size': 78.46018098373213}, {'class_id': 3, 'bbox': array([317, 181, 674, 320]), 'bbox_area': 49623, 'wh_ratio': 0.38935574229691877, 'mask_area': 6267, 'mask_ratio': 0.12629224351611149, 'size': 222.7622050528321}, {'class_id': 4, 'bbox': array([255,   2, 542, 311]), 'bbox_area': 88683, 'wh_ratio': 1.0766550522648084, 'mask_area': 14036, 'mask_ratio': 0.15827159658559137, 'size': 297.7969106622834}, {'class_id': 4, 'bbox': array([241, 384, 282, 445]), 'bbox_area': 2501, 'wh_ratio': 1.4878048780487805, 'mask_area': 1223, 'mask_ratio': 0.48900439824070374, 'size': 50.00999900019995}, {'class_id': 4, 'bbox': array([237, 530, 274, 590]), 'bbox_area': 2220, 'wh_ratio': 1.6216216216216217, 'mask_area': 990, 'mask_ratio': 0.44594594594594594, 'size': 47.11687595755899}, {'class_id': 4, 'bbox': array([ 254,  669,  552, 1022]), 'bbox_area': 105194, 'wh_ratio': 1.1845637583892616, 'mask_area': 12974, 'mask_ratio': 0.12333403045801092, 'size': 324.33624527641064}, {'class_id': 5, 'bbox': array([264, 318, 362, 366]), 'bbox_area': 4704, 'wh_ratio': 0.4897959183673469, 'mask_area': 3584, 'mask_ratio': 0.7619047619047619, 'size': 68.58571279792898}, {'class_id': 2, 'bbox': array([390, 811, 602, 903]), 'bbox_area': 19504, 'wh_ratio': 0.4339622641509434, 'mask_area': 8748, 'mask_ratio': 0.4485233798195242, 'size': 139.65672200076872}, {'class_id': 2, 'bbox': array([321, 699, 466, 762]), 'bbox_area': 9135, 'wh_ratio': 0.43448275862068964, 'mask_area': 4766, 'mask_ratio': 0.5217296113847838, 'size': 95.57719393244396}, {'class_id': 2, 'bbox': array([324, 227, 462, 277]), 'bbox_area': 6900, 'wh_ratio': 0.36231884057971014, 'mask_area': 3641, 'mask_ratio': 0.5276811594202898, 'size': 83.06623862918076}, {'class_id': 2, 'bbox': array([394,  88, 593, 189]), 'bbox_area': 20099, 'wh_ratio': 0.507537688442211, 'mask_area': 8484, 'mask_ratio': 0.4221105527638191, 'size': 141.7709420156331}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00030.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 319,  801,  738, 1022]), 'bbox_area': 92599, 'wh_ratio': 0.5274463007159904, 'mask_area': 63430, 'mask_ratio': 0.6849965982354022, 'size': 304.300837987673}, {'class_id': 7, 'bbox': array([328,   2, 756, 186]), 'bbox_area': 78752, 'wh_ratio': 0.42990654205607476, 'mask_area': 55900, 'mask_ratio': 0.7098232425843153, 'size': 280.6278674686461}, {'class_id': 2, 'bbox': array([323, 699, 466, 758]), 'bbox_area': 8437, 'wh_ratio': 0.4125874125874126, 'mask_area': 4573, 'mask_ratio': 0.5420173047291691, 'size': 91.8531436587774}, {'class_id': 2, 'bbox': array([389, 811, 601, 904]), 'bbox_area': 19716, 'wh_ratio': 0.4386792452830189, 'mask_area': 8850, 'mask_ratio': 0.44887401095556906, 'size': 140.4136745477448}, {'class_id': 2, 'bbox': array([325, 230, 465, 276]), 'bbox_area': 6440, 'wh_ratio': 0.32857142857142857, 'mask_area': 3385, 'mask_ratio': 0.5256211180124224, 'size': 80.24961059095551}, {'class_id': 2, 'bbox': array([398,  87, 592, 186]), 'bbox_area': 19206, 'wh_ratio': 0.5103092783505154, 'mask_area': 7915, 'mask_ratio': 0.41211079870873685, 'size': 138.58571354941319}, {'class_id': 3, 'bbox': array([318, 182, 675, 322]), 'bbox_area': 49980, 'wh_ratio': 0.39215686274509803, 'mask_area': 6077, 'mask_ratio': 0.12158863545418168, 'size': 223.56207191739838}, {'class_id': 3, 'bbox': array([270, 332, 500, 357]), 'bbox_area': 5750, 'wh_ratio': 0.10869565217391304, 'mask_area': 2312, 'mask_ratio': 0.40208695652173915, 'size': 75.82875444051551}, {'class_id': 3, 'bbox': array([243, 387, 537, 427]), 'bbox_area': 11760, 'wh_ratio': 0.1360544217687075, 'mask_area': 3967, 'mask_ratio': 0.3373299319727891, 'size': 108.44353369380767}, {'class_id': 3, 'bbox': array([241, 479, 617, 503]), 'bbox_area': 9024, 'wh_ratio': 0.06382978723404255, 'mask_area': 5828, 'mask_ratio': 0.6458333333333334, 'size': 94.99473669630335}, {'class_id': 3, 'bbox': array([238, 557, 497, 591]), 'bbox_area': 8806, 'wh_ratio': 0.13127413127413126, 'mask_area': 4338, 'mask_ratio': 0.49261866908925733, 'size': 93.84028985462481}, {'class_id': 3, 'bbox': array([308, 668, 667, 799]), 'bbox_area': 47029, 'wh_ratio': 0.3649025069637883, 'mask_area': 6803, 'mask_ratio': 0.14465542537583193, 'size': 216.86170708541422}, {'class_id': 4, 'bbox': array([265,   2, 542, 309]), 'bbox_area': 85039, 'wh_ratio': 1.108303249097473, 'mask_area': 12612, 'mask_ratio': 0.1483084231940639, 'size': 291.61447152019053}, {'class_id': 4, 'bbox': array([241, 386, 282, 446]), 'bbox_area': 2460, 'wh_ratio': 1.4634146341463414, 'mask_area': 1122, 'mask_ratio': 0.4560975609756098, 'size': 49.59838707054898}, {'class_id': 4, 'bbox': array([238, 529, 275, 590]), 'bbox_area': 2257, 'wh_ratio': 1.6486486486486487, 'mask_area': 1104, 'mask_ratio': 0.48914488258750555, 'size': 47.507894080878806}, {'class_id': 4, 'bbox': array([ 259,  669,  552, 1022]), 'bbox_area': 103429, 'wh_ratio': 1.204778156996587, 'mask_area': 14269, 'mask_ratio': 0.13795937309652032, 'size': 321.6037935099647}, {'class_id': 5, 'bbox': array([265, 320, 361, 362]), 'bbox_area': 4032, 'wh_ratio': 0.4375, 'mask_area': 3178, 'mask_ratio': 0.7881944444444444, 'size': 63.49803146555017}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00031.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  803,  738, 1022]), 'bbox_area': 91323, 'wh_ratio': 0.5251798561151079, 'mask_area': 62427, 'mask_ratio': 0.6835846391380047, 'size': 302.19695564316993}, {'class_id': 7, 'bbox': array([328,   1, 754, 185]), 'bbox_area': 78384, 'wh_ratio': 0.431924882629108, 'mask_area': 55653, 'mask_ratio': 0.7100045927740355, 'size': 279.9714271135539}, {'class_id': 2, 'bbox': array([325, 226, 462, 278]), 'bbox_area': 7124, 'wh_ratio': 0.3795620437956204, 'mask_area': 3661, 'mask_ratio': 0.5138966872543514, 'size': 84.40379138403678}, {'class_id': 2, 'bbox': array([398,  90, 593, 186]), 'bbox_area': 18720, 'wh_ratio': 0.49230769230769234, 'mask_area': 7967, 'mask_ratio': 0.4255876068376068, 'size': 136.82105101189657}, {'class_id': 2, 'bbox': array([322, 699, 466, 759]), 'bbox_area': 8640, 'wh_ratio': 0.4166666666666667, 'mask_area': 4696, 'mask_ratio': 0.5435185185185185, 'size': 92.951600308978}, {'class_id': 2, 'bbox': array([390, 814, 600, 906]), 'bbox_area': 19320, 'wh_ratio': 0.4380952380952381, 'mask_area': 8645, 'mask_ratio': 0.447463768115942, 'size': 138.99640283115244}, {'class_id': 3, 'bbox': array([306, 666, 671, 796]), 'bbox_area': 47450, 'wh_ratio': 0.3561643835616438, 'mask_area': 5954, 'mask_ratio': 0.12547945205479452, 'size': 217.83020910791967}, {'class_id': 3, 'bbox': array([238, 558, 497, 590]), 'bbox_area': 8288, 'wh_ratio': 0.12355212355212356, 'mask_area': 3763, 'mask_ratio': 0.45402992277992277, 'size': 91.03845341392834}, {'class_id': 3, 'bbox': array([238, 477, 606, 504]), 'bbox_area': 9936, 'wh_ratio': 0.07336956521739131, 'mask_area': 7503, 'mask_ratio': 0.7551328502415459, 'size': 99.6794863550169}, {'class_id': 3, 'bbox': array([241, 386, 534, 427]), 'bbox_area': 12013, 'wh_ratio': 0.13993174061433447, 'mask_area': 4705, 'mask_ratio': 0.39165903604428537, 'size': 109.60383204979651}, {'class_id': 3, 'bbox': array([270, 332, 500, 358]), 'bbox_area': 5980, 'wh_ratio': 0.11304347826086956, 'mask_area': 2409, 'mask_ratio': 0.40284280936454847, 'size': 77.33045971672482}, {'class_id': 3, 'bbox': array([318, 181, 674, 320]), 'bbox_area': 49484, 'wh_ratio': 0.3904494382022472, 'mask_area': 6567, 'mask_ratio': 0.1327095626869291, 'size': 222.44999438075965}, {'class_id': 4, 'bbox': array([259,   4, 543, 307]), 'bbox_area': 86052, 'wh_ratio': 1.0669014084507042, 'mask_area': 13096, 'mask_ratio': 0.1521870496908846, 'size': 293.3462118385032}, {'class_id': 4, 'bbox': array([240, 385, 281, 446]), 'bbox_area': 2501, 'wh_ratio': 1.4878048780487805, 'mask_area': 1194, 'mask_ratio': 0.4774090363854458, 'size': 50.00999900019995}, {'class_id': 4, 'bbox': array([238, 529, 277, 589]), 'bbox_area': 2340, 'wh_ratio': 1.5384615384615385, 'mask_area': 1134, 'mask_ratio': 0.4846153846153846, 'size': 48.373546489791295}, {'class_id': 4, 'bbox': array([ 258,  665,  552, 1023]), 'bbox_area': 105252, 'wh_ratio': 1.217687074829932, 'mask_area': 13690, 'mask_ratio': 0.13006878729145288, 'size': 324.4256463351811}, {'class_id': 5, 'bbox': array([259, 319, 362, 364]), 'bbox_area': 4635, 'wh_ratio': 0.4368932038834951, 'mask_area': 3407, 'mask_ratio': 0.735059331175836, 'size': 68.0808343074613}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00032.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  802,  734, 1022]), 'bbox_area': 90860, 'wh_ratio': 0.5326876513317191, 'mask_area': 62002, 'mask_ratio': 0.6823904908650671, 'size': 301.4299255216708}, {'class_id': 7, 'bbox': array([325,   2, 760, 186]), 'bbox_area': 80040, 'wh_ratio': 0.42298850574712643, 'mask_area': 56676, 'mask_ratio': 0.708095952023988, 'size': 282.9134143161119}, {'class_id': 2, 'bbox': array([326, 228, 461, 276]), 'bbox_area': 6480, 'wh_ratio': 0.35555555555555557, 'mask_area': 3209, 'mask_ratio': 0.4952160493827161, 'size': 80.49844718999243}, {'class_id': 2, 'bbox': array([398,  90, 593, 189]), 'bbox_area': 19305, 'wh_ratio': 0.5076923076923077, 'mask_area': 8258, 'mask_ratio': 0.42776482776482777, 'size': 138.94243412291294}, {'class_id': 2, 'bbox': array([319, 700, 466, 760]), 'bbox_area': 8820, 'wh_ratio': 0.40816326530612246, 'mask_area': 4919, 'mask_ratio': 0.5577097505668934, 'size': 93.91485505499116}, {'class_id': 2, 'bbox': array([387, 813, 601, 904]), 'bbox_area': 19474, 'wh_ratio': 0.4252336448598131, 'mask_area': 9083, 'mask_ratio': 0.4664167608092842, 'size': 139.54927445171472}, {'class_id': 3, 'bbox': array([307, 667, 669, 796]), 'bbox_area': 46698, 'wh_ratio': 0.356353591160221, 'mask_area': 6557, 'mask_ratio': 0.14041286564735106, 'size': 216.0972003520638}, {'class_id': 3, 'bbox': array([238, 555, 509, 590]), 'bbox_area': 9485, 'wh_ratio': 0.12915129151291513, 'mask_area': 4274, 'mask_ratio': 0.45060622034791775, 'size': 97.3909646733207}, {'class_id': 3, 'bbox': array([230, 481, 571, 502]), 'bbox_area': 7161, 'wh_ratio': 0.06158357771260997, 'mask_area': 5023, 'mask_ratio': 0.7014383465996369, 'size': 84.62269199216013}, {'class_id': 3, 'bbox': array([238, 384, 534, 427]), 'bbox_area': 12728, 'wh_ratio': 0.14527027027027026, 'mask_area': 4974, 'mask_ratio': 0.3907919547454431, 'size': 112.81843820936363}, {'class_id': 3, 'bbox': array([270, 331, 502, 358]), 'bbox_area': 6264, 'wh_ratio': 0.11637931034482758, 'mask_area': 2703, 'mask_ratio': 0.4315134099616858, 'size': 79.14543574963753}, {'class_id': 3, 'bbox': array([315, 178, 674, 323]), 'bbox_area': 52055, 'wh_ratio': 0.403899721448468, 'mask_area': 7504, 'mask_ratio': 0.14415522043991932, 'size': 228.15564862610785}, {'class_id': 4, 'bbox': array([254,   4, 542, 310]), 'bbox_area': 88128, 'wh_ratio': 1.0625, 'mask_area': 12499, 'mask_ratio': 0.1418277959331881, 'size': 296.86360504447157}, {'class_id': 4, 'bbox': array([238, 384, 282, 445]), 'bbox_area': 2684, 'wh_ratio': 1.3863636363636365, 'mask_area': 1179, 'mask_ratio': 0.4392697466467958, 'size': 51.807335387954474}, {'class_id': 4, 'bbox': array([238, 529, 274, 589]), 'bbox_area': 2160, 'wh_ratio': 1.6666666666666667, 'mask_area': 1193, 'mask_ratio': 0.5523148148148148, 'size': 46.475800154489}, {'class_id': 4, 'bbox': array([ 254,  665,  550, 1022]), 'bbox_area': 105672, 'wh_ratio': 1.2060810810810811, 'mask_area': 14391, 'mask_ratio': 0.13618555530320237, 'size': 325.07229965040085}, {'class_id': 5, 'bbox': array([262, 321, 357, 366]), 'bbox_area': 4275, 'wh_ratio': 0.47368421052631576, 'mask_area': 3116, 'mask_ratio': 0.7288888888888889, 'size': 65.3834841531101}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00033.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 323,  800,  743, 1022]), 'bbox_area': 93240, 'wh_ratio': 0.5285714285714286, 'mask_area': 63290, 'mask_ratio': 0.6787859287859288, 'size': 305.35225559998736}, {'class_id': 7, 'bbox': array([328,   2, 763, 186]), 'bbox_area': 80040, 'wh_ratio': 0.42298850574712643, 'mask_area': 56315, 'mask_ratio': 0.7035857071464268, 'size': 282.9134143161119}, {'class_id': 2, 'bbox': array([324, 227, 461, 278]), 'bbox_area': 6987, 'wh_ratio': 0.3722627737226277, 'mask_area': 3593, 'mask_ratio': 0.5142407327894661, 'size': 83.58827668997608}, {'class_id': 2, 'bbox': array([397,  90, 594, 187]), 'bbox_area': 19109, 'wh_ratio': 0.49238578680203043, 'mask_area': 8333, 'mask_ratio': 0.4360772410905856, 'size': 138.23530663329103}, {'class_id': 2, 'bbox': array([321, 700, 466, 758]), 'bbox_area': 8410, 'wh_ratio': 0.4, 'mask_area': 4605, 'mask_ratio': 0.5475624256837098, 'size': 91.706052144883}, {'class_id': 2, 'bbox': array([389, 813, 601, 903]), 'bbox_area': 19080, 'wh_ratio': 0.42452830188679247, 'mask_area': 8550, 'mask_ratio': 0.4481132075471698, 'size': 138.13037319865606}, {'class_id': 3, 'bbox': array([307, 670, 670, 798]), 'bbox_area': 46464, 'wh_ratio': 0.3526170798898072, 'mask_area': 6474, 'mask_ratio': 0.13933367768595042, 'size': 215.55509736491967}, {'class_id': 3, 'bbox': array([238, 557, 502, 590]), 'bbox_area': 8712, 'wh_ratio': 0.125, 'mask_area': 4044, 'mask_ratio': 0.4641873278236915, 'size': 93.33809511662427}, {'class_id': 3, 'bbox': array([238, 478, 550, 503]), 'bbox_area': 7800, 'wh_ratio': 0.08012820512820513, 'mask_area': 6158, 'mask_ratio': 0.7894871794871795, 'size': 88.31760866327846}, {'class_id': 3, 'bbox': array([241, 386, 524, 428]), 'bbox_area': 11886, 'wh_ratio': 0.14840989399293286, 'mask_area': 4016, 'mask_ratio': 0.33787649335352515, 'size': 109.02293336725076}, {'class_id': 3, 'bbox': array([272, 334, 498, 359]), 'bbox_area': 5650, 'wh_ratio': 0.11061946902654868, 'mask_area': 2538, 'mask_ratio': 0.44920353982300887, 'size': 75.16648189186454}, {'class_id': 3, 'bbox': array([317, 181, 673, 322]), 'bbox_area': 50196, 'wh_ratio': 0.3960674157303371, 'mask_area': 6805, 'mask_ratio': 0.1355685712008925, 'size': 224.0446384094027}, {'class_id': 4, 'bbox': array([254,   5, 540, 306]), 'bbox_area': 86086, 'wh_ratio': 1.0524475524475525, 'mask_area': 12490, 'mask_ratio': 0.14508747066886601, 'size': 293.40415811641117}, {'class_id': 4, 'bbox': array([240, 387, 282, 444]), 'bbox_area': 2394, 'wh_ratio': 1.3571428571428572, 'mask_area': 1137, 'mask_ratio': 0.474937343358396, 'size': 48.92851929090027}, {'class_id': 4, 'bbox': array([238, 528, 275, 594]), 'bbox_area': 2442, 'wh_ratio': 1.7837837837837838, 'mask_area': 1367, 'mask_ratio': 0.5597870597870598, 'size': 49.416596402423345}, {'class_id': 4, 'bbox': array([ 254,  667,  550, 1022]), 'bbox_area': 105080, 'wh_ratio': 1.1993243243243243, 'mask_area': 13169, 'mask_ratio': 0.12532356299961933, 'size': 324.1604540964243}, {'class_id': 5, 'bbox': array([262, 321, 362, 365]), 'bbox_area': 4400, 'wh_ratio': 0.44, 'mask_area': 3333, 'mask_ratio': 0.7575, 'size': 66.332495807108}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00034.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 322,  802,  738, 1022]), 'bbox_area': 91520, 'wh_ratio': 0.5288461538461539, 'mask_area': 62423, 'mask_ratio': 0.682069493006993, 'size': 302.5227264190246}, {'class_id': 7, 'bbox': array([328,   1, 756, 184]), 'bbox_area': 78324, 'wh_ratio': 0.42757009345794394, 'mask_area': 55781, 'mask_ratio': 0.712182728154844, 'size': 279.8642528083928}, {'class_id': 2, 'bbox': array([325, 227, 467, 278]), 'bbox_area': 7242, 'wh_ratio': 0.3591549295774648, 'mask_area': 3714, 'mask_ratio': 0.5128417564208783, 'size': 85.09994124557313}, {'class_id': 2, 'bbox': array([397,  91, 593, 189]), 'bbox_area': 19208, 'wh_ratio': 0.5, 'mask_area': 8397, 'mask_ratio': 0.437161599333611, 'size': 138.5929291125633}, {'class_id': 2, 'bbox': array([323, 702, 463, 759]), 'bbox_area': 7980, 'wh_ratio': 0.40714285714285714, 'mask_area': 4341, 'mask_ratio': 0.543984962406015, 'size': 89.33084573650918}, {'class_id': 2, 'bbox': array([389, 812, 598, 902]), 'bbox_area': 18810, 'wh_ratio': 0.430622009569378, 'mask_area': 8462, 'mask_ratio': 0.44986709197235514, 'size': 137.14955340794953}, {'class_id': 3, 'bbox': array([306, 666, 670, 798]), 'bbox_area': 48048, 'wh_ratio': 0.3626373626373626, 'mask_area': 7244, 'mask_ratio': 0.15076590076590077, 'size': 219.19854014112414}, {'class_id': 3, 'bbox': array([237, 558, 502, 588]), 'bbox_area': 7950, 'wh_ratio': 0.11320754716981132, 'mask_area': 3951, 'mask_ratio': 0.4969811320754717, 'size': 89.16277250063504}, {'class_id': 3, 'bbox': array([233, 479, 530, 499]), 'bbox_area': 5940, 'wh_ratio': 0.06734006734006734, 'mask_area': 3780, 'mask_ratio': 0.6363636363636364, 'size': 77.07139547199077}, {'class_id': 3, 'bbox': array([240, 386, 506, 424]), 'bbox_area': 10108, 'wh_ratio': 0.14285714285714285, 'mask_area': 3886, 'mask_ratio': 0.38444796201028886, 'size': 100.53854982045445}, {'class_id': 3, 'bbox': array([270, 334, 498, 358]), 'bbox_area': 5472, 'wh_ratio': 0.10526315789473684, 'mask_area': 2748, 'mask_ratio': 0.5021929824561403, 'size': 73.97296803562772}, {'class_id': 3, 'bbox': array([313, 182, 674, 325]), 'bbox_area': 51623, 'wh_ratio': 0.3961218836565097, 'mask_area': 7730, 'mask_ratio': 0.1497394572186816, 'size': 227.20695411892655}, {'class_id': 4, 'bbox': array([259,   3, 542, 309]), 'bbox_area': 86598, 'wh_ratio': 1.0812720848056536, 'mask_area': 13941, 'mask_ratio': 0.16098524215339846, 'size': 294.27538123329316}, {'class_id': 4, 'bbox': array([240, 384, 282, 444]), 'bbox_area': 2520, 'wh_ratio': 1.4285714285714286, 'mask_area': 1324, 'mask_ratio': 0.5253968253968254, 'size': 50.19960159204454}, {'class_id': 4, 'bbox': array([237, 529, 275, 589]), 'bbox_area': 2280, 'wh_ratio': 1.5789473684210527, 'mask_area': 1164, 'mask_ratio': 0.5105263157894737, 'size': 47.74934554525329}, {'class_id': 4, 'bbox': array([ 256,  666,  552, 1022]), 'bbox_area': 105376, 'wh_ratio': 1.2027027027027026, 'mask_area': 15051, 'mask_ratio': 0.14283138475554205, 'size': 324.61669704437577}, {'class_id': 5, 'bbox': array([258, 318, 363, 365]), 'bbox_area': 4935, 'wh_ratio': 0.44761904761904764, 'mask_area': 3700, 'mask_ratio': 0.7497467071935157, 'size': 70.24955515873393}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00035.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 316,  801,  738, 1022]), 'bbox_area': 93262, 'wh_ratio': 0.523696682464455, 'mask_area': 63379, 'mask_ratio': 0.6795801076537068, 'size': 305.3882774436504}, {'class_id': 3, 'bbox': array([305, 666, 667, 798]), 'bbox_area': 47784, 'wh_ratio': 0.36464088397790057, 'mask_area': 8246, 'mask_ratio': 0.17256822367319605, 'size': 218.5955168799214}, {'class_id': 7, 'bbox': array([327,   2, 762, 187]), 'bbox_area': 80475, 'wh_ratio': 0.42528735632183906, 'mask_area': 57203, 'mask_ratio': 0.7108170239204722, 'size': 283.6811590500857}, {'class_id': 3, 'bbox': array([238, 557, 498, 589]), 'bbox_area': 8320, 'wh_ratio': 0.12307692307692308, 'mask_area': 3760, 'mask_ratio': 0.4519230769230769, 'size': 91.21403400793105}, {'class_id': 3, 'bbox': array([232, 479, 512, 498]), 'bbox_area': 5320, 'wh_ratio': 0.06785714285714285, 'mask_area': 4107, 'mask_ratio': 0.7719924812030076, 'size': 72.93833011524187}, {'class_id': 3, 'bbox': array([241, 384, 491, 424]), 'bbox_area': 10000, 'wh_ratio': 0.16, 'mask_area': 3873, 'mask_ratio': 0.3873, 'size': 100.0}, {'class_id': 3, 'bbox': array([272, 331, 500, 359]), 'bbox_area': 6384, 'wh_ratio': 0.12280701754385964, 'mask_area': 2890, 'mask_ratio': 0.4526942355889724, 'size': 79.89993742175272}, {'class_id': 3, 'bbox': array([318, 182, 674, 323]), 'bbox_area': 50196, 'wh_ratio': 0.3960674157303371, 'mask_area': 6994, 'mask_ratio': 0.13933381145908041, 'size': 224.0446384094027}, {'class_id': 2, 'bbox': array([324, 226, 463, 276]), 'bbox_area': 6950, 'wh_ratio': 0.3597122302158273, 'mask_area': 3612, 'mask_ratio': 0.5197122302158274, 'size': 83.36666000266533}, {'class_id': 2, 'bbox': array([396,  86, 593, 189]), 'bbox_area': 20291, 'wh_ratio': 0.5228426395939086, 'mask_area': 8348, 'mask_ratio': 0.4114139273569563, 'size': 142.44648117801998}, {'class_id': 2, 'bbox': array([321, 696, 465, 765]), 'bbox_area': 9936, 'wh_ratio': 0.4791666666666667, 'mask_area': 4965, 'mask_ratio': 0.49969806763285024, 'size': 99.67948635501689}, {'class_id': 2, 'bbox': array([389, 812, 601, 903]), 'bbox_area': 19292, 'wh_ratio': 0.42924528301886794, 'mask_area': 8504, 'mask_ratio': 0.4408044785403276, 'size': 138.8956442801573}, {'class_id': 4, 'bbox': array([261,   3, 542, 309]), 'bbox_area': 85986, 'wh_ratio': 1.0889679715302492, 'mask_area': 12913, 'mask_ratio': 0.15017560998302049, 'size': 293.23369519889764}, {'class_id': 4, 'bbox': array([240, 384, 281, 446]), 'bbox_area': 2542, 'wh_ratio': 1.5121951219512195, 'mask_area': 1157, 'mask_ratio': 0.45515342250196694, 'size': 50.41825066382213}, {'class_id': 4, 'bbox': array([238, 529, 275, 586]), 'bbox_area': 2109, 'wh_ratio': 1.5405405405405406, 'mask_area': 1145, 'mask_ratio': 0.5429113323850165, 'size': 45.92385001282014}, {'class_id': 4, 'bbox': array([ 256,  667,  553, 1022]), 'bbox_area': 105435, 'wh_ratio': 1.1952861952861953, 'mask_area': 13986, 'mask_ratio': 0.1326504481434059, 'size': 324.7075607373502}, {'class_id': 5, 'bbox': array([264, 319, 359, 363]), 'bbox_area': 4180, 'wh_ratio': 0.4631578947368421, 'mask_area': 3103, 'mask_ratio': 0.7423444976076555, 'size': 64.65291950097846}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00036.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 319,  798,  744, 1023]), 'bbox_area': 95625, 'wh_ratio': 0.5294117647058824, 'mask_area': 64487, 'mask_ratio': 0.6743738562091504, 'size': 309.23292192132453}, {'class_id': 7, 'bbox': array([328,   2, 762, 186]), 'bbox_area': 79856, 'wh_ratio': 0.423963133640553, 'mask_area': 56813, 'mask_ratio': 0.7114430975756362, 'size': 282.58803937888104}, {'class_id': 2, 'bbox': array([325, 222, 466, 276]), 'bbox_area': 7614, 'wh_ratio': 0.3829787234042553, 'mask_area': 3673, 'mask_ratio': 0.48240084055686894, 'size': 87.25823743349393}, {'class_id': 2, 'bbox': array([397,  91, 594, 189]), 'bbox_area': 19306, 'wh_ratio': 0.49746192893401014, 'mask_area': 8221, 'mask_ratio': 0.42582616803066403, 'size': 138.94603268895446}, {'class_id': 2, 'bbox': array([321, 698, 466, 759]), 'bbox_area': 8845, 'wh_ratio': 0.4206896551724138, 'mask_area': 4926, 'mask_ratio': 0.5569248162803844, 'size': 94.04786015641186}, {'class_id': 2, 'bbox': array([390, 814, 602, 905]), 'bbox_area': 19292, 'wh_ratio': 0.42924528301886794, 'mask_area': 8975, 'mask_ratio': 0.4652187435206303, 'size': 138.8956442801573}, {'class_id': 3, 'bbox': array([306, 670, 670, 798]), 'bbox_area': 46592, 'wh_ratio': 0.3516483516483517, 'mask_area': 6327, 'mask_ratio': 0.13579584478021978, 'size': 215.85180101171267}, {'class_id': 3, 'bbox': array([243, 480, 498, 499]), 'bbox_area': 4845, 'wh_ratio': 0.07450980392156863, 'mask_area': 3755, 'mask_ratio': 0.7750257997936016, 'size': 69.60603422117941}, {'class_id': 3, 'bbox': array([242, 384, 486, 422]), 'bbox_area': 9272, 'wh_ratio': 0.1557377049180328, 'mask_area': 3017, 'mask_ratio': 0.32538826574633306, 'size': 96.29122493768578}, {'class_id': 3, 'bbox': array([269, 331, 498, 359]), 'bbox_area': 6412, 'wh_ratio': 0.1222707423580786, 'mask_area': 2720, 'mask_ratio': 0.42420461634435436, 'size': 80.0749648766704}, {'class_id': 3, 'bbox': array([318, 180, 673, 318]), 'bbox_area': 48990, 'wh_ratio': 0.38873239436619716, 'mask_area': 6001, 'mask_ratio': 0.12249438660951215, 'size': 221.33684736166276}, {'class_id': 4, 'bbox': array([258,   4, 540, 309]), 'bbox_area': 86010, 'wh_ratio': 1.0815602836879432, 'mask_area': 13657, 'mask_ratio': 0.1587838623415882, 'size': 293.2746153351838}, {'class_id': 4, 'bbox': array([240, 384, 282, 445]), 'bbox_area': 2562, 'wh_ratio': 1.4523809523809523, 'mask_area': 1221, 'mask_ratio': 0.4765807962529274, 'size': 50.61620293937505}, {'class_id': 4, 'bbox': array([238, 530, 274, 589]), 'bbox_area': 2124, 'wh_ratio': 1.6388888888888888, 'mask_area': 988, 'mask_ratio': 0.4651600753295669, 'size': 46.086874487211645}, {'class_id': 4, 'bbox': array([ 256,  666,  550, 1023]), 'bbox_area': 104958, 'wh_ratio': 1.2142857142857142, 'mask_area': 14475, 'mask_ratio': 0.13791230778025496, 'size': 323.9722210313718}, {'class_id': 5, 'bbox': array([258, 318, 362, 366]), 'bbox_area': 4992, 'wh_ratio': 0.46153846153846156, 'mask_area': 3525, 'mask_ratio': 0.7061298076923077, 'size': 70.65408693062278}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00037.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 322,  803,  738, 1019]), 'bbox_area': 89856, 'wh_ratio': 0.5192307692307693, 'mask_area': 61227, 'mask_ratio': 0.6813902243589743, 'size': 299.7599039231231}, {'class_id': 7, 'bbox': array([328,   2, 756, 184]), 'bbox_area': 77896, 'wh_ratio': 0.4252336448598131, 'mask_area': 56313, 'mask_ratio': 0.7229254390469344, 'size': 279.09854890342945}, {'class_id': 2, 'bbox': array([323, 226, 465, 276]), 'bbox_area': 7100, 'wh_ratio': 0.352112676056338, 'mask_area': 3877, 'mask_ratio': 0.546056338028169, 'size': 84.26149773176358}, {'class_id': 2, 'bbox': array([394,  91, 590, 190]), 'bbox_area': 19404, 'wh_ratio': 0.5051020408163265, 'mask_area': 8970, 'mask_ratio': 0.46227581941867657, 'size': 139.2982411949268}, {'class_id': 2, 'bbox': array([321, 698, 466, 760]), 'bbox_area': 8990, 'wh_ratio': 0.42758620689655175, 'mask_area': 4688, 'mask_ratio': 0.52146829810901, 'size': 94.81561052906848}, {'class_id': 2, 'bbox': array([387, 814, 602, 903]), 'bbox_area': 19135, 'wh_ratio': 0.413953488372093, 'mask_area': 8961, 'mask_ratio': 0.46830415469035797, 'size': 138.32931721077784}, {'class_id': 3, 'bbox': array([306, 670, 666, 796]), 'bbox_area': 45360, 'wh_ratio': 0.35, 'mask_area': 6215, 'mask_ratio': 0.13701499118165786, 'size': 212.9788721915862}, {'class_id': 3, 'bbox': array([238, 479, 486, 498]), 'bbox_area': 4712, 'wh_ratio': 0.07661290322580645, 'mask_area': 3925, 'mask_ratio': 0.8329796264855688, 'size': 68.64400920692205}, {'class_id': 3, 'bbox': array([241, 385, 477, 422]), 'bbox_area': 8732, 'wh_ratio': 0.15677966101694915, 'mask_area': 3608, 'mask_ratio': 0.41319285387082, 'size': 93.44517108978934}, {'class_id': 3, 'bbox': array([270, 332, 501, 358]), 'bbox_area': 6006, 'wh_ratio': 0.11255411255411256, 'mask_area': 2900, 'mask_ratio': 0.4828504828504829, 'size': 77.49838707999025}, {'class_id': 3, 'bbox': array([319, 181, 671, 320]), 'bbox_area': 48928, 'wh_ratio': 0.39488636363636365, 'mask_area': 6269, 'mask_ratio': 0.12812704381948986, 'size': 221.19674500317583}, {'class_id': 4, 'bbox': array([259,   4, 542, 307]), 'bbox_area': 85749, 'wh_ratio': 1.0706713780918728, 'mask_area': 14401, 'mask_ratio': 0.16794364948862378, 'size': 292.82930181250646}, {'class_id': 4, 'bbox': array([240, 385, 282, 446]), 'bbox_area': 2562, 'wh_ratio': 1.4523809523809523, 'mask_area': 1262, 'mask_ratio': 0.492583918813427, 'size': 50.61620293937505}, {'class_id': 4, 'bbox': array([237, 527, 275, 590]), 'bbox_area': 2394, 'wh_ratio': 1.6578947368421053, 'mask_area': 1240, 'mask_ratio': 0.5179615705931495, 'size': 48.92851929090027}, {'class_id': 4, 'bbox': array([ 259,  667,  552, 1022]), 'bbox_area': 104015, 'wh_ratio': 1.21160409556314, 'mask_area': 14271, 'mask_ratio': 0.1372013651877133, 'size': 322.51356560616176}, {'class_id': 5, 'bbox': array([260, 319, 361, 365]), 'bbox_area': 4646, 'wh_ratio': 0.45544554455445546, 'mask_area': 3523, 'mask_ratio': 0.7582866982350409, 'size': 68.1615727518079}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00038.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  802,  738, 1022]), 'bbox_area': 91740, 'wh_ratio': 0.5275779376498801, 'mask_area': 62971, 'mask_ratio': 0.6864072378460868, 'size': 302.88611721239386}, {'class_id': 7, 'bbox': array([330,   2, 756, 186]), 'bbox_area': 78384, 'wh_ratio': 0.431924882629108, 'mask_area': 55232, 'mask_ratio': 0.704633598693611, 'size': 279.9714271135539}, {'class_id': 2, 'bbox': array([325, 228, 462, 276]), 'bbox_area': 6576, 'wh_ratio': 0.35036496350364965, 'mask_area': 3426, 'mask_ratio': 0.5209854014598541, 'size': 81.09253973085318}, {'class_id': 2, 'bbox': array([398,  91, 593, 186]), 'bbox_area': 18525, 'wh_ratio': 0.48717948717948717, 'mask_area': 8022, 'mask_ratio': 0.43303643724696356, 'size': 136.10657588816198}, {'class_id': 2, 'bbox': array([321, 703, 466, 762]), 'bbox_area': 8555, 'wh_ratio': 0.4068965517241379, 'mask_area': 4629, 'mask_ratio': 0.5410870835768556, 'size': 92.49324299644812}, {'class_id': 2, 'bbox': array([390, 812, 595, 905]), 'bbox_area': 19065, 'wh_ratio': 0.45365853658536587, 'mask_area': 8631, 'mask_ratio': 0.45271439811172304, 'size': 138.07606599262596}, {'class_id': 3, 'bbox': array([306, 669, 668, 794]), 'bbox_area': 45250, 'wh_ratio': 0.3453038674033149, 'mask_area': 5946, 'mask_ratio': 0.13140331491712706, 'size': 212.72047386182646}, {'class_id': 3, 'bbox': array([243, 479, 472, 498]), 'bbox_area': 4351, 'wh_ratio': 0.08296943231441048, 'mask_area': 3904, 'mask_ratio': 0.8972649965525167, 'size': 65.96211033616193}, {'class_id': 3, 'bbox': array([242, 386, 474, 422]), 'bbox_area': 8352, 'wh_ratio': 0.15517241379310345, 'mask_area': 3352, 'mask_ratio': 0.4013409961685824, 'size': 91.38927727036689}, {'class_id': 3, 'bbox': array([270, 331, 498, 358]), 'bbox_area': 6156, 'wh_ratio': 0.11842105263157894, 'mask_area': 2804, 'mask_ratio': 0.4554905782975958, 'size': 78.46018098373213}, {'class_id': 3, 'bbox': array([321, 178, 673, 320]), 'bbox_area': 49984, 'wh_ratio': 0.4034090909090909, 'mask_area': 5858, 'mask_ratio': 0.11719750320102433, 'size': 223.57101779971393}, {'class_id': 4, 'bbox': array([255,   4, 542, 309]), 'bbox_area': 87535, 'wh_ratio': 1.0627177700348431, 'mask_area': 13887, 'mask_ratio': 0.15864511338321813, 'size': 295.86314403791494}, {'class_id': 4, 'bbox': array([241, 387, 281, 446]), 'bbox_area': 2360, 'wh_ratio': 1.475, 'mask_area': 1094, 'mask_ratio': 0.4635593220338983, 'size': 48.57983120596448}, {'class_id': 4, 'bbox': array([238, 528, 274, 589]), 'bbox_area': 2196, 'wh_ratio': 1.6944444444444444, 'mask_area': 1060, 'mask_ratio': 0.48269581056466304, 'size': 46.86149805543993}, {'class_id': 4, 'bbox': array([ 253,  666,  552, 1022]), 'bbox_area': 106444, 'wh_ratio': 1.1906354515050168, 'mask_area': 14004, 'mask_ratio': 0.13156213595881402, 'size': 326.2575669620553}, {'class_id': 5, 'bbox': array([258, 319, 365, 366]), 'bbox_area': 5029, 'wh_ratio': 0.4392523364485981, 'mask_area': 3452, 'mask_ratio': 0.6864187711274607, 'size': 70.9154426059656}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00039.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  801,  738, 1024]), 'bbox_area': 92991, 'wh_ratio': 0.5347721822541966, 'mask_area': 63302, 'mask_ratio': 0.6807325440096353, 'size': 304.94425720121376}, {'class_id': 7, 'bbox': array([329,   2, 757, 186]), 'bbox_area': 78752, 'wh_ratio': 0.42990654205607476, 'mask_area': 55989, 'mask_ratio': 0.710953372612759, 'size': 280.6278674686461}, {'class_id': 2, 'bbox': array([323, 227, 462, 276]), 'bbox_area': 6811, 'wh_ratio': 0.35251798561151076, 'mask_area': 3579, 'mask_ratio': 0.5254734987520188, 'size': 82.52878285786119}, {'class_id': 2, 'bbox': array([397,  93, 593, 187]), 'bbox_area': 18424, 'wh_ratio': 0.47959183673469385, 'mask_area': 7767, 'mask_ratio': 0.4215696917064698, 'size': 135.73503600765721}, {'class_id': 2, 'bbox': array([322, 699, 465, 762]), 'bbox_area': 9009, 'wh_ratio': 0.4405594405594406, 'mask_area': 4886, 'mask_ratio': 0.5423465423465423, 'size': 94.91575211733826}, {'class_id': 2, 'bbox': array([389, 814, 599, 902]), 'bbox_area': 18480, 'wh_ratio': 0.41904761904761906, 'mask_area': 8715, 'mask_ratio': 0.4715909090909091, 'size': 135.94116374373144}, {'class_id': 3, 'bbox': array([306, 669, 664, 797]), 'bbox_area': 45824, 'wh_ratio': 0.3575418994413408, 'mask_area': 7321, 'mask_ratio': 0.15976344273743018, 'size': 214.06541056415443}, {'class_id': 3, 'bbox': array([241, 478, 456, 498]), 'bbox_area': 4300, 'wh_ratio': 0.09302325581395349, 'mask_area': 3539, 'mask_ratio': 0.8230232558139535, 'size': 65.57438524302}, {'class_id': 3, 'bbox': array([241, 386, 477, 420]), 'bbox_area': 8024, 'wh_ratio': 0.1440677966101695, 'mask_area': 2906, 'mask_ratio': 0.36216350947158527, 'size': 89.57678270623477}, {'class_id': 3, 'bbox': array([270, 332, 498, 358]), 'bbox_area': 5928, 'wh_ratio': 0.11403508771929824, 'mask_area': 2503, 'mask_ratio': 0.42223346828609987, 'size': 76.99350621968063}, {'class_id': 3, 'bbox': array([318, 179, 673, 319]), 'bbox_area': 49700, 'wh_ratio': 0.39436619718309857, 'mask_area': 6150, 'mask_ratio': 0.12374245472837023, 'size': 222.93496809607953}, {'class_id': 4, 'bbox': array([254,   2, 540, 309]), 'bbox_area': 87802, 'wh_ratio': 1.0734265734265733, 'mask_area': 13892, 'mask_ratio': 0.15821963053233412, 'size': 296.31402261789776}, {'class_id': 4, 'bbox': array([240, 382, 280, 446]), 'bbox_area': 2560, 'wh_ratio': 1.6, 'mask_area': 1181, 'mask_ratio': 0.461328125, 'size': 50.596442562694065}, {'class_id': 4, 'bbox': array([239, 529, 275, 588]), 'bbox_area': 2124, 'wh_ratio': 1.6388888888888888, 'mask_area': 1132, 'mask_ratio': 0.5329566854990584, 'size': 46.086874487211645}, {'class_id': 4, 'bbox': array([ 256,  666,  553, 1023]), 'bbox_area': 106029, 'wh_ratio': 1.202020202020202, 'mask_area': 15315, 'mask_ratio': 0.1444416150298503, 'size': 325.62094527225975}, {'class_id': 5, 'bbox': array([260, 318, 358, 368]), 'bbox_area': 4900, 'wh_ratio': 0.5102040816326531, 'mask_area': 3412, 'mask_ratio': 0.6963265306122449, 'size': 70.0}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00040.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 318,  799,  741, 1022]), 'bbox_area': 94329, 'wh_ratio': 0.5271867612293144, 'mask_area': 64274, 'mask_ratio': 0.6813811235145077, 'size': 307.13026552262806}, {'class_id': 7, 'bbox': array([325,   3, 756, 186]), 'bbox_area': 78873, 'wh_ratio': 0.4245939675174014, 'mask_area': 56180, 'mask_ratio': 0.7122843051487835, 'size': 280.8433727186739}, {'class_id': 2, 'bbox': array([323, 226, 462, 277]), 'bbox_area': 7089, 'wh_ratio': 0.3669064748201439, 'mask_area': 3806, 'mask_ratio': 0.5368881365495839, 'size': 84.19619943916709}, {'class_id': 2, 'bbox': array([397,  88, 590, 191]), 'bbox_area': 19879, 'wh_ratio': 0.533678756476684, 'mask_area': 8485, 'mask_ratio': 0.42683233563056494, 'size': 140.99290762304324}, {'class_id': 2, 'bbox': array([389, 809, 601, 902]), 'bbox_area': 19716, 'wh_ratio': 0.4386792452830189, 'mask_area': 8776, 'mask_ratio': 0.44512071414079935, 'size': 140.4136745477448}, {'class_id': 2, 'bbox': array([321, 700, 466, 759]), 'bbox_area': 8555, 'wh_ratio': 0.4068965517241379, 'mask_area': 4830, 'mask_ratio': 0.5645821157218002, 'size': 92.49324299644812}, {'class_id': 3, 'bbox': array([241, 480, 442, 498]), 'bbox_area': 3618, 'wh_ratio': 0.08955223880597014, 'mask_area': 3175, 'mask_ratio': 0.8775566611387506, 'size': 60.14981296729026}, {'class_id': 3, 'bbox': array([242, 390, 486, 422]), 'bbox_area': 7808, 'wh_ratio': 0.13114754098360656, 'mask_area': 2927, 'mask_ratio': 0.3748719262295082, 'size': 88.36288813749809}, {'class_id': 3, 'bbox': array([270, 332, 498, 361]), 'bbox_area': 6612, 'wh_ratio': 0.12719298245614036, 'mask_area': 3021, 'mask_ratio': 0.45689655172413796, 'size': 81.31420540102448}, {'class_id': 3, 'bbox': array([317, 181, 673, 320]), 'bbox_area': 49484, 'wh_ratio': 0.3904494382022472, 'mask_area': 6555, 'mask_ratio': 0.13246706005981732, 'size': 222.44999438075965}, {'class_id': 4, 'bbox': array([258,   5, 542, 312]), 'bbox_area': 87188, 'wh_ratio': 1.0809859154929577, 'mask_area': 13798, 'mask_ratio': 0.15825572326466944, 'size': 295.2761419417424}, {'class_id': 4, 'bbox': array([240, 384, 282, 446]), 'bbox_area': 2604, 'wh_ratio': 1.4761904761904763, 'mask_area': 1217, 'mask_ratio': 0.467357910906298, 'size': 51.0294032886923}, {'class_id': 4, 'bbox': array([237, 528, 274, 589]), 'bbox_area': 2257, 'wh_ratio': 1.6486486486486487, 'mask_area': 1029, 'mask_ratio': 0.4559149313247674, 'size': 47.507894080878806}, {'class_id': 4, 'bbox': array([ 258,  666,  551, 1022]), 'bbox_area': 104308, 'wh_ratio': 1.2150170648464165, 'mask_area': 14496, 'mask_ratio': 0.13897304137745906, 'size': 322.9674906240565}, {'class_id': 5, 'bbox': array([262, 321, 358, 366]), 'bbox_area': 4320, 'wh_ratio': 0.46875, 'mask_area': 3367, 'mask_ratio': 0.7793981481481481, 'size': 65.72670690061993}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00041.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 318,  801,  741, 1023]), 'bbox_area': 93906, 'wh_ratio': 0.524822695035461, 'mask_area': 64480, 'mask_ratio': 0.68664409089941, 'size': 306.44085889450184}, {'class_id': 7, 'bbox': array([329,   2, 761, 186]), 'bbox_area': 79488, 'wh_ratio': 0.42592592592592593, 'mask_area': 56484, 'mask_ratio': 0.7105978260869565, 'size': 281.93616298729756}, {'class_id': 2, 'bbox': array([325, 226, 466, 276]), 'bbox_area': 7050, 'wh_ratio': 0.3546099290780142, 'mask_area': 3788, 'mask_ratio': 0.5373049645390071, 'size': 83.96427811873332}, {'class_id': 2, 'bbox': array([398,  91, 593, 189]), 'bbox_area': 19110, 'wh_ratio': 0.5025641025641026, 'mask_area': 7967, 'mask_ratio': 0.41690214547357407, 'size': 138.2389236069205}, {'class_id': 2, 'bbox': array([323, 702, 466, 758]), 'bbox_area': 8008, 'wh_ratio': 0.3916083916083916, 'mask_area': 4581, 'mask_ratio': 0.5720529470529471, 'size': 89.48742928478838}, {'class_id': 2, 'bbox': array([391, 814, 601, 904]), 'bbox_area': 18900, 'wh_ratio': 0.42857142857142855, 'mask_area': 8971, 'mask_ratio': 0.47465608465608466, 'size': 137.4772708486752}, {'class_id': 3, 'bbox': array([308, 670, 665, 796]), 'bbox_area': 44982, 'wh_ratio': 0.35294117647058826, 'mask_area': 5367, 'mask_ratio': 0.11931439242363612, 'size': 212.08960370560365}, {'class_id': 3, 'bbox': array([244, 478, 428, 497]), 'bbox_area': 3496, 'wh_ratio': 0.10326086956521739, 'mask_area': 2853, 'mask_ratio': 0.8160755148741419, 'size': 59.12698199637793}, {'class_id': 3, 'bbox': array([241, 385, 500, 424]), 'bbox_area': 10101, 'wh_ratio': 0.15057915057915058, 'mask_area': 3658, 'mask_ratio': 0.3621423621423621, 'size': 100.50373127401788}, {'class_id': 3, 'bbox': array([270, 333, 500, 359]), 'bbox_area': 5980, 'wh_ratio': 0.11304347826086956, 'mask_area': 3117, 'mask_ratio': 0.5212374581939799, 'size': 77.33045971672482}, {'class_id': 3, 'bbox': array([318, 178, 671, 319]), 'bbox_area': 49773, 'wh_ratio': 0.39943342776203966, 'mask_area': 6716, 'mask_ratio': 0.134932593976654, 'size': 223.09863289585618}, {'class_id': 4, 'bbox': array([256,   4, 542, 309]), 'bbox_area': 87230, 'wh_ratio': 1.0664335664335665, 'mask_area': 14340, 'mask_ratio': 0.1643929840651152, 'size': 295.347253246073}, {'class_id': 4, 'bbox': array([241, 386, 280, 445]), 'bbox_area': 2301, 'wh_ratio': 1.5128205128205128, 'mask_area': 1163, 'mask_ratio': 0.505432420686658, 'size': 47.968739820845826}, {'class_id': 4, 'bbox': array([238, 528, 274, 588]), 'bbox_area': 2160, 'wh_ratio': 1.6666666666666667, 'mask_area': 1064, 'mask_ratio': 0.4925925925925926, 'size': 46.475800154489}, {'class_id': 4, 'bbox': array([ 255,  664,  552, 1022]), 'bbox_area': 106326, 'wh_ratio': 1.2053872053872055, 'mask_area': 15364, 'mask_ratio': 0.14449899366100483, 'size': 326.0766780988791}, {'class_id': 5, 'bbox': array([258, 318, 357, 363]), 'bbox_area': 4455, 'wh_ratio': 0.45454545454545453, 'mask_area': 3268, 'mask_ratio': 0.7335578002244669, 'size': 66.74578638386097}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00042.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  800,  740, 1022]), 'bbox_area': 93018, 'wh_ratio': 0.5298329355608592, 'mask_area': 63445, 'mask_ratio': 0.6820722870842203, 'size': 304.9885243742787}, {'class_id': 7, 'bbox': array([327,   2, 761, 186]), 'bbox_area': 79856, 'wh_ratio': 0.423963133640553, 'mask_area': 56691, 'mask_ratio': 0.7099153476257263, 'size': 282.58803937888104}, {'class_id': 2, 'bbox': array([322, 225, 463, 276]), 'bbox_area': 7191, 'wh_ratio': 0.3617021276595745, 'mask_area': 3850, 'mask_ratio': 0.5353914615491586, 'size': 84.79976415061542}, {'class_id': 2, 'bbox': array([396,  89, 593, 190]), 'bbox_area': 19897, 'wh_ratio': 0.5126903553299492, 'mask_area': 8439, 'mask_ratio': 0.424134291601749, 'size': 141.05672617780408}, {'class_id': 2, 'bbox': array([319, 700, 474, 762]), 'bbox_area': 9610, 'wh_ratio': 0.4, 'mask_area': 5181, 'mask_ratio': 0.5391259105098856, 'size': 98.03060746521976}, {'class_id': 2, 'bbox': array([389, 812, 604, 905]), 'bbox_area': 19995, 'wh_ratio': 0.4325581395348837, 'mask_area': 9019, 'mask_ratio': 0.45106276569142284, 'size': 141.40367746278736}, {'class_id': 3, 'bbox': array([310, 666, 666, 798]), 'bbox_area': 46992, 'wh_ratio': 0.3707865168539326, 'mask_area': 7482, 'mask_ratio': 0.15921859039836567, 'size': 216.77638247742763}, {'class_id': 3, 'bbox': array([238, 478, 413, 497]), 'bbox_area': 3325, 'wh_ratio': 0.10857142857142857, 'mask_area': 3034, 'mask_ratio': 0.9124812030075188, 'size': 57.66281297335398}, {'class_id': 3, 'bbox': array([242, 386, 518, 425]), 'bbox_area': 10764, 'wh_ratio': 0.14130434782608695, 'mask_area': 3945, 'mask_ratio': 0.3664994425863991, 'size': 103.7496987947435}, {'class_id': 3, 'bbox': array([270, 331, 500, 358]), 'bbox_area': 6210, 'wh_ratio': 0.11739130434782609, 'mask_area': 2777, 'mask_ratio': 0.4471819645732689, 'size': 78.80355321938218}, {'class_id': 3, 'bbox': array([317, 179, 673, 322]), 'bbox_area': 50908, 'wh_ratio': 0.40168539325842695, 'mask_area': 7148, 'mask_ratio': 0.1404101516461067, 'size': 225.62801244526352}, {'class_id': 4, 'bbox': array([255,   2, 544, 310]), 'bbox_area': 89012, 'wh_ratio': 1.065743944636678, 'mask_area': 14165, 'mask_ratio': 0.15913584685210982, 'size': 298.34878917133216}, {'class_id': 4, 'bbox': array([240, 383, 281, 446]), 'bbox_area': 2583, 'wh_ratio': 1.5365853658536586, 'mask_area': 1188, 'mask_ratio': 0.45993031358885017, 'size': 50.823223038292255}, {'class_id': 4, 'bbox': array([237, 528, 274, 591]), 'bbox_area': 2331, 'wh_ratio': 1.7027027027027026, 'mask_area': 1143, 'mask_ratio': 0.49034749034749037, 'size': 48.28043081829325}, {'class_id': 4, 'bbox': array([ 255,  662,  551, 1022]), 'bbox_area': 106560, 'wh_ratio': 1.2162162162162162, 'mask_area': 14950, 'mask_ratio': 0.14029654654654655, 'size': 326.43529220965064}, {'class_id': 5, 'bbox': array([261, 318, 359, 364]), 'bbox_area': 4508, 'wh_ratio': 0.46938775510204084, 'mask_area': 3547, 'mask_ratio': 0.7868234250221828, 'size': 67.14164132637806}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00043.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 322,  803,  738, 1022]), 'bbox_area': 91104, 'wh_ratio': 0.5264423076923077, 'mask_area': 62647, 'mask_ratio': 0.687642694063927, 'size': 301.8343916786157}, {'class_id': 7, 'bbox': array([327,   2, 755, 186]), 'bbox_area': 78752, 'wh_ratio': 0.42990654205607476, 'mask_area': 56338, 'mask_ratio': 0.7153850060950833, 'size': 280.6278674686461}, {'class_id': 2, 'bbox': array([325, 228, 465, 274]), 'bbox_area': 6440, 'wh_ratio': 0.32857142857142857, 'mask_area': 3491, 'mask_ratio': 0.5420807453416149, 'size': 80.24961059095551}, {'class_id': 2, 'bbox': array([398,  90, 595, 190]), 'bbox_area': 19700, 'wh_ratio': 0.5076142131979695, 'mask_area': 8500, 'mask_ratio': 0.43147208121827413, 'size': 140.356688476182}, {'class_id': 2, 'bbox': array([322, 700, 466, 760]), 'bbox_area': 8640, 'wh_ratio': 0.4166666666666667, 'mask_area': 4670, 'mask_ratio': 0.5405092592592593, 'size': 92.951600308978}, {'class_id': 2, 'bbox': array([388, 811, 598, 907]), 'bbox_area': 20160, 'wh_ratio': 0.45714285714285713, 'mask_area': 8921, 'mask_ratio': 0.4425099206349206, 'size': 141.9859147943908}, {'class_id': 3, 'bbox': array([310, 663, 670, 797]), 'bbox_area': 48240, 'wh_ratio': 0.37222222222222223, 'mask_area': 7864, 'mask_ratio': 0.16301824212271973, 'size': 219.63606261267753}, {'class_id': 3, 'bbox': array([263, 618, 428, 644]), 'bbox_area': 4290, 'wh_ratio': 0.15757575757575756, 'mask_area': 2102, 'mask_ratio': 0.48997668997668997, 'size': 65.49809157525127}, {'class_id': 3, 'bbox': array([241, 478, 402, 496]), 'bbox_area': 2898, 'wh_ratio': 0.11180124223602485, 'mask_area': 2718, 'mask_ratio': 0.937888198757764, 'size': 53.83307533477908}, {'class_id': 3, 'bbox': array([241, 386, 520, 424]), 'bbox_area': 10602, 'wh_ratio': 0.13620071684587814, 'mask_area': 3842, 'mask_ratio': 0.36238445576306355, 'size': 102.96601381038309}, {'class_id': 3, 'bbox': array([272, 332, 500, 359]), 'bbox_area': 6156, 'wh_ratio': 0.11842105263157894, 'mask_area': 2908, 'mask_ratio': 0.4723846653671215, 'size': 78.46018098373213}, {'class_id': 3, 'bbox': array([318, 179, 674, 322]), 'bbox_area': 50908, 'wh_ratio': 0.40168539325842695, 'mask_area': 7227, 'mask_ratio': 0.141961970613656, 'size': 225.62801244526352}, {'class_id': 4, 'bbox': array([259,   6, 542, 313]), 'bbox_area': 86881, 'wh_ratio': 1.0848056537102473, 'mask_area': 15077, 'mask_ratio': 0.17353621620377296, 'size': 294.75583115521226}, {'class_id': 4, 'bbox': array([240, 386, 278, 446]), 'bbox_area': 2280, 'wh_ratio': 1.5789473684210527, 'mask_area': 1243, 'mask_ratio': 0.5451754385964912, 'size': 47.74934554525329}, {'class_id': 4, 'bbox': array([237, 529, 274, 590]), 'bbox_area': 2257, 'wh_ratio': 1.6486486486486487, 'mask_area': 1235, 'mask_ratio': 0.5471865307930882, 'size': 47.507894080878806}, {'class_id': 4, 'bbox': array([ 256,  666,  550, 1023]), 'bbox_area': 104958, 'wh_ratio': 1.2142857142857142, 'mask_area': 15510, 'mask_ratio': 0.14777339507231463, 'size': 323.9722210313718}, {'class_id': 5, 'bbox': array([260, 319, 357, 369]), 'bbox_area': 4850, 'wh_ratio': 0.5154639175257731, 'mask_area': 3542, 'mask_ratio': 0.7303092783505155, 'size': 69.6419413859206}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00044.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 318,  801,  738, 1023]), 'bbox_area': 93240, 'wh_ratio': 0.5285714285714286, 'mask_area': 63990, 'mask_ratio': 0.6862934362934363, 'size': 305.35225559998736}, {'class_id': 7, 'bbox': array([328,   1, 760, 187]), 'bbox_area': 80352, 'wh_ratio': 0.4305555555555556, 'mask_area': 56697, 'mask_ratio': 0.705607825567503, 'size': 283.4642834644252}, {'class_id': 2, 'bbox': array([325, 228, 465, 276]), 'bbox_area': 6720, 'wh_ratio': 0.34285714285714286, 'mask_area': 3626, 'mask_ratio': 0.5395833333333333, 'size': 81.97560612767678}, {'class_id': 2, 'bbox': array([398,  88, 592, 190]), 'bbox_area': 19788, 'wh_ratio': 0.5257731958762887, 'mask_area': 8840, 'mask_ratio': 0.44673539518900346, 'size': 140.6698261888455}, {'class_id': 2, 'bbox': array([320, 700, 466, 762]), 'bbox_area': 9052, 'wh_ratio': 0.4246575342465753, 'mask_area': 4846, 'mask_ratio': 0.5353513035793195, 'size': 95.14199913813037}, {'class_id': 2, 'bbox': array([389, 811, 601, 904]), 'bbox_area': 19716, 'wh_ratio': 0.4386792452830189, 'mask_area': 8663, 'mask_ratio': 0.43938932846419154, 'size': 140.4136745477448}, {'class_id': 3, 'bbox': array([302, 671, 668, 796]), 'bbox_area': 45750, 'wh_ratio': 0.34153005464480873, 'mask_area': 6032, 'mask_ratio': 0.13184699453551912, 'size': 213.8924963620744}, {'class_id': 3, 'bbox': array([264, 618, 431, 644]), 'bbox_area': 4342, 'wh_ratio': 0.15568862275449102, 'mask_area': 1916, 'mask_ratio': 0.44127130354675265, 'size': 65.89385403814228}, {'class_id': 3, 'bbox': array([240, 478, 391, 496]), 'bbox_area': 2718, 'wh_ratio': 0.11920529801324503, 'mask_area': 2324, 'mask_ratio': 0.8550404709345106, 'size': 52.134441590948306}, {'class_id': 3, 'bbox': array([240, 384, 507, 424]), 'bbox_area': 10680, 'wh_ratio': 0.149812734082397, 'mask_area': 4007, 'mask_ratio': 0.375187265917603, 'size': 103.34408546211051}, {'class_id': 3, 'bbox': array([270, 332, 496, 358]), 'bbox_area': 5876, 'wh_ratio': 0.11504424778761062, 'mask_area': 2479, 'mask_ratio': 0.4218856364874064, 'size': 76.6550715869472}, {'class_id': 3, 'bbox': array([317, 181, 673, 322]), 'bbox_area': 50196, 'wh_ratio': 0.3960674157303371, 'mask_area': 6344, 'mask_ratio': 0.1263845724758945, 'size': 224.0446384094027}, {'class_id': 4, 'bbox': array([258,   3, 544, 307]), 'bbox_area': 86944, 'wh_ratio': 1.062937062937063, 'mask_area': 13200, 'mask_ratio': 0.15182186234817813, 'size': 294.86267990371385}, {'class_id': 4, 'bbox': array([240, 384, 281, 446]), 'bbox_area': 2542, 'wh_ratio': 1.5121951219512195, 'mask_area': 1220, 'mask_ratio': 0.47993705743509046, 'size': 50.41825066382213}, {'class_id': 4, 'bbox': array([236, 528, 274, 589]), 'bbox_area': 2318, 'wh_ratio': 1.605263157894737, 'mask_area': 1183, 'mask_ratio': 0.5103537532355479, 'size': 48.14561246884289}, {'class_id': 4, 'bbox': array([ 258,  666,  554, 1023]), 'bbox_area': 105672, 'wh_ratio': 1.2060810810810811, 'mask_area': 14976, 'mask_ratio': 0.14172155348625937, 'size': 325.07229965040085}, {'class_id': 5, 'bbox': array([260, 319, 346, 365]), 'bbox_area': 3956, 'wh_ratio': 0.5348837209302325, 'mask_area': 3172, 'mask_ratio': 0.8018200202224469, 'size': 62.896740774065556}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00045.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  801,  737, 1022]), 'bbox_area': 91936, 'wh_ratio': 0.53125, 'mask_area': 64045, 'mask_ratio': 0.6966259136790811, 'size': 303.2094985319556}, {'class_id': 7, 'bbox': array([307, 667, 669, 797]), 'bbox_area': 47060, 'wh_ratio': 0.35911602209944754, 'mask_area': 6836, 'mask_ratio': 0.14526136846578835, 'size': 216.9331694324314}, {'class_id': 3, 'bbox': array([262, 618, 483, 644]), 'bbox_area': 5746, 'wh_ratio': 0.11764705882352941, 'mask_area': 2916, 'mask_ratio': 0.5074834667594849, 'size': 75.8023746329889}, {'class_id': 3, 'bbox': array([238, 479, 384, 496]), 'bbox_area': 2482, 'wh_ratio': 0.11643835616438356, 'mask_area': 2296, 'mask_ratio': 0.9250604351329573, 'size': 49.8196748283246}, {'class_id': 3, 'bbox': array([240, 385, 498, 424]), 'bbox_area': 10062, 'wh_ratio': 0.1511627906976744, 'mask_area': 4112, 'mask_ratio': 0.40866626913138543, 'size': 100.30952098380293}, {'class_id': 3, 'bbox': array([271, 331, 502, 359]), 'bbox_area': 6468, 'wh_ratio': 0.12121212121212122, 'mask_area': 2717, 'mask_ratio': 0.4200680272108844, 'size': 80.4238770515324}, {'class_id': 3, 'bbox': array([316, 178, 673, 320]), 'bbox_area': 50694, 'wh_ratio': 0.39775910364145656, 'mask_area': 6332, 'mask_ratio': 0.12490630054838837, 'size': 225.15328112199475}, {'class_id': 2, 'bbox': array([394,  85, 592, 187]), 'bbox_area': 20196, 'wh_ratio': 0.5151515151515151, 'mask_area': 7896, 'mask_ratio': 0.3909685086155674, 'size': 142.11263138792413}, {'class_id': 2, 'bbox': array([325, 226, 462, 278]), 'bbox_area': 7124, 'wh_ratio': 0.3795620437956204, 'mask_area': 3465, 'mask_ratio': 0.4863840539023021, 'size': 84.40379138403678}, {'class_id': 2, 'bbox': array([321, 698, 469, 759]), 'bbox_area': 9028, 'wh_ratio': 0.41216216216216217, 'mask_area': 4931, 'mask_ratio': 0.546189632255206, 'size': 95.01578816175761}, {'class_id': 2, 'bbox': array([389, 812, 602, 902]), 'bbox_area': 19170, 'wh_ratio': 0.4225352112676056, 'mask_area': 9090, 'mask_ratio': 0.47417840375586856, 'size': 138.4557691105719}, {'class_id': 4, 'bbox': array([240, 384, 281, 446]), 'bbox_area': 2542, 'wh_ratio': 1.5121951219512195, 'mask_area': 1264, 'mask_ratio': 0.4972462627852085, 'size': 50.41825066382213}, {'class_id': 4, 'bbox': array([259,   2, 542, 307]), 'bbox_area': 86315, 'wh_ratio': 1.0777385159010602, 'mask_area': 13135, 'mask_ratio': 0.1521751723338933, 'size': 293.7941456190031}, {'class_id': 4, 'bbox': array([237, 530, 275, 589]), 'bbox_area': 2242, 'wh_ratio': 1.5526315789473684, 'mask_area': 1028, 'mask_ratio': 0.45851917930419267, 'size': 47.34976240700686}, {'class_id': 4, 'bbox': array([ 258,  663,  550, 1022]), 'bbox_area': 104828, 'wh_ratio': 1.2294520547945205, 'mask_area': 15097, 'mask_ratio': 0.14401686572289846, 'size': 323.77152438100546}, {'class_id': 5, 'bbox': array([262, 320, 357, 364]), 'bbox_area': 4180, 'wh_ratio': 0.4631578947368421, 'mask_area': 3395, 'mask_ratio': 0.812200956937799, 'size': 64.65291950097846}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00046.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 322,  801,  738, 1022]), 'bbox_area': 91936, 'wh_ratio': 0.53125, 'mask_area': 63335, 'mask_ratio': 0.6889031500174034, 'size': 303.2094985319556}, {'class_id': 7, 'bbox': array([325,   2, 757, 185]), 'bbox_area': 79056, 'wh_ratio': 0.4236111111111111, 'mask_area': 56204, 'mask_ratio': 0.7109390811576604, 'size': 281.1689883326396}, {'class_id': 2, 'bbox': array([326, 226, 462, 276]), 'bbox_area': 6800, 'wh_ratio': 0.36764705882352944, 'mask_area': 3503, 'mask_ratio': 0.5151470588235294, 'size': 82.4621125123532}, {'class_id': 2, 'bbox': array([398,  87, 592, 190]), 'bbox_area': 19982, 'wh_ratio': 0.5309278350515464, 'mask_area': 8202, 'mask_ratio': 0.4104694224802322, 'size': 141.35770230164326}, {'class_id': 2, 'bbox': array([319, 700, 466, 762]), 'bbox_area': 9114, 'wh_ratio': 0.4217687074829932, 'mask_area': 4910, 'mask_ratio': 0.5387316216809305, 'size': 95.467271878901}, {'class_id': 2, 'bbox': array([390, 810, 598, 904]), 'bbox_area': 19552, 'wh_ratio': 0.4519230769230769, 'mask_area': 8975, 'mask_ratio': 0.4590323240589198, 'size': 139.82846634358827}, {'class_id': 3, 'bbox': array([308, 670, 666, 798]), 'bbox_area': 45824, 'wh_ratio': 0.3575418994413408, 'mask_area': 6787, 'mask_ratio': 0.14811016061452514, 'size': 214.06541056415443}, {'class_id': 3, 'bbox': array([265, 618, 492, 642]), 'bbox_area': 5448, 'wh_ratio': 0.10572687224669604, 'mask_area': 2618, 'mask_ratio': 0.48054331864904554, 'size': 73.81056834898374}, {'class_id': 3, 'bbox': array([242, 477, 378, 496]), 'bbox_area': 2584, 'wh_ratio': 0.13970588235294118, 'mask_area': 2389, 'mask_ratio': 0.9245356037151703, 'size': 50.83306010855534}, {'class_id': 3, 'bbox': array([242, 386, 494, 422]), 'bbox_area': 9072, 'wh_ratio': 0.14285714285714285, 'mask_area': 3640, 'mask_ratio': 0.4012345679012346, 'size': 95.24704719832526}, {'class_id': 3, 'bbox': array([270, 331, 502, 358]), 'bbox_area': 6264, 'wh_ratio': 0.11637931034482758, 'mask_area': 2828, 'mask_ratio': 0.45146871008939976, 'size': 79.14543574963753}, {'class_id': 3, 'bbox': array([317, 180, 673, 319]), 'bbox_area': 49484, 'wh_ratio': 0.3904494382022472, 'mask_area': 6096, 'mask_ratio': 0.1231913345727912, 'size': 222.44999438075965}, {'class_id': 4, 'bbox': array([260,   2, 543, 306]), 'bbox_area': 86032, 'wh_ratio': 1.0742049469964665, 'mask_area': 13248, 'mask_ratio': 0.15398921331597545, 'size': 293.3121204450985}, {'class_id': 4, 'bbox': array([239, 385, 282, 446]), 'bbox_area': 2623, 'wh_ratio': 1.4186046511627908, 'mask_area': 1270, 'mask_ratio': 0.48417842165459396, 'size': 51.215232109207506}, {'class_id': 4, 'bbox': array([237, 530, 274, 588]), 'bbox_area': 2146, 'wh_ratio': 1.5675675675675675, 'mask_area': 1021, 'mask_ratio': 0.47576887232059645, 'size': 46.32493928760188}, {'class_id': 4, 'bbox': array([ 252,  665,  550, 1023]), 'bbox_area': 106684, 'wh_ratio': 1.2013422818791946, 'mask_area': 15672, 'mask_ratio': 0.14690112856660792, 'size': 326.6251674320275}, {'class_id': 5, 'bbox': array([260, 320, 356, 365]), 'bbox_area': 4320, 'wh_ratio': 0.46875, 'mask_area': 3177, 'mask_ratio': 0.7354166666666667, 'size': 65.72670690061993}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00047.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 319,  801,  741, 1022]), 'bbox_area': 93262, 'wh_ratio': 0.523696682464455, 'mask_area': 63238, 'mask_ratio': 0.678068237867513, 'size': 305.3882774436504}, {'class_id': 7, 'bbox': array([328,   2, 757, 186]), 'bbox_area': 78936, 'wh_ratio': 0.4289044289044289, 'mask_area': 55696, 'mask_ratio': 0.7055842708016621, 'size': 280.95551249263644}, {'class_id': 2, 'bbox': array([323, 224, 463, 276]), 'bbox_area': 7280, 'wh_ratio': 0.37142857142857144, 'mask_area': 3809, 'mask_ratio': 0.5232142857142857, 'size': 85.32291603080617}, {'class_id': 2, 'bbox': array([396,  92, 594, 190]), 'bbox_area': 19404, 'wh_ratio': 0.494949494949495, 'mask_area': 8341, 'mask_ratio': 0.4298598227169656, 'size': 139.2982411949268}, {'class_id': 2, 'bbox': array([321, 702, 466, 759]), 'bbox_area': 8265, 'wh_ratio': 0.3931034482758621, 'mask_area': 5091, 'mask_ratio': 0.6159709618874774, 'size': 90.91204540653565}, {'class_id': 2, 'bbox': array([385, 811, 602, 905]), 'bbox_area': 20398, 'wh_ratio': 0.43317972350230416, 'mask_area': 9461, 'mask_ratio': 0.4638199823512109, 'size': 142.8215669988255}, {'class_id': 3, 'bbox': array([306, 666, 668, 797]), 'bbox_area': 47422, 'wh_ratio': 0.36187845303867405, 'mask_area': 6615, 'mask_ratio': 0.1394922188014002, 'size': 217.76592938290418}, {'class_id': 3, 'bbox': array([265, 615, 495, 644]), 'bbox_area': 6670, 'wh_ratio': 0.12608695652173912, 'mask_area': 3357, 'mask_ratio': 0.5032983508245877, 'size': 81.67006795638167}, {'class_id': 3, 'bbox': array([236, 479, 375, 496]), 'bbox_area': 2363, 'wh_ratio': 0.1223021582733813, 'mask_area': 2113, 'mask_ratio': 0.894202285230639, 'size': 48.61069841094654}, {'class_id': 3, 'bbox': array([241, 386, 491, 423]), 'bbox_area': 9250, 'wh_ratio': 0.148, 'mask_area': 3921, 'mask_ratio': 0.4238918918918919, 'size': 96.17692030835671}, {'class_id': 3, 'bbox': array([270, 332, 502, 360]), 'bbox_area': 6496, 'wh_ratio': 0.1206896551724138, 'mask_area': 3184, 'mask_ratio': 0.49014778325123154, 'size': 80.59776671843953}, {'class_id': 3, 'bbox': array([318, 185, 665, 318]), 'bbox_area': 46151, 'wh_ratio': 0.38328530259365995, 'mask_area': 2171, 'mask_ratio': 0.047041234209442916, 'size': 214.82783804712088}, {'class_id': 4, 'bbox': array([256,   3, 543, 310]), 'bbox_area': 88109, 'wh_ratio': 1.0696864111498259, 'mask_area': 14603, 'mask_ratio': 0.1657378928372811, 'size': 296.83160209115204}, {'class_id': 4, 'bbox': array([240, 386, 282, 446]), 'bbox_area': 2520, 'wh_ratio': 1.4285714285714286, 'mask_area': 1160, 'mask_ratio': 0.4603174603174603, 'size': 50.19960159204454}, {'class_id': 4, 'bbox': array([237, 529, 274, 590]), 'bbox_area': 2257, 'wh_ratio': 1.6486486486486487, 'mask_area': 1091, 'mask_ratio': 0.4833850243686309, 'size': 47.507894080878806}, {'class_id': 4, 'bbox': array([ 256,  667,  554, 1023]), 'bbox_area': 106088, 'wh_ratio': 1.1946308724832215, 'mask_area': 14985, 'mask_ratio': 0.14125065982957544, 'size': 325.7115288103877}, {'class_id': 5, 'bbox': array([261, 318, 358, 363]), 'bbox_area': 4365, 'wh_ratio': 0.4639175257731959, 'mask_area': 3378, 'mask_ratio': 0.7738831615120275, 'size': 66.06814663663572}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00048.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  801,  740, 1020]), 'bbox_area': 91761, 'wh_ratio': 0.522673031026253, 'mask_area': 62891, 'mask_ratio': 0.6853783197654777, 'size': 302.92078172353905}, {'class_id': 7, 'bbox': array([329,   2, 756, 186]), 'bbox_area': 78568, 'wh_ratio': 0.4309133489461358, 'mask_area': 55671, 'mask_ratio': 0.7085709194583036, 'size': 280.29983945767793}, {'class_id': 2, 'bbox': array([322, 226, 462, 277]), 'bbox_area': 7140, 'wh_ratio': 0.36428571428571427, 'mask_area': 3591, 'mask_ratio': 0.5029411764705882, 'size': 84.49852069711044}, {'class_id': 2, 'bbox': array([398,  91, 593, 186]), 'bbox_area': 18525, 'wh_ratio': 0.48717948717948717, 'mask_area': 7796, 'mask_ratio': 0.4208367071524966, 'size': 136.10657588816198}, {'class_id': 2, 'bbox': array([321, 698, 467, 762]), 'bbox_area': 9344, 'wh_ratio': 0.4383561643835616, 'mask_area': 5012, 'mask_ratio': 0.5363869863013698, 'size': 96.66436778875658}, {'class_id': 2, 'bbox': array([389, 811, 602, 904]), 'bbox_area': 19809, 'wh_ratio': 0.43661971830985913, 'mask_area': 9241, 'mask_ratio': 0.4665051239335655, 'size': 140.7444492688788}, {'class_id': 3, 'bbox': array([310, 666, 671, 798]), 'bbox_area': 47652, 'wh_ratio': 0.3656509695290859, 'mask_area': 7348, 'mask_ratio': 0.15420129270544783, 'size': 218.2933805684451}, {'class_id': 3, 'bbox': array([262, 618, 497, 644]), 'bbox_area': 6110, 'wh_ratio': 0.11063829787234042, 'mask_area': 2864, 'mask_ratio': 0.46873977086743046, 'size': 78.16648898345122}, {'class_id': 3, 'bbox': array([240, 478, 388, 498]), 'bbox_area': 2960, 'wh_ratio': 0.13513513513513514, 'mask_area': 2875, 'mask_ratio': 0.9712837837837838, 'size': 54.405882034941776}, {'class_id': 3, 'bbox': array([238, 383, 500, 426]), 'bbox_area': 11266, 'wh_ratio': 0.16412213740458015, 'mask_area': 4577, 'mask_ratio': 0.406266642996627, 'size': 106.1414151026827}, {'class_id': 3, 'bbox': array([269, 332, 502, 359]), 'bbox_area': 6291, 'wh_ratio': 0.11587982832618025, 'mask_area': 2786, 'mask_ratio': 0.4428548720394214, 'size': 79.31582439841371}, {'class_id': 3, 'bbox': array([318, 178, 674, 322]), 'bbox_area': 51264, 'wh_ratio': 0.4044943820224719, 'mask_area': 8238, 'mask_ratio': 0.16069756554307116, 'size': 226.4155471693585}, {'class_id': 4, 'bbox': array([254,   5, 544, 307]), 'bbox_area': 87580, 'wh_ratio': 1.0413793103448277, 'mask_area': 12837, 'mask_ratio': 0.14657456040191824, 'size': 295.9391829413604}, {'class_id': 4, 'bbox': array([240, 383, 282, 446]), 'bbox_area': 2646, 'wh_ratio': 1.5, 'mask_area': 1335, 'mask_ratio': 0.5045351473922902, 'size': 51.43928459844675}, {'class_id': 4, 'bbox': array([238, 528, 274, 589]), 'bbox_area': 2196, 'wh_ratio': 1.6944444444444444, 'mask_area': 1056, 'mask_ratio': 0.4808743169398907, 'size': 46.86149805543993}, {'class_id': 4, 'bbox': array([ 256,  666,  553, 1022]), 'bbox_area': 105732, 'wh_ratio': 1.1986531986531987, 'mask_area': 15944, 'mask_ratio': 0.15079635304354405, 'size': 325.1645737161415}, {'class_id': 5, 'bbox': array([258, 318, 357, 366]), 'bbox_area': 4752, 'wh_ratio': 0.48484848484848486, 'mask_area': 3707, 'mask_ratio': 0.7800925925925926, 'size': 68.93475175845634}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00049.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 318,  799,  741, 1022]), 'bbox_area': 94329, 'wh_ratio': 0.5271867612293144, 'mask_area': 64199, 'mask_ratio': 0.680586033987427, 'size': 307.13026552262806}, {'class_id': 7, 'bbox': array([327,   2, 755, 189]), 'bbox_area': 80036, 'wh_ratio': 0.4369158878504673, 'mask_area': 57471, 'mask_ratio': 0.7180643710330351, 'size': 282.90634492708006}, {'class_id': 2, 'bbox': array([323, 222, 462, 277]), 'bbox_area': 7645, 'wh_ratio': 0.39568345323741005, 'mask_area': 4010, 'mask_ratio': 0.5245258338783518, 'size': 87.43569065318808}, {'class_id': 2, 'bbox': array([398,  88, 592, 187]), 'bbox_area': 19206, 'wh_ratio': 0.5103092783505154, 'mask_area': 8157, 'mask_ratio': 0.4247110278038113, 'size': 138.58571354941319}, {'class_id': 2, 'bbox': array([322, 701, 466, 759]), 'bbox_area': 8352, 'wh_ratio': 0.4027777777777778, 'mask_area': 4687, 'mask_ratio': 0.5611829501915708, 'size': 91.3892772703669}, {'class_id': 2, 'bbox': array([389, 811, 602, 905]), 'bbox_area': 20022, 'wh_ratio': 0.4413145539906103, 'mask_area': 9335, 'mask_ratio': 0.4662371391469384, 'size': 141.4991166050163}, {'class_id': 3, 'bbox': array([302, 666, 666, 799]), 'bbox_area': 48412, 'wh_ratio': 0.36538461538461536, 'mask_area': 7220, 'mask_ratio': 0.14913657770800628, 'size': 220.02727103702398}, {'class_id': 3, 'bbox': array([264, 618, 495, 643]), 'bbox_area': 5775, 'wh_ratio': 0.10822510822510822, 'mask_area': 3033, 'mask_ratio': 0.5251948051948052, 'size': 75.99342076785331}, {'class_id': 3, 'bbox': array([238, 477, 393, 496]), 'bbox_area': 2945, 'wh_ratio': 0.12258064516129032, 'mask_area': 2763, 'mask_ratio': 0.9382003395585738, 'size': 54.26785420486054}, {'class_id': 3, 'bbox': array([241, 384, 502, 424]), 'bbox_area': 10440, 'wh_ratio': 0.1532567049808429, 'mask_area': 3552, 'mask_ratio': 0.34022988505747126, 'size': 102.17631819555841}, {'class_id': 3, 'bbox': array([267, 330, 501, 360]), 'bbox_area': 7020, 'wh_ratio': 0.1282051282051282, 'mask_area': 3177, 'mask_ratio': 0.45256410256410257, 'size': 83.78544026261365}, {'class_id': 3, 'bbox': array([317, 180, 678, 324]), 'bbox_area': 51984, 'wh_ratio': 0.3988919667590028, 'mask_area': 7379, 'mask_ratio': 0.14194752231455832, 'size': 228.0}, {'class_id': 4, 'bbox': array([258,   3, 541, 307]), 'bbox_area': 86032, 'wh_ratio': 1.0742049469964665, 'mask_area': 13424, 'mask_ratio': 0.1560349637344244, 'size': 293.3121204450985}, {'class_id': 4, 'bbox': array([240, 383, 282, 446]), 'bbox_area': 2646, 'wh_ratio': 1.5, 'mask_area': 1297, 'mask_ratio': 0.49017384731670444, 'size': 51.43928459844675}, {'class_id': 4, 'bbox': array([235, 528, 274, 591]), 'bbox_area': 2457, 'wh_ratio': 1.6153846153846154, 'mask_area': 1175, 'mask_ratio': 0.47822547822547823, 'size': 49.56813492557492}, {'class_id': 4, 'bbox': array([ 256,  666,  552, 1022]), 'bbox_area': 105376, 'wh_ratio': 1.2027027027027026, 'mask_area': 14658, 'mask_ratio': 0.13910188278165805, 'size': 324.61669704437577}, {'class_id': 5, 'bbox': array([260, 319, 362, 364]), 'bbox_area': 4590, 'wh_ratio': 0.4411764705882353, 'mask_area': 3637, 'mask_ratio': 0.7923747276688453, 'size': 67.7495387438173}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00050.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  801,  741, 1023]), 'bbox_area': 93240, 'wh_ratio': 0.5285714285714286, 'mask_area': 62830, 'mask_ratio': 0.6738524238524238, 'size': 305.35225559998736}, {'class_id': 7, 'bbox': array([326,   2, 766, 187]), 'bbox_area': 81400, 'wh_ratio': 0.42045454545454547, 'mask_area': 57555, 'mask_ratio': 0.707063882063882, 'size': 285.3068523537421}, {'class_id': 2, 'bbox': array([326, 226, 462, 276]), 'bbox_area': 6800, 'wh_ratio': 0.36764705882352944, 'mask_area': 3528, 'mask_ratio': 0.5188235294117647, 'size': 82.4621125123532}, {'class_id': 2, 'bbox': array([398,  88, 590, 190]), 'bbox_area': 19584, 'wh_ratio': 0.53125, 'mask_area': 8606, 'mask_ratio': 0.4394403594771242, 'size': 139.9428454762872}, {'class_id': 2, 'bbox': array([321, 701, 466, 758]), 'bbox_area': 8265, 'wh_ratio': 0.3931034482758621, 'mask_area': 4765, 'mask_ratio': 0.5765275257108288, 'size': 90.91204540653565}, {'class_id': 2, 'bbox': array([389, 811, 602, 907]), 'bbox_area': 20448, 'wh_ratio': 0.4507042253521127, 'mask_area': 9545, 'mask_ratio': 0.4667938184663537, 'size': 142.99650345375582}, {'class_id': 3, 'bbox': array([308, 668, 671, 796]), 'bbox_area': 46464, 'wh_ratio': 0.3526170798898072, 'mask_area': 7182, 'mask_ratio': 0.15457128099173553, 'size': 215.55509736491967}, {'class_id': 3, 'bbox': array([264, 615, 496, 647]), 'bbox_area': 7424, 'wh_ratio': 0.13793103448275862, 'mask_area': 3740, 'mask_ratio': 0.5037715517241379, 'size': 86.16263691415206}, {'class_id': 3, 'bbox': array([242, 477, 392, 498]), 'bbox_area': 3150, 'wh_ratio': 0.14, 'mask_area': 3000, 'mask_ratio': 0.9523809523809523, 'size': 56.12486080160912}, {'class_id': 3, 'bbox': array([241, 384, 514, 425]), 'bbox_area': 11193, 'wh_ratio': 0.15018315018315018, 'mask_area': 4113, 'mask_ratio': 0.36746180648619675, 'size': 105.79697538209682}, {'class_id': 3, 'bbox': array([270, 331, 500, 360]), 'bbox_area': 6670, 'wh_ratio': 0.12608695652173912, 'mask_area': 2881, 'mask_ratio': 0.43193403298350824, 'size': 81.67006795638167}, {'class_id': 3, 'bbox': array([317, 179, 674, 320]), 'bbox_area': 50337, 'wh_ratio': 0.3949579831932773, 'mask_area': 6636, 'mask_ratio': 0.13183145598664997, 'size': 224.35908717945884}, {'class_id': 4, 'bbox': array([259,   2, 541, 307]), 'bbox_area': 86010, 'wh_ratio': 1.0815602836879432, 'mask_area': 14644, 'mask_ratio': 0.17025927217765377, 'size': 293.2746153351838}, {'class_id': 4, 'bbox': array([240, 384, 282, 446]), 'bbox_area': 2604, 'wh_ratio': 1.4761904761904763, 'mask_area': 1219, 'mask_ratio': 0.4681259600614439, 'size': 51.0294032886923}, {'class_id': 4, 'bbox': array([237, 527, 275, 591]), 'bbox_area': 2432, 'wh_ratio': 1.6842105263157894, 'mask_area': 1317, 'mask_ratio': 0.5415296052631579, 'size': 49.31531202375181}, {'class_id': 4, 'bbox': array([ 259,  664,  552, 1022]), 'bbox_area': 104894, 'wh_ratio': 1.2218430034129693, 'mask_area': 15926, 'mask_ratio': 0.1518294659370412, 'size': 323.8734320687635}, {'class_id': 5, 'bbox': array([261, 319, 362, 366]), 'bbox_area': 4747, 'wh_ratio': 0.46534653465346537, 'mask_area': 3805, 'mask_ratio': 0.8015588792921845, 'size': 68.89847603539573}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00051.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  799,  741, 1022]), 'bbox_area': 93660, 'wh_ratio': 0.530952380952381, 'mask_area': 64915, 'mask_ratio': 0.6930920350202862, 'size': 306.0392131737369}, {'class_id': 7, 'bbox': array([329,   1, 762, 186]), 'bbox_area': 80105, 'wh_ratio': 0.42725173210161665, 'mask_area': 57052, 'mask_ratio': 0.7122152175269958, 'size': 283.02826713952084}, {'class_id': 2, 'bbox': array([321, 227, 466, 276]), 'bbox_area': 7105, 'wh_ratio': 0.33793103448275863, 'mask_area': 4026, 'mask_ratio': 0.566643209007741, 'size': 84.29116205154607}, {'class_id': 2, 'bbox': array([394,  89, 595, 187]), 'bbox_area': 19698, 'wh_ratio': 0.48756218905472637, 'mask_area': 8619, 'mask_ratio': 0.4375571123971977, 'size': 140.34956359034396}, {'class_id': 2, 'bbox': array([321, 698, 466, 762]), 'bbox_area': 9280, 'wh_ratio': 0.4413793103448276, 'mask_area': 5317, 'mask_ratio': 0.5729525862068966, 'size': 96.33275663033835}, {'class_id': 2, 'bbox': array([389, 811, 602, 905]), 'bbox_area': 20022, 'wh_ratio': 0.4413145539906103, 'mask_area': 9528, 'mask_ratio': 0.47587653581060835, 'size': 141.4991166050163}, {'class_id': 3, 'bbox': array([308, 669, 669, 794]), 'bbox_area': 45125, 'wh_ratio': 0.3462603878116344, 'mask_area': 6255, 'mask_ratio': 0.13861495844875346, 'size': 212.42645786248}, {'class_id': 3, 'bbox': array([262, 618, 497, 644]), 'bbox_area': 6110, 'wh_ratio': 0.11063829787234042, 'mask_area': 2950, 'mask_ratio': 0.4828150572831424, 'size': 78.16648898345122}, {'class_id': 3, 'bbox': array([230, 478, 430, 496]), 'bbox_area': 3600, 'wh_ratio': 0.09, 'mask_area': 3044, 'mask_ratio': 0.8455555555555555, 'size': 60.0}, {'class_id': 3, 'bbox': array([240, 383, 523, 426]), 'bbox_area': 12169, 'wh_ratio': 0.1519434628975265, 'mask_area': 3888, 'mask_ratio': 0.3195003697920947, 'size': 110.31319050775387}, {'class_id': 3, 'bbox': array([270, 331, 502, 359]), 'bbox_area': 6496, 'wh_ratio': 0.1206896551724138, 'mask_area': 2728, 'mask_ratio': 0.41995073891625617, 'size': 80.59776671843953}, {'class_id': 3, 'bbox': array([318, 178, 673, 322]), 'bbox_area': 51120, 'wh_ratio': 0.4056338028169014, 'mask_area': 7223, 'mask_ratio': 0.14129499217527386, 'size': 226.09732417700124}, {'class_id': 4, 'bbox': array([258,   4, 542, 308]), 'bbox_area': 86336, 'wh_ratio': 1.0704225352112675, 'mask_area': 14880, 'mask_ratio': 0.17234988880652335, 'size': 293.82988275531136}, {'class_id': 4, 'bbox': array([238, 384, 280, 446]), 'bbox_area': 2604, 'wh_ratio': 1.4761904761904763, 'mask_area': 1176, 'mask_ratio': 0.45161290322580644, 'size': 51.0294032886923}, {'class_id': 4, 'bbox': array([238, 529, 277, 589]), 'bbox_area': 2340, 'wh_ratio': 1.5384615384615385, 'mask_area': 1268, 'mask_ratio': 0.5418803418803418, 'size': 48.373546489791295}, {'class_id': 4, 'bbox': array([ 256,  666,  553, 1022]), 'bbox_area': 105732, 'wh_ratio': 1.1986531986531987, 'mask_area': 14844, 'mask_ratio': 0.1403926909544887, 'size': 325.1645737161415}, {'class_id': 5, 'bbox': array([259, 317, 358, 365]), 'bbox_area': 4752, 'wh_ratio': 0.48484848484848486, 'mask_area': 3603, 'mask_ratio': 0.7582070707070707, 'size': 68.93475175845634}], 'objs': []}, {'file': 'D:\\\\Projects\\\\Mask_RCNN\\\\my_mask_rcnn\\\\dataset_1119\\\\val\\\\scene00052.jpeg', 'gts': [{'class_id': 7, 'bbox': array([ 321,  801,  744, 1022]), 'bbox_area': 93483, 'wh_ratio': 0.5224586288416075, 'mask_area': 63923, 'mask_ratio': 0.6837927751569804, 'size': 305.7498977922969}, {'class_id': 7, 'bbox': array([327,   2, 757, 186]), 'bbox_area': 79120, 'wh_ratio': 0.42790697674418604, 'mask_area': 56382, 'mask_ratio': 0.7126137512639029, 'size': 281.28277586798663}, {'class_id': 2, 'bbox': array([327, 226, 462, 276]), 'bbox_area': 6750, 'wh_ratio': 0.37037037037037035, 'mask_area': 3549, 'mask_ratio': 0.5257777777777778, 'size': 82.15838362577492}, {'class_id': 2, 'bbox': array([394,  83, 594, 189]), 'bbox_area': 21200, 'wh_ratio': 0.53, 'mask_area': 8362, 'mask_ratio': 0.39443396226415095, 'size': 145.60219778561037}, {'class_id': 2, 'bbox': array([321, 700, 466, 760]), 'bbox_area': 8700, 'wh_ratio': 0.41379310344827586, 'mask_area': 4808, 'mask_ratio': 0.5526436781609195, 'size': 93.27379053088815}, {'class_id': 2, 'bbox': array([389, 809, 602, 905]), 'bbox_area': 20448, 'wh_ratio': 0.4507042253521127, 'mask_area': 9630, 'mask_ratio': 0.4709507042253521, 'size': 142.99650345375582}, {'class_id': 3, 'bbox': array([306, 667, 671, 798]), 'bbox_area': 47815, 'wh_ratio': 0.3589041095890411, 'mask_area': 7956, 'mask_ratio': 0.1663912998013176, 'size': 218.66641260147844}, {'class_id': 3, 'bbox': array([264, 618, 497, 646]), 'bbox_area': 6524, 'wh_ratio': 0.12017167381974249, 'mask_area': 3396, 'mask_ratio': 0.5205395462906193, 'size': 80.77128202523468}, {'class_id': 3, 'bbox': array([238, 479, 439, 497]), 'bbox_area': 3618, 'wh_ratio': 0.08955223880597014, 'mask_area': 3571, 'mask_ratio': 0.9870093974571587, 'size': 60.14981296729026}, {'class_id': 3, 'bbox': array([241, 382, 530, 427]), 'bbox_area': 13005, 'wh_ratio': 0.15570934256055363, 'mask_area': 4680, 'mask_ratio': 0.35986159169550175, 'size': 114.03946685248927}, {'class_id': 3, 'bbox': array([272, 330, 502, 358]), 'bbox_area': 6440, 'wh_ratio': 0.12173913043478261, 'mask_area': 2854, 'mask_ratio': 0.443167701863354, 'size': 80.24961059095551}, {'class_id': 3, 'bbox': array([318, 179, 676, 322]), 'bbox_area': 51194, 'wh_ratio': 0.3994413407821229, 'mask_area': 7147, 'mask_ratio': 0.13960620385201392, 'size': 226.26091133909983}, {'class_id': 4, 'bbox': array([252,   4, 542, 309]), 'bbox_area': 88450, 'wh_ratio': 1.0517241379310345, 'mask_area': 13366, 'mask_ratio': 0.1511136235161108, 'size': 297.40544715926103}, {'class_id': 4, 'bbox': array([239, 382, 281, 444]), 'bbox_area': 2604, 'wh_ratio': 1.4761904761904763, 'mask_area': 1138, 'mask_ratio': 0.43701996927803377, 'size': 51.0294032886923}, {'class_id': 4, 'bbox': array([238, 530, 274, 590]), 'bbox_area': 2160, 'wh_ratio': 1.6666666666666667, 'mask_area': 1105, 'mask_ratio': 0.5115740740740741, 'size': 46.475800154489}, {'class_id': 4, 'bbox': array([ 260,  666,  550, 1022]), 'bbox_area': 103240, 'wh_ratio': 1.2275862068965517, 'mask_area': 14386, 'mask_ratio': 0.13934521503293298, 'size': 321.30981933330327}, {'class_id': 5, 'bbox': array([262, 318, 362, 365]), 'bbox_area': 4700, 'wh_ratio': 0.47, 'mask_area': 3320, 'mask_ratio': 0.7063829787234043, 'size': 68.55654600401044}], 'objs': []}]\n"
     ]
    }
   ],
   "source": [
    "# Import Mask RCNN\n",
    "# sys.path.append(ROOT_DIR)  # To find local version of the library\n",
    "import os, json\n",
    "import numpy as np\n",
    "import skimage.draw\n",
    "import math\n",
    "from mrcnn.config import Config\n",
    "from mrcnn import model as modellib, utils\n",
    "from mrcnn import visualize\n",
    "\n",
    "# Path to data set\n",
    "DATASET = \"D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\\\\"\n",
    "\n",
    "class MyConfig(Config):\n",
    "    \"\"\"Configuration for training on the toy  dataset.\n",
    "    Derives from the base Config class and overrides some values.\n",
    "    \"\"\"\n",
    "    # Give the configuration a recognizable name\n",
    "    NAME = \"mask\"\n",
    "\n",
    "    # We use a GPU with 12GB memory, which can fit two images.\n",
    "    # Adjust down if you use a smaller GPU.\n",
    "    IMAGES_PER_GPU = 1\n",
    "\n",
    "    # Number of classes (including background)\n",
    "    NUM_CLASSES = 1 + 7  # Background + my\n",
    "\n",
    "    # Number of training steps per epoch\n",
    "    STEPS_PER_EPOCH = 100\n",
    "\n",
    "    # Skip detections with < 90% confidence\n",
    "    DETECTION_MIN_CONFIDENCE = 0.9\n",
    "    BACKBONE = \"resnet101\"\n",
    "    IMAGE_RESIZE_MODE = \"square\"\n",
    "    IMAGE_MIN_DIM = 800  \n",
    "    IMAGE_MAX_DIM = 1024\n",
    "    IMAGE_MIN_SCALE = 0\n",
    "\n",
    "    # BACKBONE = \"resnet50\"\n",
    "    # BACKBONE_STRIDES = [4, 8, 16, 32, 64]\n",
    "   # # BACKBONE_STRIDES = [2, 4, 8, 16, 32]\n",
    "    # RPN_ANCHOR_SCALES = (10, 32, 64, 128, 256)\n",
    "    # RPN_ANCHOR_STRIDE = 2\n",
    "    # RPN_NMS_THRESHOLD = 0.9\n",
    "    # RPN_TRAIN_ANCHORS_PER_IMAGE = 512\n",
    "    # TRAIN_ROIS_PER_IMAGE = 512\n",
    "\n",
    "############################################################\n",
    "#  Dataset\n",
    "############################################################\n",
    "\n",
    "class MyDataset(utils.Dataset):\n",
    "\n",
    "    def print_size(self, poly):\n",
    "        for p in poly:\n",
    "            a = np.array(p['all_points_y'])\n",
    "            height = a.max() - a.min()\n",
    "            a = np.array(p['all_points_x'])\n",
    "            width = a.max() - a.min()\n",
    "            self.areas.append(height * width)\n",
    "            #if height * width < 4096:\n",
    "            #    print(width, height)\n",
    "\n",
    "    def load_my(self, dataset_dir, subset, class_dict):\n",
    "        \"\"\"Load a subset of the My dataset.\n",
    "        dataset_dir: Root directory of the dataset.\n",
    "        subset: Subset to load: train or val\n",
    "        \"\"\"\n",
    "        self.areas = []\n",
    "        # Add classes. We have only one class to add.\n",
    "        for (k, v) in class_dict.items():\n",
    "            self.add_class(\"my\", v, k)\n",
    "\n",
    "        # Train or validation dataset?\n",
    "        assert subset in [\"train\", \"val\"]\n",
    "        dataset_dir = os.path.join(dataset_dir, subset)\n",
    "\n",
    "        # Load annotations\n",
    "        # VGG Image Annotator (up to version 1.6) saves each image in the form:\n",
    "        # { 'filename': '28503151_5b5b7ec140_b.jpg',\n",
    "        #   'regions': {\n",
    "        #       '0': {\n",
    "        #           'region_attributes': {},\n",
    "        #           'shape_attributes': {\n",
    "        #               'all_points_x': [...],\n",
    "        #               'all_points_y': [...],\n",
    "        #               'name': 'polygon'}},\n",
    "        #       ... more regions ...\n",
    "        #   },\n",
    "        #   'size': 100202\n",
    "        # }\n",
    "        # We mostly care about the x and y coordinates of each region\n",
    "        # Note: In VIA 2.0, regions was changed from a dict to a list.\n",
    "        annotations = json.load(open(os.path.join(dataset_dir, \"via_region_data.json\")))\n",
    "        annotations = list(annotations.values())  # don't need the dict keys\n",
    "\n",
    "        # The VIA tool saves images in the JSON even if they don't have any\n",
    "        # annotations. Skip unannotated images.\n",
    "        annotations = [a for a in annotations if a['regions']]\n",
    "\n",
    "        # Add images\n",
    "        for a in annotations:\n",
    "            # Get the x, y coordinaets of points of the polygons that make up\n",
    "            # the outline of each object instance. These are stores in the\n",
    "            # shape_attributes (see json format above)\n",
    "            # The if condition is needed to support VIA versions 1.x and 2.x.\n",
    "            # print(a['regions'])\n",
    "            # print(a['filename'])\n",
    "            if type(a['regions']) is dict:\n",
    "                polygons = [r['shape_attributes'] for r in a['regions'].values()]\n",
    "            else:\n",
    "                if a['regions']:\n",
    "                    class_ids = []\n",
    "                    polygons = []\n",
    "                    for r in a['regions']:\n",
    "                        polygons.append(r['shape_attributes'])\n",
    "                        class_type = r['region_attributes']['type']\n",
    "                        class_ids.append(class_dict[class_type])\n",
    "                        \n",
    "                    self.print_size(polygons)\n",
    "                    # print(class_ids)\n",
    "                        \n",
    "\n",
    "            # load_mask() needs the image size to convert polygons to masks.\n",
    "            # Unfortunately, VIA doesn't include it in JSON, so we must read\n",
    "            # the image. This is only managable since the dataset is tiny.\n",
    "                    image_path = os.path.join(dataset_dir, a['filename'])\n",
    "                    image = skimage.io.imread(image_path)\n",
    "                    height, width = image.shape[:2]\n",
    "\n",
    "                    self.add_image(\n",
    "                        \"my\",\n",
    "                        image_id=a['filename'],  # use file name as a unique image id\n",
    "                        path=image_path,\n",
    "                        width=width, height=height,\n",
    "                        polygons=polygons,\n",
    "                        class_ids=class_ids)\n",
    "        self.areas.sort()\n",
    "        print(np.unique(np.round(np.sqrt(self.areas))))\n",
    "\n",
    "    def load_mask(self, image_id):\n",
    "        \"\"\"Generate instance masks for an image.\n",
    "       Returns:\n",
    "        masks: A bool array of shape [height, width, instance count] with\n",
    "            one mask per instance.\n",
    "        class_ids: a 1D array of class IDs of the instance masks.\n",
    "        \"\"\"\n",
    "        # If not a my dataset image, delegate to parent class.\n",
    "        image_info = self.image_info[image_id]\n",
    "        if image_info[\"source\"] != \"my\":\n",
    "            return super(self.__class__, self).load_mask(image_id)\n",
    "\n",
    "        # Convert polygons to a bitmap mask of shape\n",
    "        # [height, width, instance_count]\n",
    "        info = self.image_info[image_id]\n",
    "        mask = np.zeros([info[\"height\"], info[\"width\"], len(info[\"polygons\"])],\n",
    "                        dtype=np.uint8)\n",
    "        for i, p in enumerate(info[\"polygons\"]):\n",
    "            # Get indexes of pixels inside the polygon and set them to 1\n",
    "            rr, cc = skimage.draw.polygon(p['all_points_y'], p['all_points_x'])\n",
    "            mask[rr, cc, i] = 1\n",
    "        \n",
    "        class_ids = np.array(info['class_ids'])\n",
    "        # Return mask, and array of class IDs of each instance. Since we have\n",
    "        # one class ID only, we return an array of 1s\n",
    "        return mask.astype(np.bool), class_ids.astype(np.int32)\n",
    "\n",
    "    def image_reference(self, image_id):\n",
    "        \"\"\"Return the path of the image.\"\"\"\n",
    "        info = self.image_info[image_id]\n",
    "        if info[\"source\"] == \"my\":\n",
    "            return info[\"path\"]\n",
    "        else:\n",
    "            super(self.__class__, self).image_reference(image_id)\n",
    "\n",
    "target = {\n",
    "    'file' : '',\n",
    "    'gts' : [],\n",
    "    'objs' : []\n",
    "}\n",
    "obj = {\n",
    "    'class_id' : 1,\n",
    "    'bbox' : [],\n",
    "    'bbox_area' : 123,\n",
    "    'wh_ratio' : 0.34,\n",
    "    'mask_area' : 123,\n",
    "    'mask_ratio' : 0.1,\n",
    "    'score' : 1\n",
    "}\n",
    "config = MyConfig()\n",
    "class_dict = {}\n",
    "label_file = open(DATASET+'label.txt')\n",
    "label_lines = label_file.readlines()\n",
    "label_id = 1\n",
    "for label_line in label_lines:\n",
    "    label_line = label_line.replace('\\n', '')\n",
    "    class_dict[label_line] = label_id\n",
    "    label_id = label_id + 1\n",
    "# Validation dataset\n",
    "dataset_val = MyDataset()\n",
    "dataset_val.load_my(DATASET, \"val\", class_dict)\n",
    "dataset_val.prepare()\n",
    "gt_targets = []\n",
    "for image_id in dataset_val.image_ids:\n",
    "    image, image_meta, gt_class_id, gt_box, gt_mask = modellib.load_image_gt(dataset_val, config, image_id, use_mini_mask=False)\n",
    "    print(dataset_val.image_reference(image_id))\n",
    "    target = {\n",
    "        'file' : dataset_val.image_reference(image_id),\n",
    "        'gts' : [],\n",
    "        'objs' : []\n",
    "    }\n",
    "    for i in range(0,len(gt_class_id)):\n",
    "        box_height = gt_box[i][2] - gt_box[i][0]\n",
    "        box_width = gt_box[i][3] - gt_box[i][1]\n",
    "        box_area = box_height * box_width\n",
    "        box_ratio = box_width / box_height\n",
    "        mask_area = np.sum(gt_mask[:,:,i]==True)\n",
    "        mask_ratio = mask_area / box_area\n",
    "        obj = {}\n",
    "        obj['class_id'] = gt_class_id[i]\n",
    "        obj['bbox'] = gt_box[i]\n",
    "        obj['bbox_area'] = box_area\n",
    "        obj['wh_ratio'] = box_ratio\n",
    "        obj['mask_area'] = mask_area\n",
    "        obj['mask_ratio'] = mask_ratio\n",
    "        obj['size'] = (obj['bbox'][3] - obj['bbox'][1]) / math.sqrt(obj['wh_ratio'])\n",
    "        target['gts'].append(obj)\n",
    "        print(gt_class_id[i], gt_box[i], box_area, box_ratio, mask_area, mask_ratio)\n",
    "    gt_targets.append(target)\n",
    "print(gt_targets)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 标注目标的正方形尺寸\n",
    "\n",
    "上面已知标注的目标尺寸与宽高比，可以计算目标的正方形尺寸\n",
    "\n",
    "$$Size = \\lceil \\frac {width}{ \\sqrt {wh\\_ratio}} \\rceil$$\n",
    "\n",
    "尺寸$Size$向上取整。\n",
    "\n",
    "宽高比$wh\\_ratio$设定为小数点后一位，进位后舍弃后面的小数。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "# print(gt_targets)\n",
    "for target in gt_targets:\n",
    "    target['gt_wh_ratios'] = []\n",
    "    target['gt_sizes'] = []\n",
    "    for i in range(0, 8):\n",
    "        #target['obj_sizes'] = [obj['size'] for obj in target['objs'] if obj['class_id'] == 7]\n",
    "        #target['obj_wh_ratios'] = [obj['wh_ratio'] for obj in target['objs'] if obj['class_id'] == 7]\n",
    "        target['gt_wh_ratios'].append([gt['wh_ratio'] for gt in target['gts'] if gt['class_id'] == i])#gt_wh_ratios\n",
    "        target['gt_sizes'].append([gt['size'] for gt in target['gts'] if gt['class_id'] == i])#gt_wh_ratiosgt_sizes\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 尺寸、宽高比分布\n",
    "\n",
    "横坐标为尺寸，纵坐标为宽高比，把所有标注目标绘制坐标图\n",
    "\n",
    "主要查看目标的分布情况，猜测目标有7类，又在同一场景，不会有多少点"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.cla()\n",
    "plt.clf()\n",
    "plt.close()\n",
    "for i in range(0, 8):\n",
    "    for target in gt_targets:\n",
    "        # plt.title(target['file'])\n",
    "        #print('标注目标尺寸：', np.sort(target['obj_sizes']))\n",
    "        #print('标注目标宽高比：', np.sort(target['obj_wh_ratios']))\n",
    "        plt.plot(target['gt_sizes'][i], target['gt_wh_ratios'][i], 'ro')\n",
    "        #plt.plot(target['obj_sizes'], target['obj_wh_ratios'], 'go')\n",
    "        plt.axis([0, 512, 0, 2])\n",
    "    plt.title(i)\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 目标检测数据\n",
    "\n",
    "对数据集，如验证集（val）、测试集（test）等都可以，每张图片检测后把结果收集。\n",
    "\n",
    "检测结果：类别、ROIS、目标框面积、宽高比、掩码面积、面积比、得分"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\tensorflow\\python\\framework\\op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "WARNING:tensorflow:From c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\mask_rcnn-2.1-py3.7.egg\\mrcnn\\model.py:772: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Re-starting from epoch 500\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 5.777140378952026\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00028.jpeg\n",
      "3 [310 673 670 806] 47880 0.36944444444444446 6720 0.14035087719298245 0.99999785\n",
      "4 [ 250  674  544 1010] 98784 1.1428571428571428 12149 0.12298550372529965 0.9999875\n",
      "4 [239 530 273 591] 2074 1.7941176470588236 873 0.42092574734811955 0.99998224\n",
      "2 [387 802 611 901] 22176 0.4419642857142857 7608 0.3430735930735931 0.999969\n",
      "3 [244 479 611 504] 9175 0.0681198910081744 5424 0.5911716621253406 0.99993825\n",
      "2 [317 700 477 760] 9600 0.375 4799 0.4998958333333333 0.9999223\n",
      "7 [ 306  801  747 1022] 97461 0.5011337868480725 60741 0.6232339089481946 0.9999187\n",
      "3 [232 385 532 426] 12300 0.13666666666666666 3790 0.308130081300813 0.9997745\n",
      "3 [269 334 503 360] 6084 0.1111111111111111 2452 0.40302432610124916 0.99977285\n",
      "2 [321 228 463 273] 6390 0.31690140845070425 3425 0.5359937402190923 0.99962544\n",
      "4 [242 388 279 451] 2331 1.7027027027027026 961 0.41226941226941227 0.999521\n",
      "3 [329 182 677 313] 45588 0.3764367816091954 4743 0.10404053698341668 0.9988943\n",
      "2 [412  87 593 184] 17557 0.5359116022099447 6210 0.35370507489890074 0.9985568\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3268 0.6685761047463176 0.997235\n",
      "4 [261  17 566 318] 91805 0.9868852459016394 10062 0.10960187353629977 0.9748989\n",
      "7 [338   0 771 186] 80538 0.4295612009237875 51890 0.6442921353895056 0.9738632\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4938673973083496\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00029.jpeg\n",
      "3 [303 669 673 811] 52540 0.3837837837837838 6516 0.12401979444232965 0.9999925\n",
      "4 [241 661 543 995] 100868 1.1059602649006623 11178 0.11081809890153468 0.99998784\n",
      "4 [239 531 273 590] 2006 1.7352941176470589 841 0.41924227318045865 0.99998593\n",
      "2 [391 804 614 901] 21631 0.4349775784753363 7600 0.35134760297720863 0.9999684\n",
      "7 [ 297  801  727 1023] 95460 0.5162790697674419 59069 0.6187827362245967 0.99991465\n",
      "3 [268 334 502 360] 6084 0.1111111111111111 2445 0.40187376725838264 0.9997695\n",
      "3 [233 384 528 426] 12390 0.1423728813559322 3552 0.28668280871670704 0.9997477\n",
      "2 [312 700 463 762] 9362 0.4105960264900662 4228 0.45161290322580644 0.9996505\n",
      "2 [321 227 465 273] 6624 0.3194444444444444 3222 0.48641304347826086 0.99957186\n",
      "2 [385  84 589 179] 19380 0.46568627450980393 6771 0.3493808049535604 0.99947494\n",
      "3 [225 481 602 501] 7540 0.05305039787798409 5083 0.6741379310344827 0.99943334\n",
      "3 [330 183 678 314] 45588 0.3764367816091954 4524 0.09923664122137404 0.9992029\n",
      "4 [243 387 279 451] 2304 1.7777777777777777 956 0.4149305555555556 0.9991147\n",
      "5 [255 320 358 367] 4841 0.4563106796116505 3251 0.6715554637471597 0.9970898\n",
      "4 [262  39 566 322] 86032 0.930921052631579 8965 0.10420541193974335 0.99061865\n",
      "7 [339   2 754 204] 83830 0.4867469879518072 51523 0.6146129070738399 0.9463216\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4768483638763428\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00030.jpeg\n",
      "4 [239 531 272 592] 2013 1.8484848484848484 851 0.4227521112767014 0.999977\n",
      "4 [ 254  656  550 1009] 104488 1.1925675675675675 12547 0.12008077482581732 0.9999703\n",
      "3 [321 670 662 825] 52855 0.45454545454545453 4572 0.08650080408665216 0.9999505\n",
      "2 [384 799 610 903] 23504 0.46017699115044247 7788 0.3313478556841389 0.9998543\n",
      "4 [242 387 278 453] 2376 1.8333333333333333 958 0.4031986531986532 0.99982774\n",
      "3 [236 384 523 426] 12054 0.14634146341463414 3450 0.28621204579392734 0.99981683\n",
      "7 [ 299  792  751 1024] 104864 0.5132743362831859 55618 0.530382209337809 0.9996538\n",
      "2 [323 227 464 273] 6486 0.3262411347517731 3241 0.4996916435399322 0.9995074\n",
      "3 [267 334 504 360] 6162 0.10970464135021098 2501 0.40587471600129826 0.999405\n",
      "3 [330 184 682 314] 45760 0.3693181818181818 4718 0.10310314685314685 0.99934775\n",
      "3 [224 481 583 502] 7539 0.0584958217270195 4673 0.6198434805677145 0.99901533\n",
      "2 [412  87 593 184] 17557 0.5359116022099447 6259 0.3564959845076038 0.9984794\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3245 0.6638707037643208 0.99713045\n",
      "2 [297 701 470 766] 11245 0.37572254335260113 3519 0.31293908403734993 0.9956507\n",
      "7 [339   0 771 188] 81216 0.4351851851851852 51693 0.6364878841607565 0.9824187\n",
      "4 [257  14 568 327] 97343 1.0064308681672025 10610 0.10899602436744296 0.9673616\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4828464984893799\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00031.jpeg\n",
      "2 [384 796 614 905] 25070 0.47391304347826085 7756 0.3093737534902274 0.99999535\n",
      "4 [238 530 273 592] 2170 1.7714285714285714 891 0.41059907834101383 0.9999826\n",
      "4 [ 247  675  547 1016] 102300 1.1366666666666667 12644 0.12359726295210166 0.99993443\n",
      "3 [315 666 671 820] 54824 0.43258426966292135 4070 0.07423756019261638 0.9999132\n",
      "3 [240 385 521 426] 11521 0.14590747330960854 3256 0.28261435639267424 0.9998721\n",
      "2 [320 227 464 273] 6624 0.3194444444444444 3274 0.49426328502415456 0.99972016\n",
      "3 [332 184 683 312] 44928 0.3646723646723647 4767 0.10610309829059829 0.9996698\n",
      "2 [386  84 589 180] 19488 0.4729064039408867 6696 0.3435960591133005 0.99946934\n",
      "3 [269 333 498 360] 6183 0.11790393013100436 2500 0.40433446546983665 0.9993079\n",
      "4 [243 388 279 452] 2304 1.7777777777777777 863 0.3745659722222222 0.9991584\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3256 0.6661211129296236 0.99628663\n",
      "3 [226 481 582 501] 7120 0.056179775280898875 4306 0.6047752808988764 0.99296135\n",
      "7 [ 310  806  726 1024] 90688 0.5240384615384616 58460 0.6446277346506705 0.9927297\n",
      "2 [300 700 462 772] 11664 0.4444444444444444 3765 0.3227880658436214 0.98736316\n",
      "7 [336   1 771 189] 81780 0.432183908045977 51940 0.6351186109073123 0.9747726\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.48484134674072266\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00032.jpeg\n",
      "3 [320 664 688 820] 57408 0.42391304347826086 4892 0.08521460423634336 0.9999927\n",
      "4 [249 658 544 972] 92630 1.064406779661017 11499 0.12413904782467883 0.99998987\n",
      "4 [238 530 273 590] 2100 1.7142857142857142 864 0.4114285714285714 0.9999691\n",
      "2 [394 803 596 902] 19998 0.4900990099009901 7652 0.38263826382638266 0.9999473\n",
      "3 [330 185 678 311] 43848 0.3620689655172414 4534 0.10340266374749134 0.9997429\n",
      "3 [244 385 528 426] 11644 0.1443661971830986 3172 0.27241497767090345 0.9996892\n",
      "2 [320 226 465 272] 6670 0.31724137931034485 3222 0.48305847076461766 0.9995103\n",
      "4 [243 388 279 452] 2304 1.7777777777777777 878 0.3810763888888889 0.99935776\n",
      "2 [320 705 469 768] 9387 0.4228187919463087 3592 0.3826568658783424 0.9990376\n",
      "2 [384  84 588 180] 19584 0.47058823529411764 6663 0.3402267156862745 0.9987741\n",
      "3 [269 333 501 360] 6264 0.11637931034482758 2441 0.38968710089399744 0.99860746\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3212 0.6571194762684124 0.996863\n",
      "7 [332   0 771 194] 85166 0.4419134396355353 52516 0.6166310499495103 0.9794988\n",
      "3 [236 479 549 501] 6886 0.07028753993610223 3608 0.5239616613418531 0.9734927\n",
      "7 [ 323  806  726 1024] 87854 0.5409429280397022 55701 0.6340178022628452 0.97283584\n",
      "4 [251  42 572 334] 93732 0.9096573208722741 8906 0.09501557632398754 0.97030634\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.48682069778442383\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00033.jpeg\n",
      "4 [ 252  683  547 1022] 100005 1.1491525423728814 12587 0.12586370681465928 0.999987\n",
      "3 [315 676 683 817] 51888 0.38315217391304346 4107 0.07915124884366327 0.9999763\n",
      "2 [392 804 596 900] 19584 0.47058823529411764 7559 0.3859783496732026 0.99994516\n",
      "4 [238 529 273 590] 2135 1.7428571428571429 801 0.37517564402810305 0.9998363\n",
      "3 [238 385 528 425] 11600 0.13793103448275862 3076 0.26517241379310347 0.9997937\n",
      "3 [328 185 678 311] 44100 0.36 4666 0.10580498866213152 0.9997024\n",
      "2 [321 227 466 273] 6670 0.31724137931034485 3259 0.48860569715142427 0.99965084\n",
      "4 [243 388 279 452] 2304 1.7777777777777777 845 0.3667534722222222 0.99944323\n",
      "2 [384  83 589 180] 19885 0.47317073170731705 6746 0.3392506914759869 0.9986883\n",
      "3 [266 334 507 360] 6266 0.1078838174273859 2351 0.3751994893073731 0.9984402\n",
      "2 [303 700 464 768] 10948 0.422360248447205 3425 0.3128425283156741 0.9961888\n",
      "5 [256 320 360 367] 4888 0.4519230769230769 3260 0.6669394435351882 0.9959997\n",
      "4 [257  42 566 331] 89301 0.9352750809061489 8989 0.10065956708211554 0.98734456\n",
      "3 [248 480 561 500] 6260 0.06389776357827476 2982 0.47635782747603833 0.95994794\n",
      "7 [356   0 764 190] 77520 0.46568627450980393 50091 0.6461687306501548 0.95792055\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4928429126739502\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00034.jpeg\n",
      "4 [238 529 274 593] 2304 1.7777777777777777 964 0.4184027777777778 0.99999154\n",
      "4 [ 254  693  580 1011] 103668 0.9754601226993865 11327 0.10926226029247212 0.99998534\n",
      "2 [322 695 470 765] 10360 0.47297297297297297 2974 0.2870656370656371 0.9999852\n",
      "2 [403 807 601 899] 18216 0.46464646464646464 7565 0.4152942468159859 0.99985135\n",
      "3 [322 185 675 310] 44125 0.35410764872521244 4427 0.10032861189801699 0.99969757\n",
      "3 [238 383 525 426] 12341 0.14982578397212543 3185 0.2580828133862734 0.9996911\n",
      "4 [243 388 278 452] 2240 1.8285714285714285 860 0.38392857142857145 0.9996207\n",
      "3 [327 658 655 827] 55432 0.5152439024390244 5553 0.10017679318805023 0.9996125\n",
      "2 [320 226 467 272] 6762 0.3129251700680272 3251 0.48077491866311745 0.99951184\n",
      "3 [265 334 508 360] 6318 0.10699588477366255 2446 0.38714783159227606 0.9993024\n",
      "2 [384  83 589 180] 19885 0.47317073170731705 6855 0.34473221020870004 0.99856526\n",
      "5 [255 320 360 367] 4935 0.44761904761904764 3227 0.6539007092198581 0.99673814\n",
      "4 [257  43 564 332] 88723 0.9413680781758957 8989 0.10131532973411629 0.98572993\n",
      "7 [339   0 777 194] 84972 0.4429223744292237 52348 0.6160617615214423 0.98162526\n",
      "2 [ 340  815  686 1022] 71622 0.5982658959537572 36566 0.5105414537432632 0.95513755\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4698474407196045\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00035.jpeg\n",
      "4 [239 529 274 594] 2275 1.8571428571428572 987 0.4338461538461538 0.9999908\n",
      "4 [ 243  685  553 1020] 103850 1.0806451612903225 12028 0.11582089552238806 0.9998771\n",
      "3 [265 333 505 360] 6480 0.1125 2517 0.38842592592592595 0.9997818\n",
      "3 [326 680 668 819] 47538 0.4064327485380117 3853 0.08105094871471244 0.999767\n",
      "2 [400 806 603 896] 18270 0.4433497536945813 7220 0.39518336070060206 0.9997106\n",
      "3 [331 185 678 314] 44763 0.37175792507204614 4442 0.09923374215311753 0.9996586\n",
      "2 [321 226 467 273] 6862 0.3219178082191781 3281 0.478140483823958 0.99938285\n",
      "4 [243 388 279 452] 2304 1.7777777777777777 845 0.3667534722222222 0.999366\n",
      "3 [239 383 531 425] 12264 0.14383561643835616 3031 0.2471461187214612 0.9986192\n",
      "2 [413  87 593 184] 17460 0.5388888888888889 6309 0.361340206185567 0.997811\n",
      "2 [330 700 484 763] 9702 0.4090909090909091 2474 0.2549989692846836 0.99740404\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3188 0.6522094926350246 0.9967121\n",
      "4 [259  39 564 329] 88450 0.9508196721311475 9080 0.10265686828716789 0.98261523\n",
      "7 [332   2 748 215] 88608 0.5120192307692307 51253 0.578424070061394 0.9219425\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.47887253761291504\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00036.jpeg\n",
      "4 [238 529 274 593] 2304 1.7777777777777777 974 0.4227430555555556 0.9999851\n",
      "3 [316 682 666 814] 46200 0.37714285714285717 1838 0.03978354978354978 0.99998105\n",
      "2 [328 702 488 768] 10560 0.4125 2252 0.21325757575757576 0.99995565\n",
      "4 [ 249  683  545 1023] 100640 1.1486486486486487 12073 0.11996224165341812 0.99989367\n",
      "3 [261 334 520 359] 6475 0.09652509652509653 2412 0.3725096525096525 0.99977225\n",
      "4 [243 388 279 451] 2268 1.75 854 0.3765432098765432 0.9997043\n",
      "2 [400 808 604 896] 17952 0.43137254901960786 6955 0.3874220142602496 0.9996215\n",
      "3 [329 185 676 313] 44416 0.3688760806916426 4336 0.09762247838616715 0.9995734\n",
      "2 [322 227 467 273] 6670 0.31724137931034485 3259 0.48860569715142427 0.99953556\n",
      "2 [413  87 592 184] 17363 0.5418994413407822 6285 0.36197661694407646 0.99808717\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3197 0.654050736497545 0.9968093\n",
      "3 [236 382 529 427] 13185 0.15358361774744028 3088 0.2342055365946151 0.99375546\n",
      "7 [338   0 778 194] 85360 0.4409090909090909 52388 0.6137300843486411 0.9824638\n",
      "4 [260  41 561 329] 86688 0.9568106312292359 8996 0.10377445551864156 0.97858673\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4838430881500244\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00037.jpeg\n",
      "4 [238 530 274 594] 2304 1.7777777777777777 954 0.4140625 0.9999869\n",
      "4 [ 249  678  547 1024] 103108 1.1610738255033557 12606 0.12226015440121038 0.99993944\n",
      "2 [395 807 605 894] 18270 0.4142857142857143 7065 0.3866995073891626 0.9997948\n",
      "3 [261 334 518 359] 6425 0.09727626459143969 2493 0.38801556420233463 0.99971443\n",
      "2 [322 227 466 273] 6624 0.3194444444444444 3260 0.4921497584541063 0.9996007\n",
      "4 [243 387 279 452] 2340 1.8055555555555556 852 0.3641025641025641 0.9995881\n",
      "3 [327 185 680 310] 44125 0.35410764872521244 4406 0.09985269121813031 0.9995571\n",
      "3 [342 680 670 815] 44280 0.4115853658536585 2569 0.05801716350496838 0.9989955\n",
      "2 [384  83 589 179] 19680 0.4682926829268293 6773 0.34415650406504067 0.99833447\n",
      "3 [235 381 528 426] 13185 0.15358361774744028 3031 0.2298824421691316 0.9972512\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3190 0.6526186579378068 0.9964709\n",
      "2 [321 700 498 773] 12921 0.4124293785310734 2417 0.18705982509093724 0.9947849\n",
      "4 [262  39 560 324] 84930 0.9563758389261745 8998 0.10594607323678323 0.97695094\n",
      "7 [350   0 774 195] 82680 0.45990566037735847 51287 0.6203072085147557 0.9555095\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Time: 0.48484373092651367\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00038.jpeg\n",
      "4 [ 252  682  546 1024] 100548 1.163265306122449 12477 0.12408998687194175 0.9999918\n",
      "4 [238 530 274 596] 2376 1.8333333333333333 967 0.406986531986532 0.99998164\n",
      "3 [262 334 516 359] 6350 0.0984251968503937 2436 0.3836220472440945 0.99979407\n",
      "2 [394 806 601 895] 18423 0.42995169082125606 7204 0.3910329479455029 0.9997789\n",
      "4 [243 388 279 451] 2268 1.75 854 0.3765432098765432 0.99963427\n",
      "2 [322 227 466 273] 6624 0.3194444444444444 3251 0.49079106280193235 0.9995571\n",
      "3 [324 672 673 835] 56887 0.4670487106017192 3127 0.05496862200502751 0.9993223\n",
      "3 [326 185 684 311] 45108 0.35195530726256985 4329 0.09596967278531525 0.9992761\n",
      "3 [238 383 525 426] 12341 0.14982578397212543 3036 0.2460092375010129 0.99857426\n",
      "2 [385  83 589 180] 19788 0.47549019607843135 6774 0.34232868405093997 0.9985487\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3201 0.6548690671031097 0.9966011\n",
      "2 [322 700 482 767] 10720 0.41875 2444 0.22798507462686568 0.9946267\n",
      "4 [262  38 564 325] 86674 0.9503311258278145 9089 0.10486420379814015 0.9830856\n",
      "7 [339   0 766 190] 81130 0.4449648711943794 51714 0.6374214224084802 0.9709447\n",
      "7 [ 341  812  725 1024] 81408 0.5520833333333334 48412 0.59468356918239 0.9125658\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4678511619567871\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00039.jpeg\n",
      "4 [239 531 274 594] 2205 1.8 962 0.436281179138322 0.9999839\n",
      "4 [ 269  673  587 1024] 111618 1.1037735849056605 12907 0.11563547098138294 0.99989474\n",
      "3 [263 335 515 359] 6048 0.09523809523809523 2352 0.3888888888888889 0.9996755\n",
      "2 [322 227 466 273] 6624 0.3194444444444444 3254 0.491243961352657 0.9995894\n",
      "2 [394 806 594 895] 17800 0.445 6639 0.37297752808988766 0.9995635\n",
      "4 [243 387 279 451] 2304 1.7777777777777777 831 0.3606770833333333 0.9992079\n",
      "3 [327 184 683 311] 45212 0.35674157303370785 4280 0.09466513315049102 0.99902546\n",
      "3 [232 384 528 427] 12728 0.14527027027027026 3056 0.24010056568196103 0.99898094\n",
      "2 [413  86 592 184] 17542 0.547486033519553 6343 0.3615893284688177 0.9978016\n",
      "5 [256 320 360 367] 4888 0.4519230769230769 3199 0.6544599018003273 0.99562436\n",
      "4 [259  38 564 327] 88145 0.9475409836065574 9022 0.10235407567076975 0.98278975\n",
      "3 [305 678 676 815] 50827 0.3692722371967655 3235 0.06364727408660752 0.9816442\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4968400001525879\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00040.jpeg\n",
      "4 [ 256  680  544 1024] 99072 1.1944444444444444 12554 0.12671592377260982 0.999977\n",
      "4 [238 529 274 592] 2268 1.75 963 0.4246031746031746 0.9999548\n",
      "2 [320 227 466 272] 6570 0.3082191780821918 3219 0.4899543378995434 0.99966943\n",
      "2 [385  84 590 180] 19680 0.4682926829268293 6836 0.34735772357723577 0.99942446\n",
      "3 [260 333 513 361] 7084 0.11067193675889328 2486 0.35093167701863354 0.9992623\n",
      "3 [327 183 681 311] 45312 0.3615819209039548 4368 0.09639830508474577 0.99878746\n",
      "2 [395 807 592 895] 17336 0.4467005076142132 6375 0.3677318874019382 0.99873346\n",
      "3 [239 383 520 424] 11521 0.14590747330960854 3072 0.266643520527732 0.9979717\n",
      "4 [243 387 280 452] 2405 1.7567567567567568 845 0.35135135135135137 0.99769175\n",
      "5 [256 320 360 367] 4888 0.4519230769230769 3192 0.6530278232405892 0.99499595\n",
      "3 [320 682 672 820] 48576 0.39204545454545453 3978 0.08189229249011858 0.9927366\n",
      "4 [255  35 565 326] 90210 0.9387096774193548 9204 0.10202859993348852 0.9770264\n",
      "7 [324   9 746 199] 80180 0.45023696682464454 52486 0.6546021451733599 0.9175803\n",
      "7 [ 325  809  658 1024] 71595 0.6456456456456456 42626 0.5953767721209582 0.90038115\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.47284746170043945\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00041.jpeg\n",
      "4 [ 259  672  537 1010] 93964 1.2158273381294964 12705 0.135211357541186 0.9999558\n",
      "4 [239 529 274 593] 2240 1.8285714285714285 953 0.42544642857142856 0.9999554\n",
      "2 [320 227 466 272] 6570 0.3082191780821918 3250 0.4946727549467275 0.9997571\n",
      "3 [262 334 512 359] 6250 0.1 2382 0.38112 0.9997367\n",
      "3 [327 183 678 313] 45630 0.37037037037037035 4389 0.09618671926364233 0.99900514\n",
      "2 [413  86 592 184] 17542 0.547486033519553 6302 0.35925208072055637 0.99835134\n",
      "3 [304 682 682 815] 50274 0.35185185185185186 3990 0.07936507936507936 0.99824345\n",
      "2 [400 808 596 895] 17052 0.44387755102040816 5911 0.34664555477363357 0.99754924\n",
      "3 [231 381 542 429] 14928 0.15434083601286175 3344 0.2240085744908896 0.99741936\n",
      "4 [243 387 280 452] 2405 1.7567567567567568 840 0.3492723492723493 0.99592495\n",
      "5 [255 320 359 367] 4888 0.4519230769230769 3211 0.6569148936170213 0.9943982\n",
      "2 [312 700 467 768] 10540 0.43870967741935485 4767 0.45227703984819734 0.99000716\n",
      "4 [256  35 564 326] 89628 0.9448051948051948 9256 0.10327129914758781 0.98770356\n",
      "7 [356   0 780 184] 78016 0.4339622641509434 50695 0.6498026045939295 0.95735735\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.46884632110595703\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00042.jpeg\n",
      "2 [323 705 486 769] 10432 0.39263803680981596 4572 0.43826687116564417 0.9999918\n",
      "3 [263 333 516 361] 7084 0.11067193675889328 2426 0.3424618859401468 0.99977714\n",
      "4 [239 529 274 593] 2240 1.8285714285714285 874 0.39017857142857143 0.99955684\n",
      "2 [322 227 464 273] 6532 0.323943661971831 3238 0.4957134109001837 0.99954754\n",
      "3 [327 182 675 313] 45588 0.3764367816091954 4490 0.09849083092041766 0.9990476\n",
      "4 [ 245  667  536 1001] 97194 1.147766323024055 12026 0.1237319176080828 0.9988607\n",
      "2 [384  83 590 180] 19982 0.470873786407767 6811 0.3408567710939846 0.9987109\n",
      "2 [393 808 591 896] 17424 0.4444444444444444 6104 0.35032139577594124 0.99840754\n",
      "3 [237 382 530 426] 12892 0.15017064846416384 2971 0.23045299410487124 0.9975867\n",
      "5 [255 320 359 366] 4784 0.4423076923076923 3124 0.653010033444816 0.99526477\n",
      "4 [243 387 280 452] 2405 1.7567567567567568 811 0.33721413721413723 0.99330336\n",
      "3 [300 673 687 819] 56502 0.3772609819121447 4449 0.07874057555484762 0.9859117\n",
      "4 [256  34 566 328] 91140 0.9483870967741935 9235 0.10132762782532367 0.98022217\n",
      "7 [304   7 748 205] 87912 0.44594594594594594 53139 0.6044567294567295 0.95863223\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.45485568046569824\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00043.jpeg\n",
      "3 [235 383 532 425] 12474 0.1414141414141414 2940 0.2356902356902357 0.9997682\n",
      "3 [264 333 515 361] 7028 0.11155378486055777 2494 0.35486624928856003 0.99974936\n",
      "4 [239 529 275 591] 2232 1.7222222222222223 903 0.40456989247311825 0.9995517\n",
      "4 [ 260  693  586 1024] 107906 1.0153374233128833 11553 0.10706540878171741 0.9994498\n",
      "2 [324 227 465 273] 6486 0.3262411347517731 3212 0.4952204748689485 0.9993517\n",
      "2 [316 702 468 761] 8968 0.3881578947368421 3644 0.4063336306868867 0.99921155\n",
      "3 [326 180 674 314] 46632 0.3850574712643678 4402 0.09439869617430091 0.99864215\n",
      "2 [414  87 593 185] 17542 0.547486033519553 6279 0.35794094173982444 0.9980439\n",
      "2 [389 804 588 901] 19303 0.48743718592964824 6098 0.31590944412785577 0.99793255\n",
      "5 [255 320 360 366] 4830 0.4380952380952381 3161 0.6544513457556936 0.9953583\n",
      "4 [243 387 280 452] 2405 1.7567567567567568 787 0.32723492723492725 0.993321\n",
      "4 [255  35 568 326] 91083 0.9297124600638977 9049 0.09934894546732101 0.9756007\n",
      "7 [308   6 746 208] 88476 0.4611872146118721 52837 0.597190198471902 0.9723626\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.47884607315063477\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00044.jpeg\n",
      "4 [ 242  692  557 1021] 103635 1.0444444444444445 11926 0.11507695276692237 0.99994504\n",
      "3 [261 334 512 360] 6526 0.10358565737051793 2442 0.3741955255899479 0.9996331\n",
      "4 [239 530 275 591] 2196 1.6944444444444444 897 0.40846994535519127 0.9996275\n",
      "3 [244 384 539 428] 12980 0.14915254237288136 2862 0.2204930662557781 0.99956113\n",
      "2 [324 227 464 274] 6580 0.3357142857142857 3345 0.5083586626139818 0.9993818\n",
      "2 [319 701 476 765] 10048 0.40764331210191085 3936 0.39171974522292996 0.999226\n",
      "3 [324 181 671 314] 46151 0.38328530259365995 4705 0.10194795345712986 0.9982699\n",
      "2 [415  87 592 185] 17346 0.5536723163841808 6266 0.3612360198316615 0.9979462\n",
      "2 [387 805 573 899] 17484 0.5053763440860215 5149 0.2944978265843057 0.9962547\n",
      "5 [254 320 359 366] 4830 0.4380952380952381 3186 0.6596273291925466 0.9957793\n",
      "4 [243 387 280 454] 2479 1.8108108108108107 847 0.3416700282371924 0.99198955\n",
      "4 [288  13 575 303] 83230 1.0104529616724738 9437 0.11338459690015619 0.979673\n",
      "7 [304   7 749 211] 90780 0.4584269662921348 52396 0.5771755893368583 0.96648383\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.46485137939453125\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00045.jpeg\n",
      "4 [ 234  695  559 1018] 104975 0.9938461538461538 11607 0.11056918313884258 0.9999459\n",
      "4 [240 529 275 592] 2205 1.8 940 0.42630385487528344 0.9998971\n",
      "2 [327 701 462 770] 9315 0.5111111111111111 3766 0.40429414922168544 0.99989605\n",
      "3 [266 335 507 360] 6025 0.1037344398340249 2407 0.39950207468879667 0.99971706\n",
      "2 [325 227 465 274] 6580 0.3357142857142857 3326 0.5054711246200608 0.999413\n",
      "3 [247 384 531 426] 11928 0.14788732394366197 2983 0.250083836351442 0.9992079\n",
      "4 [243 389 279 454] 2340 1.8055555555555556 848 0.3623931623931624 0.99894804\n",
      "3 [326 181 671 313] 45540 0.3826086956521739 4603 0.10107597716293368 0.9983342\n",
      "2 [414  87 592 185] 17444 0.550561797752809 6263 0.35903462508598943 0.9982791\n",
      "5 [254 320 360 366] 4876 0.4339622641509434 3212 0.6587366694011485 0.99611133\n",
      "2 [395 805 578 900] 17385 0.5191256830601093 5026 0.2890997986770204 0.9959721\n",
      "4 [254  35 567 327] 91396 0.9329073482428115 9231 0.10100004376559149 0.9778881\n",
      "7 [300   7 752 211] 92208 0.45132743362831856 52547 0.5698746312684366 0.9713488\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.46382617950439453\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00046.jpeg\n",
      "4 [ 243  687  563 1024] 107840 1.053125 11079 0.10273553412462907 0.99998057\n",
      "4 [240 530 275 592] 2170 1.7714285714285714 943 0.4345622119815668 0.9999428\n",
      "4 [242 389 278 453] 2304 1.7777777777777777 890 0.3862847222222222 0.9996891\n",
      "2 [320 694 449 769] 9675 0.5813953488372093 2733 0.2824806201550388 0.9995852\n",
      "3 [260 334 508 360] 6448 0.10483870967741936 2487 0.3857009925558313 0.9993067\n",
      "2 [326 227 464 274] 6486 0.34057971014492755 3550 0.5473327166204132 0.99930656\n",
      "3 [239 383 535 426] 12728 0.14527027027027026 3049 0.23955059710873663 0.9992291\n",
      "2 [386 802 591 901] 20295 0.48292682926829267 6218 0.30638088199063807 0.9988858\n",
      "2 [413  87 593 184] 17460 0.5388888888888889 6350 0.36368843069873996 0.99843687\n",
      "3 [327 182 673 313] 45326 0.3786127167630058 4426 0.0976481489652738 0.9978452\n",
      "5 [256 320 359 368] 4944 0.46601941747572817 3175 0.6421925566343042 0.9970168\n",
      "4 [283  14 567 298] 80656 1.0 9725 0.12057379488196786 0.99180716\n",
      "7 [309   7 748 215] 91312 0.47380410022779046 52487 0.5748094445417908 0.9795574\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.47484707832336426\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00047.jpeg\n",
      "4 [240 530 275 590] 2100 1.7142857142857142 947 0.45095238095238094 0.99995506\n",
      "4 [ 244  687  593 1024] 117613 0.9656160458452722 10791 0.0917500616428456 0.9999299\n",
      "4 [242 389 278 452] 2268 1.75 929 0.4096119929453263 0.9997389\n",
      "2 [325 227 465 274] 6580 0.3357142857142857 3339 0.5074468085106383 0.9995258\n",
      "3 [261 334 506 360] 6370 0.10612244897959183 2458 0.3858712715855573 0.9994804\n",
      "3 [240 383 531 425] 12222 0.14432989690721648 3175 0.2597774504991 0.99938715\n",
      "2 [313 700 446 765] 8645 0.48872180451127817 3049 0.35268941584731056 0.99868757\n",
      "2 [413  87 593 184] 17460 0.5388888888888889 6357 0.3640893470790378 0.9985592\n",
      "3 [327 181 674 313] 45804 0.3804034582132565 4443 0.09700026198585276 0.998156\n",
      "2 [390 804 584 901] 18818 0.5 5674 0.30151982144755024 0.99761474\n",
      "5 [254 320 359 367] 4935 0.44761904761904764 3226 0.6536980749746707 0.9970497\n",
      "4 [261  39 561 321] 84600 0.94 9005 0.10644208037825059 0.9916277\n",
      "7 [301   4 746 209] 91225 0.4606741573033708 52502 0.5755220608385859 0.9785041\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.5058157444000244\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00048.jpeg\n",
      "4 [ 258  690  570 1024] 104208 1.0705128205128205 11436 0.1097420543528328 0.9999902\n",
      "2 [319 705 449 759] 7020 0.4153846153846154 3041 0.43319088319088317 0.999948\n",
      "3 [266 333 507 358] 6025 0.1037344398340249 2125 0.35269709543568467 0.99987984\n",
      "2 [320 228 467 273] 6615 0.30612244897959184 3069 0.4639455782312925 0.99981636\n",
      "3 [239 385 534 424] 11505 0.13220338983050847 2897 0.25180356366797046 0.9995957\n",
      "2 [393 807 592 894] 17313 0.4371859296482412 5617 0.3244382833708774 0.99945086\n",
      "4 [242 387 279 451] 2368 1.7297297297297298 956 0.40371621621621623 0.9992261\n",
      "4 [240 531 276 592] 2196 1.6944444444444444 958 0.43624772313296906 0.99904317\n",
      "5 [259 318 358 365] 4653 0.47474747474747475 3135 0.6737588652482269 0.99888295\n",
      "3 [323 183 673 316] 46550 0.38 5127 0.11013963480128894 0.9987129\n",
      "2 [412  90 593 185] 17195 0.5248618784530387 6267 0.36446641465542307 0.99636734\n",
      "4 [264  35 561 319] 84348 0.9562289562289562 8432 0.09996680419215631 0.99565244\n",
      "3 [269 629 688 799] 71230 0.40572792362768495 4230 0.05938509055173382 0.9728918\n",
      "7 [302  11 758 222] 96216 0.46271929824561403 51884 0.5392450320113079 0.9688668\n",
      "7 [ 324  814  696 1024] 78120 0.5645161290322581 39767 0.5090501792114696 0.9031895\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4718482494354248\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00049.jpeg\n",
      "4 [ 257  680  578 1022] 109782 1.0654205607476634 10342 0.09420487875972382 0.99999845\n",
      "3 [267 333 509 358] 6050 0.10330578512396695 2131 0.35223140495867766 0.9999138\n",
      "2 [320 228 467 273] 6615 0.30612244897959184 3081 0.4657596371882086 0.99983895\n",
      "4 [240 532 276 591] 2124 1.6388888888888888 943 0.4439736346516008 0.9997758\n",
      "4 [242 387 277 452] 2275 1.8571428571428572 958 0.4210989010989011 0.9995777\n",
      "2 [395 807 600 894] 17835 0.424390243902439 6348 0.35592935239697227 0.99933714\n",
      "3 [286 631 659 807] 65648 0.4718498659517426 5195 0.07913417011942481 0.99915874\n",
      "5 [259 318 357 365] 4606 0.47959183673469385 3029 0.6576204950065132 0.99887913\n",
      "3 [235 383 532 426] 12771 0.1447811447811448 2996 0.2345940020358625 0.9987753\n",
      "3 [322 182 672 316] 46900 0.38285714285714284 5211 0.11110874200426439 0.99873704\n",
      "4 [267  35 558 319] 82644 0.9759450171821306 8348 0.10101156768791443 0.9963558\n",
      "2 [412  90 593 185] 17195 0.5248618784530387 6276 0.3649898226228555 0.995611\n",
      "2 [317 703 464 767] 9408 0.43537414965986393 2388 0.2538265306122449 0.9931376\n",
      "7 [298  11 756 220] 95722 0.45633187772925765 51486 0.537870082112785 0.96832585\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4738478660583496\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00050.jpeg\n",
      "4 [ 272  687  587 1024] 106155 1.0698412698412698 9211 0.08676934670999953 0.99998987\n",
      "3 [267 333 508 358] 6025 0.1037344398340249 2158 0.35817427385892114 0.9999082\n",
      "2 [319 228 466 273] 6615 0.30612244897959184 3128 0.47286470143613 0.9998685\n",
      "4 [243 387 278 453] 2310 1.8857142857142857 982 0.4251082251082251 0.9995216\n",
      "4 [240 531 276 593] 2232 1.7222222222222223 945 0.42338709677419356 0.9990004\n",
      "3 [326 180 673 315] 46845 0.38904899135446686 5064 0.10810118475824528 0.9989649\n",
      "5 [259 318 358 365] 4653 0.47474747474747475 3063 0.6582849774339136 0.9989052\n",
      "3 [247 386 549 424] 11476 0.12582781456953643 2873 0.2503485535029627 0.99842036\n",
      "2 [411  90 593 185] 17290 0.521978021978022 6313 0.36512434933487564 0.99655473\n",
      "2 [400 806 589 892] 16254 0.455026455026455 5853 0.36009597637504615 0.99641716\n",
      "4 [267  36 559 318] 82344 0.9657534246575342 8282 0.10057806276109978 0.9943796\n",
      "7 [313   4 752 212] 91312 0.47380410022779046 51675 0.5659168564920274 0.9887377\n",
      "3 [284 640 672 830] 73720 0.4896907216494845 4641 0.06295442213781878 0.95070636\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4488561153411865\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00051.jpeg\n",
      "4 [ 268  683  563 1024] 100595 1.1559322033898305 10588 0.10525374024553905 0.99999833\n",
      "3 [268 333 508 358] 6000 0.10416666666666667 2172 0.362 0.9999151\n",
      "2 [320 228 465 273] 6525 0.3103448275862069 3090 0.4735632183908046 0.9998691\n",
      "4 [243 387 279 451] 2304 1.7777777777777777 974 0.4227430555555556 0.9997285\n",
      "3 [245 385 535 425] 11600 0.13793103448275862 3083 0.26577586206896553 0.9995486\n",
      "4 [239 531 277 592] 2318 1.605263157894737 980 0.4227782571182053 0.99912554\n",
      "3 [326 180 674 315] 46980 0.3879310344827586 5168 0.11000425713069391 0.999008\n",
      "5 [259 318 359 366] 4800 0.48 3109 0.6477083333333333 0.998978\n",
      "2 [412  90 593 185] 17195 0.5248618784530387 6324 0.3677813317824949 0.99655765\n",
      "4 [268  35 561 318] 82919 0.9658703071672355 8276 0.09980824660210567 0.9942053\n",
      "2 [399 808 591 891] 15936 0.4322916666666667 5773 0.36226154618473894 0.99390936\n",
      "7 [316  11 758 220] 92378 0.47285067873303166 51353 0.555900755591158 0.9877266\n",
      "3 [293 643 670 832] 71253 0.5013262599469496 3716 0.05215219008322457 0.9462741\n",
      "Processing 1 images\n",
      "image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000  uint8\n",
      "molded_images            shape: (1, 1024, 1024, 3)    min:    0.00000  max:  255.00000  uint8\n",
      "image_metas              shape: (1, 20)               min:    0.00000  max: 1280.00000  float64\n",
      "anchors                  shape: (1, 261888, 4)        min:   -0.35390  max:    1.29134  float32\n",
      "Time: 0.4478590488433838\n",
      "D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\val\\scene00052.jpeg\n",
      "4 [ 281  688  577 1024] 99456 1.135135135135135 11380 0.11442245817245818 0.9999956\n",
      "3 [269 333 506 358] 5925 0.10548523206751055 2139 0.3610126582278481 0.99991095\n",
      "2 [318 227 465 273] 6762 0.3129251700680272 3127 0.4624371487725525 0.99989486\n",
      "3 [242 385 532 425] 11600 0.13793103448275862 3246 0.2798275862068966 0.9997327\n",
      "4 [243 387 278 449] 2170 1.7714285714285714 915 0.4216589861751152 0.999724\n",
      "3 [281 649 670 822] 67297 0.44473007712082263 3134 0.04656968364117271 0.9995171\n",
      "3 [333 183 674 310] 43307 0.3724340175953079 5053 0.11667859699355762 0.9995085\n",
      "5 [260 318 359 366] 4752 0.48484848484848486 3069 0.6458333333333334 0.99880254\n",
      "2 [411  90 593 185] 17290 0.521978021978022 6328 0.3659919028340081 0.9966589\n",
      "4 [240 531 278 595] 2432 1.6842105263157894 961 0.3951480263157895 0.99663585\n",
      "2 [399 809 599 892] 16600 0.415 6320 0.38072289156626504 0.9955064\n",
      "4 [266  34 562 317] 83768 0.956081081081081 8340 0.09956069143348295 0.99192655\n",
      "7 [316  11 759 221] 93030 0.47404063205417607 51589 0.554541545737934 0.988139\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "config.RPN_ANCHOR_RATIOS = [0.5, 1, 2]\n",
    "config.RPN_ANCHOR_SCALES = (32, 64, 128, 256, 512)  #调整检测使用的参数\n",
    "#config.RPN_ANCHOR_SCALES = (0, 0, 0, 300, 0)  #调整检测使用的参数\n",
    "model = modellib.MaskRCNN(mode=\"inference\", config=config,\n",
    "                                model_dir=\"D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\logs\\mask20201217T2051\\\\\")\n",
    "model.load_weights(\"D:\\Projects\\Mask_RCNN\\my_mask_rcnn\\dataset_1119\\logs\\mask20201217T2051\\mask_rcnn_mask_0500.h5\", by_name=True)\n",
    "#detect_targets = []\n",
    "for image_id in dataset_val.image_ids:\n",
    "    image, image_meta, gt_class_id, gt_box, gt_mask = modellib.load_image_gt(dataset_val, config, image_id, use_mini_mask=False)\n",
    "    # print(gt_box)\n",
    "    start_time = time.time()\n",
    "    results = model.detect_molded(np.expand_dims(image, 0), np.expand_dims(image_meta, 0), verbose=1)\n",
    "    end_time = time.time()\n",
    "    print(\"Time: %s\" % str(end_time - start_time))\n",
    "    # print(results)\n",
    "    r = results[0]\n",
    "    target = [x for x in gt_targets if x['file'] == dataset_val.image_reference(image_id)]\n",
    "    '''\n",
    "    target = {\n",
    "        'file' : dataset_val.image_reference(image_id),\n",
    "        'objs' : []\n",
    "    }\n",
    "    '''\n",
    "    print(dataset_val.image_reference(image_id))\n",
    "    for i in range(0,len(r['class_ids'])):\n",
    "        box_height = r['rois'][i][2] - r['rois'][i][0]\n",
    "        box_width = r['rois'][i][3] - r['rois'][i][1]\n",
    "        box_area = box_height * box_width\n",
    "        box_ratio = box_width / box_height\n",
    "        mask_area = np.sum(r['masks'][:,:,i]==True)\n",
    "        mask_ratio = mask_area / box_area\n",
    "        obj = {\n",
    "            'class_id' : r['class_ids'][i],\n",
    "            'bbox' : r['rois'][i],\n",
    "            'bbox_area' : box_area,\n",
    "            'wh_ratio' : box_ratio,\n",
    "            'mask_area' : mask_area,\n",
    "            'mask_ratio' : mask_ratio,\n",
    "            'score' : r['scores'][i],\n",
    "            'size' : (obj['bbox'][3] - obj['bbox'][1]) / math.sqrt(obj['wh_ratio'])\n",
    "        }\n",
    "        target[0]['objs'].append(obj)\n",
    "        print(r['class_ids'][i], r['rois'][i], box_area, box_ratio, mask_area, mask_ratio, r['scores'][i])\n",
    "#    detect_targets.append(target)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 目标框\n",
    "\n",
    "计算GT的bbox与检测的bbox的IoU。\n",
    "\n",
    "2个目标怎么匹配？IoU最大的那个"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[6720, 12149, 873, 7608, 5424, 4799, 60741, 3790, 2452, 3425, 961, 4743, 6210, 3268, 10062, 51890]\n",
      "[5563, 13409, 1166, 8716, 5445, 4294, 62338, 3385, 2280, 3120, 1230, 5989, 7705, 3364, 12823, 56305]\n",
      "[3, 4, 4, 2, 3, 2, 7, 3, 3, 2, 4, 3, 2, 5, 4, 7]\n",
      "[6516, 11178, 841, 7600, 59069, 2445, 3552, 4228, 3222, 6771, 5083, 4524, 956, 3251, 8965, 51523]\n",
      "[5994, 12974, 990, 8748, 63565, 2780, 4029, 4766, 3641, 8484, 6936, 6267, 1223, 3584, 14036, 57073]\n",
      "[3, 4, 4, 2, 7, 3, 3, 2, 2, 2, 3, 3, 4, 5, 4, 7]\n",
      "[851, 12547, 4572, 7788, 958, 3450, 55618, 3241, 2501, 4718, 4673, 6259, 3245, 3519, 51693, 10610]\n",
      "[1104, 14269, 6803, 8850, 1122, 3967, 63430, 3385, 2312, 6077, 5828, 7915, 3178, 4573, 55900, 12612]\n",
      "[4, 4, 3, 2, 4, 3, 7, 2, 3, 3, 3, 2, 5, 2, 7, 4]\n",
      "[7756, 891, 12644, 4070, 3256, 3274, 4767, 6696, 2500, 863, 3256, 4306, 58460, 3765, 51940]\n",
      "[8645, 1134, 13690, 5954, 4705, 3661, 6567, 7967, 2409, 1194, 3407, 7503, 62427, 4696, 55653]\n",
      "[2, 4, 4, 3, 3, 2, 3, 2, 3, 4, 5, 3, 7, 2, 7]\n",
      "[4892, 11499, 864, 7652, 4534, 3172, 3222, 878, 3592, 6663, 2441, 3212, 52516, 3608, 55701, 8906]\n",
      "[6557, 14391, 1193, 9083, 7504, 4974, 3209, 1179, 4919, 8258, 2703, 3116, 56676, 5023, 62002, 12499]\n",
      "[3, 4, 4, 2, 3, 3, 2, 4, 2, 2, 3, 5, 7, 3, 7, 4]\n",
      "[12587, 4107, 7559, 801, 3076, 4666, 3259, 845, 6746, 2351, 3425, 3260, 8989, 2982, 50091]\n",
      "[13169, 6474, 8550, 1367, 4016, 6805, 3593, 1137, 8333, 2538, 4605, 3333, 12490, 6158, 56315]\n",
      "[4, 3, 2, 4, 3, 3, 2, 4, 2, 3, 2, 5, 4, 3, 7]\n",
      "[964, 11327, 2974, 7565, 4427, 3185, 860, 5553, 3251, 2446, 6855, 3227, 8989, 52348, 36566]\n",
      "[1164, 15051, 4341, 8462, 7730, 3886, 1324, 7244, 3714, 2748, 8397, 3700, 13941, 55781, 62423]\n",
      "[4, 4, 2, 2, 3, 3, 4, 3, 2, 3, 2, 5, 4, 7, 2]\n",
      "[987, 12028, 2517, 3853, 7220, 4442, 3281, 845, 3031, 6309, 2474, 3188, 9080, 51253]\n",
      "[1145, 13986, 2890, 8246, 8504, 6994, 3612, 1157, 3873, 8348, 4965, 3103, 12913, 57203]\n",
      "[4, 4, 3, 3, 2, 3, 2, 4, 3, 2, 2, 5, 4, 7]\n",
      "[974, 1838, 2252, 12073, 2412, 854, 6955, 4336, 3259, 6285, 3197, 3088, 52388, 8996]\n",
      "[988, 6327, 4926, 14475, 2720, 1221, 8975, 6001, 3673, 8221, 3525, 3017, 56813, 13657]\n",
      "[4, 3, 2, 4, 3, 4, 2, 3, 2, 2, 5, 3, 7, 4]\n",
      "[954, 12606, 7065, 2493, 3260, 852, 4406, 2569, 6773, 3031, 3190, 2417, 8998, 51287]\n",
      "[1240, 14271, 8961, 2900, 3877, 1262, 6269, 6215, 8970, 3608, 3523, 4688, 14401, 56313]\n",
      "[4, 4, 2, 3, 2, 4, 3, 3, 2, 3, 5, 2, 4, 7]\n",
      "[12477, 967, 2436, 7204, 854, 3251, 3127, 4329, 3036, 6774, 3201, 2444, 9089, 51714, 48412]\n",
      "[14004, 1060, 2804, 8631, 1094, 3426, 5946, 5858, 3352, 8022, 3452, 4629, 13887, 55232, 62971]\n",
      "[4, 4, 3, 2, 4, 2, 3, 3, 3, 2, 5, 2, 4, 7, 7]\n",
      "[962, 12907, 2352, 3254, 6639, 831, 4280, 3056, 6343, 3199, 9022, 3235]\n",
      "[1132, 15315, 2503, 3579, 8715, 1181, 6150, 2906, 7767, 3412, 13892, 7321]\n",
      "[4, 4, 3, 2, 2, 4, 3, 3, 2, 5, 4, 3]\n",
      "[12554, 963, 3219, 6836, 2486, 4368, 6375, 3072, 845, 3192, 3978, 9204, 52486, 42626]\n",
      "[14496, 1029, 3806, 8485, 3021, 6555, 8776, 2927, 1217, 3367, 14496, 13798, 56180, 64274]\n",
      "[4, 4, 2, 2, 3, 3, 2, 3, 4, 5, 3, 4, 7, 7]\n",
      "[12705, 953, 3250, 2382, 4389, 6302, 3990, 5911, 3344, 840, 3211, 4767, 9256, 50695]\n",
      "[15364, 1064, 3788, 3117, 6716, 7967, 5367, 8971, 3658, 1163, 3268, 4581, 14340, 56484]\n",
      "[4, 4, 2, 3, 3, 2, 3, 2, 3, 4, 5, 2, 4, 7]\n",
      "[4572, 2426, 874, 3238, 4490, 12026, 6811, 6104, 2971, 3124, 811, 4449, 9235, 53139]\n",
      "[5181, 2777, 1143, 3850, 7148, 14950, 8439, 9019, 3945, 3547, 1188, 7482, 14165, 56691]\n",
      "[2, 3, 4, 2, 3, 4, 2, 2, 3, 5, 4, 3, 4, 7]\n",
      "[2940, 2494, 903, 11553, 3212, 3644, 4402, 6279, 6098, 3161, 787, 9049, 52837]\n",
      "[3842, 2908, 1235, 15510, 3491, 4670, 7227, 8500, 8921, 3542, 1243, 15077, 56338]\n",
      "[3, 3, 4, 4, 2, 2, 3, 2, 2, 5, 4, 4, 7]\n",
      "[11926, 2442, 897, 2862, 3345, 3936, 4705, 6266, 5149, 3186, 847, 9437, 52396]\n",
      "[14976, 2479, 1183, 4007, 3626, 4846, 6344, 8840, 8663, 3172, 1220, 13200, 56697]\n",
      "[4, 3, 4, 3, 2, 2, 3, 2, 2, 5, 4, 4, 7]\n",
      "[11607, 940, 3766, 2407, 3326, 2983, 848, 4603, 6263, 3212, 5026, 9231, 52547]\n",
      "[15097, 1028, 4931, 2717, 3465, 4112, 1264, 6332, 7896, 3395, 9090, 13135, 13135]\n",
      "[4, 4, 2, 3, 2, 3, 4, 3, 2, 5, 2, 4, 7]\n",
      "[11079, 943, 890, 2733, 2487, 3550, 3049, 6218, 6350, 4426, 3175, 9725, 52487]\n",
      "[15672, 1021, 1270, 4910, 2828, 3503, 3640, 8975, 8202, 6096, 3177, 13248, 56204]\n",
      "[4, 4, 4, 2, 3, 2, 3, 2, 2, 3, 5, 4, 7]\n",
      "[947, 10791, 929, 3339, 2458, 3175, 3049, 6357, 4443, 5674, 3226, 9005, 52502]\n",
      "[1091, 14985, 1160, 3809, 3184, 3921, 5091, 8341, 2171, 9461, 3378, 14603, 55696]\n",
      "[4, 4, 4, 2, 3, 3, 2, 2, 3, 2, 5, 4, 7]\n",
      "[11436, 3041, 2125, 3069, 2897, 5617, 956, 958, 3135, 5127, 6267, 8432, 4230, 51884, 39767]\n",
      "[15944, 5012, 2786, 3591, 4577, 9241, 1335, 1056, 3707, 8238, 7796, 12837, 7348, 55671, 62891]\n",
      "[4, 2, 3, 2, 3, 2, 4, 4, 5, 3, 2, 4, 3, 7, 7]\n",
      "[10342, 2131, 3081, 943, 958, 6348, 5195, 3029, 2996, 5211, 8348, 6276, 2388, 51486]\n",
      "[14658, 3177, 4010, 1175, 1297, 9335, 7220, 3637, 3552, 7379, 13424, 8157, 4687, 57471]\n",
      "[4, 3, 2, 4, 4, 2, 3, 5, 3, 3, 4, 2, 2, 7]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\ps\\anaconda3\\envs\\mask_rcnn\\lib\\site-packages\\ipykernel_launcher.py:29: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[9211, 2158, 3128, 982, 945, 5064, 3063, 2873, 6313, 5853, 8282, 51675, 4641]\n",
      "[15926, 2881, 3528, 1219, 1317, 6636, 3805, 4113, 8606, 9545, 14644, 57555, 7182]\n",
      "[4, 3, 2, 4, 4, 3, 5, 3, 2, 2, 4, 7, 3]\n",
      "[10588, 2172, 3090, 974, 3083, 980, 5168, 3109, 6324, 8276, 5773, 51353, 3716]\n",
      "[14844, 2728, 4026, 1176, 3888, 1268, 7223, 3603, 8619, 14880, 9528, 57052, 6255]\n",
      "[4, 3, 2, 4, 3, 4, 3, 5, 2, 4, 2, 7, 3]\n",
      "[11380, 2139, 3127, 3246, 915, 3134, 5053, 3069, 6328, 961, 6320, 8340, 51589]\n",
      "[14386, 2854, 3549, 4680, 1138, 7956, 7147, 3320, 8362, 1105, 9630, 13366, 56382]\n",
      "[4, 3, 2, 3, 4, 3, 3, 5, 2, 4, 2, 4, 7]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdYAAAEYCAYAAADh+NeeAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAspklEQVR4nO3de/xUVb3/8ddbQEEFFERDQLGkUlETED3Hy7HIpDK11KK8YGmoWenPblqdssxztFNmdtLyUqDmhWNe0DQPx0tpIYqXUlSKFOUbqIioeMGEPr8/1hrZDPOd79cve77f78D7+XjMY/asvdeatdfsvT97rb1nRhGBmZmZlWO9rq6AmZnZ2sSB1czMrEQOrGZmZiVyYDUzMyuRA6uZmVmJHFjNzMxKtFYFVkk3S5rY1fVoFEn7SGrp6nrYqiSFpG27uh4d4W2q60k6TdJlXV2PtYmklyW9vavef40Dq6R5kl6TtFTSC5L+KOk4SauVLelCScdUpU2W9I/cEM9Lmi7p3R2pS0R8MCKmdHRdcn3atZFLOqx6OUlH5YPs2VXpB+X0yWtSt47IB85/5vZdKmmOpE9XLTNW0k3583te0j2VZXL+kPTTqjx3SToqT1fW+ytVy7Tk/N+Q9L0Gr6pVKbvd8+d8V1nllfE+kt4r6XZJL0qaV2P+6ZIekrRc0mlV8wZLmiZpQd5+h1fNHyLp+rxPtEg6LqcPlTS/wyvY+ro0pNwySRog6VpJr0h6UtKnquaPk/SYpFfz57J1YZ4knSVpcX58X5IK82+XtEjSS5L+JOnAqrK/IOmJPH+WpD1bq2dEbBwRj5e57m9FWT3Wj0REX2Br4Ezga8DFNZa7CfhwjfTvR8TGwFDgWWBy9QL5Q+lOPezfAvvWqNPfgE9I6llIOxL4S6fVbHULcvv2A/4fcKGkdwFI+hfgNuB3wLbAQOB44IOF/K8AR1YfeKo8D3xNUr8a81r73A2o2lbKtC60+yvAL4CvtDJ/LvBV4Dc15v2TtB8f3Erey4AngC1I7fgfkt4bES3A85J2XpOKV2tUuSX7KfAPUpscBpwvaQcASZsB1wD/DgwAZgFXFfJOAg4CdgZ2AvYHji3MPxEYHBH98rKXSRqcy96NFFsOAfqT4su1kno0ZC3XVESs0QOYB7y/Km0saaMdWZW+MfAcsH4hbTLwvcLrDwMv5+k7gDOAPwCvkQ78/wrcC7yYn/+1kPcO4JjC688AjwJLgFuArQvzdgCmkwLCM8DXgfGkjeYN4GXgT3nZo4DHgaWkHe2wnD4D+JdCmUcBd5F21g/ntAHA08B/AZMLy/5PTn8R+D2wQ2Heh4BH8vv9HfhyTt8HaCks98W83NA6n88qeXLas8Chefou4Kdt5Qd+AvyykH4XcFTVet8AfLuwTAuwT2F6y8K803IbXJbX8yHgncCpuX7zgQ/kZQ8F7quq15eA69rYNicD5wE358/zD8DbgHPyNvEYsEth+VNIJ0ZLc7t+tDBvW9LJx4ukbfiqwrwAts3Te+a6v7eNugVwAvBX4ImcdiDwIPBSrsf4wnZ9eq7/UuB/gc3yvOG5rInAU7lu36j6DIrt3ie3y5K8jl+p2qZqtgGwHbAMWJHb8oXC/vpArvN84LRCWb3z57sYeIG0v26R51UOjgtJ2/j3gB6tvU87jkPvB+bVmX9ZsW5V83rmNhxedawKYFAh7QLg0jz9n8CphXm/BT5fVe6fgI/l6R/n9nkJuA/Yq2pfuKyVcuu14QDgl8CC/HleV8i3f96WXgD+COxUmDcP+DLwZ9L2fBXQu628wEak4+M7C8teCpyZpycBfyzM24h03H53fv1HYFJh/tHA3a18JmPzdjA2v/4EcE9V2UEKxK3tX5V9cjLwM9LxfilpP966sOy7WRkL5gAfL8wbSDquvZTb/nvAXW1uj+3ZaNvYoOdRFVhz+lOkns+ngD8X0v8PGFd4PZkcWEkb8+XAnYUDylOkINiTdJa0BDgiv/5kfj2wsPwxefog0tnqdnnZb1Y+dKAvaYf+EmnD7QvsVr2RFz7Al4B35deDyUEwl3l6YdmjSAHmU+QDL/A54Of5AykG1s/k992AdKB/sDBvIXnHAzYFRuXpfcgHQdJZ4f0UdvxWPp9invWAA0gnPbsAG5IOYK0GAVYG1rdVtUOtwPoe0s44IKcXA+tFwNFVB5NlwH7587mEdNLyDaAX8FlWBpwNSBv9doX8DwAHt7Huk0mBZnT+nG/L73Ek6SD+PeD2wvKHAlvmdvoEqTc0OM+7ItdtvVzWntU7cV6X+eSDQRt1C9LOPIAU7MaSDnL75vcYwsoD0h2kYPfOvOwdrDyYDc9lXZjn7Qy8XmmrGu1+JnBnft9hwMOsGljrtcFRVB1U8vaxY15+J9JJ6kF53rGkg9KGub1HA/3yvOtI+8VGwObAPcCxrb1PO9qz7MDaN6dtXki7EHggT+9JPk7l10cCfyi83p60L2yQXx9OOkj3JB13niYHM1YNrNXl1mvD35CC4qakfebfcvoo0snpbjnPRNJxulKXebm9t8zbwaPAcW3lJR0zXqtquy8DN+TpHwPnV81/mLyfkrbv3QrzxgBLq5a/kXRcCNLJyno5vR/phKRSry+QjgGqs38VA+tSYO+8Hj+ubF+k7W8+8On82YwiHTMqx/gr82PD/JnOp4sD690UzpwL6ScDZ1cd/JbljfBpYBrwjsIB5buFZY+gcNaS02aw8gB/BysD682sekBZD3iVNFz9SfIOUqN+p7F6YH2BNFzUp2rZUcD9hddHkQJMH9IBpn9uhz2oCqxV5WySN4T++fVTpB2qX9Vy+5DO7s/O79O/HZ/PPqRA+gLpgLsCOCnPG5Lf991t5K8E5u+z8oRhtcCap6cCZ+XpYmD9GHBNVTtPL7z+CKmH0iO/rhzYNsmvzwfOyNM7kE6oNmhj3ScDFxZefwF4tPB6R+r0iEhn7Qfm6UtIPZbVRgdyPU8FngR2bOd+E8D7Cq9/DvyolWXvAL5ZeP054Ld5engua2hh/j3AhFba/XFyTzi/nkTViEadNnjzc66z/DmV9SCdPK7SW8rpW+RtsU8h7ZPkk5z2vE+N9y01sBa28Z+QTqRGkXs0eV4PUgDatLC9vkLuCZFG2n5Rpz5LgJ0L+8JlrZTbWhsOJu3Xm9Yo+3wKJ/w5bQ4rA+884PDCvO8DP2srL7AX8HTVvM8Cd+Tpi8knfIX5f2DlcWIFhWMNMCK3u6ry9CJdivp/hTSRRhXfAJaTgt+ubexfxcB6ZWHexrkuw0gnj3dW5f058O38WbxB7kzkee3qsTbymuUQ0oZY7SbSUGfRDyJik4h4W0QcEBF/K8wrXszfknTwKnoyv1e1rYEf5xtyXsh1UV52GKkH0KaIeIXU+McBCyX9pnBz1QPAFpXrAIU8r5HOJr9JGrL7Q3G+pB6SzpT0N0kvkTZ0gM3y88GkNnpS0u/yddCKTUgHw/+MiBfbsw6ka6ybkM76zgXel9OXkHbOwa3kq3YWsF8b14C+BRwv6W1V6dOBvSX1KqQ9U5h+DXguIlYUXkPaCQCmAJ/KNzscAUyNiNfbUefq96h+XSkfSUdKerCwzYxk5WfyVdL2c4+k2ZI+U/U+J+U6PdSOOlUUt+22tsmnC9OvFuvdxvzqdt+y6n1X2Z/aaIPVSNqtcNPJi6T9pLL8paRLMFfmG4S+n+uxNengubDwPj8n9Vy7k8OAbUjtdT7wK9LJInk7vY00SkFELCXt8xNy3gl5eQAkfUnSo/kmqxdIJ92rtWt1ubTehsOA5yNiSY16bw18qdK2+f2GkT77ita2l3p5XyYdQ4r6kXqDdGB+P9Jlv6hqgzci4mbSseaAnHwM6SRjB2B90gjAjZKK61TPm9t8RLxMigdb5vXdrWp9DyON0A0inXTNr1VOPQ0JrJJ2JQWw1e7si4jHgF5q/63QxUZfQGqIoq1Ivbhq80lDS5sUHn0i4o953jva8X6VOt8SEfuSAtBjpCEh8gZxC6ve6FNxCWnI59Ia8z5Fup72ftIONjynK5d7b0QcSDrQXEfqBVYsIV0D+aWkPVpZh9orlgLR14AdJR0UEa+Sevyt3bxRnX8xqUdyep1lHiPdwPD1qvSlpGs6e72VOhfy3026vrMXqf1qtWuH5bsXLwQ+T7q0sAlpGKvymTwdEZ+NiC1JownnadWv2BwKHCTppLfwtsVtrd422WE12n0h6UBZsVVloq02oMa+Qbp0Mw0YFhH9SdeyKm32RkR8JyK2J90bsT9pyHQ+qce6WWHf7BcRO9R5n04XEU9GxP4RMSgidiMN5d5TWKS6k3AF8Ml8ItwHuB1A0l6k/e7jpB7mJqRhUVHbm+W20YYDJG1SI/980uhO8di3YURc0Y7Vrpf3L0BPSSMKy+8MzM7Ts/Nr8npvRNqma86vyltLT1buEzuThpz/EhH/jIjfkrblf23HOkFhm5e0MWkIfEFe399Vre/GEXE8sIjUOx5aq5x6Sg2skvpJ2p80Jn1ZnbP3Wr3W9rgJeKekT0nqKekTpHHvG2ss+zPg1MIda/0lHZrn3Qi8TdJJkjaQ1DffdQapRzO8crevpC0kHZA3ktdJZ10rCu/zm1bW5Xek62U/qTGvby5rMWns/j8qMyStr/RVnv4R8Qbpumbx/YiIO0hnVdcW6t0uEfEP4IekniWknthRkr4iaWCuw86SrmyliLNJG/N2dd7mO6RrFptUpbfWVu11CfDfwPKIKPtrH5WbIRYBKH3daGRlpqRDJVV2sCV52eLnsgAYB3xR0uc68P4XA59W+rrCekpf9ejQ185qKLb7VNJ+sWleny8UlqvbBqR9Y6ik9QtpfUk9p2WSxpJOesj53ytpx3zn5kukYbUVEbGQdAPWD/MxYz1J75D0b3Xep6actzepByxJvYv5JPXK89cjBYXeKtxJmudtkF9ukF9X5m2Xjw3rSzoc+ABp+6+4GfiAVn4z4CbSif93SZdM/lloo+W5XXtK+har9+yK3iy3jTa8mXSCt2lez71z/guB4/JogiRtJOnDkvq21Z718ubRu2uA7+b0PUgdhMpJ7rXASEkH53b8Fun+msfy/EuAk/O2vSWp4zE5t/W7JX1QUp+8LoeTron+Lue9F/iwpLfneu1Luufg4XasE8CHJO2Zt43TgZkRMZ8UC94p6Yj8vr0k7Sppuzx6cA1wmqQN8/54ZHverKzAeoOkpaTo/w3Sxlf5HuRhkqrPSjr0NYDcY9qf9IEsJgWF/SPiuRrLXksaurxSabj1YXLPMp/F70u6rvc06c7M9+as/5OfF0u6n9RGXyIdOJ8nXWsoHjinA/to1SFOIrk1ImoNh19CGoL7O+nOy7ur5h8BzMv1Po407FG9ftNJbTxN0uga71HPL4CtJH0k9+Dflx+PS3qedC3xploZI+Il0jWZAa0VHhFPkHa2japmrenXPy4lHehL7a0CRMQjpBOOGaQD+46k60MVuwIzJb1M6qGdmNezWMZTpOD6NVV9X7sd738P6fP8Eak38ztWH53pqGK7f4e07T1BCm5vtmU72uA2Ug/jaUmVfe5zpAPtUtKBtDi68jbgalJAeDSvU+W730eShvQeIZ2oXM3KSxK13qc1e5OG9G8i9b5fy+tVcWFO+yTp2PQaaf+qeI10sgxpNOq1wrz9SNekl5D2w/ERsagyM08/Sdo2KiNC15BGoi4vlHMLKQj+JS+/jDpDilXl1mvDI0iB9jHSddmTcv5ZpGuf/53rPpd03bpN7cj7OVJv/FlSD/34iJhdqPfBpOvLS0g3Gk0o5P056Uash0jH49/kNEi999NyuYtIX735RETcn+dfQuqw3ZHb4lzSiORjAJJ+JulndVbtctJ10+dJN4Adluu8lHTCNIF0jH+aFDcqJ1ufJ40qPk3aV64gdYrqUtXwdqfIZzN/J91w8Vpby7+Fcn8PXBQRl5RVZjvf93bSTVa3d+b7NiNJj5PuCn+izYVXz1vZoUdFxF9Lr9xabE3a3Von6dukO1e/3QzlrgvyCMIK0o1kTyn9ME9LRHyzhLLPAt4WERPrLdclP7gQEctIvc2BZZUpaUPg7aQz8c72PdKZo7XtK6zek22v44F7HVQ7ZE3a3Vp3GWnEq1nKXReMJI0KPN3Wgm3JQ9Q75eHnsaTv3l7bVr5G/eJLmyKi1i8zdYikzUlDFjdQ44apRouIWzv7PYskfZ2qm4WyOyOi1o1VXSYift2RfEo/VyfS95OL6bOpPWR6bET8qkZ6p1C6YeXmWvMi/QpWp+pou3cH3fUzBoj0DYZ2fcOgO5S7tpN0MOlS1tfy/SRrqi9p+HdL0mjZD4Hr26xHVwwFm5mZra2602/vmpmZNb0uGwruKptttlkMHz68q6thZmbZfffd91xEDOrqepRlnQusw4cPZ9asWV1dDTMzyyRV/6JeU/NQsJmZWYkcWM3MzErkwGpmZlaide4aay1vvPEGLS0tLFu2rKurskZ69+7N0KFD6dWrV9sLm5lZQziwAi0tLfTt25fhw4cjtfaHE91bRLB48WJaWlrYZptturo6ZmbrLA8FA8uWLWPgwIFNG1QBJDFw4MCm73WbmTU7B9asmYNqxdqwDmZmzc6B1czMrES+xlrDBReUW96kSR3LN2/ePPbff38efnj1//I95phjOPnkk9l+++3XsHZmZlYmB9YmddFFF3V1FcxsXVGvt9HRnsNazEPB3cjZZ5/NyJEjGTlyJOeccw4Ay5cvZ+LEiey0004ccsghvPrqqwDss88+/mlGM7NuyIG1m7jvvvv45S9/ycyZM7n77ru58MILWbJkCXPmzGHSpEn8+c9/pl+/fpx33nldXVUzM6vDgbWbuOuuu/joRz/KRhttxMYbb8zHPvYx7rzzToYNG8Yee+wBwOGHH85dd3X6/7ibmdlb4MDaTbT2h/PVX6HxV2rMzLo3B9ZuYu+99+a6667j1Vdf5ZVXXuHaa69lr7324qmnnmLGjBkAXHHFFey5555dXFMzM6unoXcFS9oEuAgYCQTwGWAOcBUwHJgHfDwiluTlTwWOBlYAX4yIW3L6aGAy0Ae4CTgxIkLSBsAlwGhgMfCJiJi3pvXuipvcRo0axVFHHcXYsWOB9HWaTTfdlO22244pU6Zw7LHHMmLECI4//vg387j3ambW/TT66zY/Bn4bEYdIWh/YEPg6cGtEnCnpFOAU4GuStgcmADsAWwL/J+mdEbECOB+YBNxNCqzjgZtJQXhJRGwraQJwFvCJBq9Tw5x88smcfPLJq6Q98sgjNZddvHgxAwYM6IxqmZnZW9CwoWBJ/YC9gYsBIuIfEfECcCAwJS82BTgoTx8IXBkRr0fEE8BcYKykwUC/iJgR6ULkJVV5KmVdDYzTOtCN23fffdlxxx39Y/tmZt1QI3usbwcWAb+UtDNwH3AisEVELASIiIWSNs/LDyH1SCtactobebo6vZJnfi5ruaQXgYHAcw1Zo25i+vTpXV0FMzNrRSNvXuoJjALOj4hdgFdIw76tqdXTjDrp9fKsWrA0SdIsSbMWLVpUv9ZmZmZroJGBtQVoiYiZ+fXVpED7TB7eJT8/W1h+WCH/UGBBTh9aI32VPJJ6Av2B56srEhEXRMSYiBgzaNCgElbNzMystoYF1oh4Gpgv6V05aRzwCDANmJjTJgLX5+lpwARJG0jaBhgB3JOHjZdK2j1fPz2yKk+lrEOA26K1L4SamZl1gkbfFfwF4Ff5juDHgU+TgvlUSUcDTwGHAkTEbElTScF3OXBCviMY4HhWft3m5vyAdGPUpZLmknqqExq8PmZmZnU1NLBGxIPAmBqzxrWy/BnAGTXSZ5G+C1udvowcmEvVXf43Dpg8eTIf+MAH2HLLLUuskJmZNYp/eambmzx5MgsWLGh7QTMz6xb8f6zdyOmnn86vfvUrhg0bxmabbcbo0aOZNWsWhx12GH369GHGjBn06dOnq6tpZmZ1OLB2E7NmzeLXv/41DzzwAMuXL2fUqFGMHj2aMWPG8IMf/IAxY2qNqJuZWXfjwNpN3HXXXRx44IFv9kg/8pGPdHGNzMysI3yNtZvwt4TMzNYODqzdxJ577skNN9zAsmXLePnll/nNb34DQN++fVm6dGkX187MzNrLQ8G1dMH/xu26664ccMAB7Lzzzmy99daMGTOG/v37c9RRR3Hcccf55iUzsybhHms38uUvf5k5c+Zw3XXXMWfOHEaPHs3BBx/MnDlzePDBBx1UzcyagHus3cikSZN45JFHWLZsGRMnTmTUqFFdXSUzM3uLHFi7kcsvv7yrq2BmZmvIgTWLCJr9P9J9Z7HZOqytn2LtgntH1lW+xgr07t2bxYsXN3VgiggWL15M7969u7oqZmbrNPdYgaFDh9LS0kKz/wl67969GTp0aNsLmplZwziwAr169WKbbbbp6mqYmdlawEPBZmZmJXJgNTMzK5EDq5mZWYkcWM3MzErkwGpmZlYiB1YzM7MSObCamZmVyIHVzMysRP6BCDOzJlHv54D9S8Ddh3usZmZmJWpoYJU0T9JDkh6UNCunDZA0XdJf8/OmheVPlTRX0hxJ+xXSR+dy5ko6V/lvaCRtIOmqnD5T0vBGro+ZmVlbOqPH+t6IeE9EjMmvTwFujYgRwK35NZK2ByYAOwDjgfMk9ch5zieNdIzIj/E5/WhgSURsC/wIOKsT1sfMzKxVXTEUfCAwJU9PAQ4qpF8ZEa9HxBPAXGCspMFAv4iYEel/3S6pylMp62pgnJr9T1XNzKypNTqwBvC/ku6TVLm2vkVELATIz5vn9CHA/ELelpw2JE9Xp6+SJyKWAy8CAxuwHmZmZu3S6LuC94iIBZI2B6ZLeqzOsrV6mlEnvV6eVQtOQX0SwFZbbVW/xmZmZmugoT3WiFiQn58FrgXGAs/k4V3y87N58RZgWCH7UGBBTh9aI32VPJJ6Av2B52vU44KIGBMRYwYNGlTOypmZmdXQsMAqaSNJfSvTwAeAh4FpwMS82ETg+jw9DZiQ7/TdhnST0j15uHippN3z9dMjq/JUyjoEuC1fhzUzM+sSjRwK3gK4Nt9L1BO4PCJ+K+leYKqko4GngEMBImK2pKnAI8By4ISIWJHLOh6YDPQBbs4PgIuBSyXNJfVUJzRwfczMzNrUsMAaEY8DO9dIXwyMayXPGcAZNdJnASNrpC8jB2YzM7PuwD9paGa2jqv3U4ngn0t8q/yThmZmZiVyYDUzMyuRA6uZmVmJHFjNzMxK5MBqZmZWIgdWMzOzEjmwmpmZlciB1czMrEQOrGZmZiVyYDUzMyuRA6uZmVmJHFjNzMxK5MBqZmZWIgdWMzOzEjmwmpmZlciB1czMrEQOrGZmZiVyYDUzMyuRA6uZmVmJHFjNzMxK5MBqZmZWIgdWMzOzEjmwmpmZlciB1czMrEQND6ySekh6QNKN+fUASdMl/TU/b1pY9lRJcyXNkbRfIX20pIfyvHMlKadvIOmqnD5T0vBGr4+ZmVk9ndFjPRF4tPD6FODWiBgB3JpfI2l7YAKwAzAeOE9Sj5znfGASMCI/xuf0o4ElEbEt8CPgrMauipmZWX0NDayShgIfBi4qJB8ITMnTU4CDCulXRsTrEfEEMBcYK2kw0C8iZkREAJdU5amUdTUwrtKbNTMz6wqN7rGeA3wV+GchbYuIWAiQnzfP6UOA+YXlWnLakDxdnb5KnohYDrwIDKyuhKRJkmZJmrVo0aI1XCUzM7PWNSywStofeDYi7mtvlhppUSe9Xp5VEyIuiIgxETFm0KBB7ayOmZnZW9ezgWXvARwg6UNAb6CfpMuAZyQNjoiFeZj32bx8CzCskH8osCCnD62RXszTIqkn0B94vlErZGZm1paG9Vgj4tSIGBoRw0k3Jd0WEYcD04CJebGJwPV5ehowId/puw3pJqV78nDxUkm75+unR1blqZR1SH6P1XqsZmZmnaWRPdbWnAlMlXQ08BRwKEBEzJY0FXgEWA6cEBErcp7jgclAH+Dm/AC4GLhU0lxST3VCZ62EmZlZLZ0SWCPiDuCOPL0YGNfKcmcAZ9RInwWMrJG+jByYzczMugP/8pKZmVmJHFjNzMxK5MBqZmZWIgdWMzOzEjmwmpmZlciB1czMrETtCqySvi+pn6Rekm6V9JykwxtdOTMzs2bT3h7rByLiJWB/0s8IvhP4SsNqZWZm1qTaG1h75ecPAVdEhH+P18zMrIb2/vLSDZIeA14DPidpELCscdUyMzNrTu3qsUbEKcC/AGMi4g3gFdKfjJuZmVlB3R6rpI/VSCu+vKbsCpmZmTWztoaCP1JnXuDAamZmtoq6gTUiPt1ZFTEzM1sbtPd7rFtIuljSzfn19vn/VM3MzKygvV+3mQzcAmyZX/8FOKkB9TEzM2tq7Q2sm0XEVOCfABGxHFjRsFqZmZk1qfYG1lckDSTdsISk3YEXG1YrMzOzJtXeH4j4EjANeIekPwCDgEMbViszM7Mm1a7AGhH3Sfo34F2AgDn52czMzArae1fwHcDQiJgdEQ8D7wHubWC9zMzMmlJ7h4L/E/itpHOBIaQf4/d3XM3MzKq0dyj4FknHAdOB54BdIuLphtbMzMysCbV3KPjfgZ8AewOnAXdI+nAD62VmZtaU2v09VmBsRMyIiJ8D+9HGD0RI6i3pHkl/kjRb0ndy+gBJ0yX9NT9vWshzqqS5kuZI2q+QPlrSQ3neucr/BCBpA0lX5fSZkoa/tdU3MzMrV3v/Nu7EiHit8PrJiNi3jWyvA++LiJ1JNzuNz99/PQW4NSJGALfm10jaHpgA7ACMB86T1COXdT4wCRiRH+Nz+tHAkojYFvgRcFZ71sfMzKxR6gZWSefk5xskTat+1Msbycv5Za/8CNL/uE7J6VOAg/L0gcCVEfF6RDwBzAXGShoM9Mu95QAuqcpTKetqYFylN2tmZtYV2rp56dL8/IOOFJ57nPcB2wI/jYiZkraIiIUAEbFQ0uZ58SHA3YXsLTntjTxdnV7JMz+XtVzSi8BA0g1WxXpMIvV42WqrrTqyKmZmZu3SVmCdLekkUmB8CLg4/05wu0TECuA9kjYBrpU0ss7itXqaUSe9Xp7qelwAXAAwZsyY1eabmZmVpa1rrFOAMaSg+kHghx15k4h4AbiDdG30mTy8S35+Ni/WAgwrZBsKLMjpQ2ukr5JHUk+gP/B8R+poZmZWhrYC6/YRcXi+E/gQYK/2FixpUO6pIqkP8H7gMdJvDk/Mi00Ers/T04AJ+U7fbUg3Kd2Th42XSto9Xz89sipPpaxDgNvydVgzM7Mu0dZQ8BuViXwN862UPRiYkq+zrgdMjYgbJc0ApuY/Sn+K/GP+ETFb0lTgEWA5cEIeSgY4nvSfsH2Am/MD4GLgUklzST3VCW+lgmZmZmVrK7DuLOmlPC2gT34t0o2//VrLGBF/Bnapkb4YGNdKnjOAM2qkzwJWuz4bEcvwv+yYmVk3UjewRkSPevPNzMxsVe395SUzMzNrBwdWMzOzEjmwmpmZlciB1czMrEQOrGZmZiVyYDUzMyuRA6uZmVmJHFjNzMxK5MBqZmZWIgdWMzOzEjmwmpmZlciB1czMrEQOrGZmZiVyYDUzMyuRA6uZmVmJHFjNzMxK5MBqZmZWIgdWMzOzEjmwmpmZlciB1czMrEQOrGZmZiVyYDUzMyuRA6uZmVmJGhZYJQ2TdLukRyXNlnRiTh8gabqkv+bnTQt5TpU0V9IcSfsV0kdLeijPO1eScvoGkq7K6TMlDW/U+piZmbVHI3usy4EvRcR2wO7ACZK2B04Bbo2IEcCt+TV53gRgB2A8cJ6kHrms84FJwIj8GJ/TjwaWRMS2wI+Asxq4PmZmZm1qWGCNiIURcX+eXgo8CgwBDgSm5MWmAAfl6QOBKyPi9Yh4ApgLjJU0GOgXETMiIoBLqvJUyroaGFfpzZqZmXWFTrnGmododwFmAltExEJIwRfYPC82BJhfyNaS04bk6er0VfJExHLgRWBgjfefJGmWpFmLFi0qaa3MzMxW1/DAKmlj4NfASRHxUr1Fa6RFnfR6eVZNiLggIsZExJhBgwa1VWUzM7MOa2hgldSLFFR/FRHX5ORn8vAu+fnZnN4CDCtkHwosyOlDa6SvkkdST6A/8Hz5a2JmZtY+jbwrWMDFwKMRcXZh1jRgYp6eCFxfSJ+Q7/TdhnST0j15uHippN1zmUdW5amUdQhwW74Oa2Zm1iV6NrDsPYAjgIckPZjTvg6cCUyVdDTwFHAoQETMljQVeIR0R/EJEbEi5zsemAz0AW7OD0iB+1JJc0k91QkNXB8zM7M2NSywRsRd1L4GCjCulTxnAGfUSJ8FjKyRvowcmM3MzLoD//KSmZlZiRxYzczMSuTAamZmViIHVjMzsxI5sJqZmZXIgdXMzKxEDqxmZmYlcmA1MzMrkQOrmZlZiRxYzczMSuTAamZmViIHVjMzsxI5sJqZmZXIgdXMzKxEjfw/1nXTBRfUnz9pUufUw8y6Jx8j1nrusZqZmZXIgdXMzKxEDqxmZmYlcmA1MzMrkW9eeovavO+gc6phZmbdlHusZmZmJXJgNTMzK5EDq5mZWYkcWM3MzErUsMAq6ReSnpX0cCFtgKTpkv6anzctzDtV0lxJcyTtV0gfLemhPO9cScrpG0i6KqfPlDS8UetiZmbWXo3ssU4GxlelnQLcGhEjgFvzayRtD0wAdsh5zpPUI+c5n3Sz7Yj8qJR5NLAkIrYFfgSc1bA1MTMza6eGBdaI+D3wfFXygcCUPD0FOKiQfmVEvB4RTwBzgbGSBgP9ImJGRARwSVWeSllXA+MqvVkzM7Ou0tnXWLeIiIUA+XnznD4EmF9YriWnDcnT1emr5ImI5cCLwMBabyppkqRZkmYtWrSopFUxMzNbXXe5ealWTzPqpNfLs3pixAURMSYixgwaNKiDVTQzM2tbZwfWZ/LwLvn52ZzeAgwrLDcUWJDTh9ZIXyWPpJ5Af1YfejYzM+tUnf2ThtOAicCZ+fn6Qvrlks4GtiTdpHRPRKyQtFTS7sBM4EjgJ1VlzQAOAW7L12HNzLpUvZ8+9c+erv0aFlglXQHsA2wmqQX4NimgTpV0NPAUcChARMyWNBV4BFgOnBARK3JRx5PuMO4D3JwfABcDl0qaS+qpTmjUulgH+M+czWwd1bDAGhGfbGXWuFaWPwM4o0b6LGBkjfRl5MBsZmbWXfjfbcysc3gUw9YR3eWuYDMzs7WCA6uZmVmJPBRsHeY7H83MVuceq5mZWYncYzWzUrR5b1LnVMOsy7nHamZmViL3WM0are7FaPfjzNY2DqxmzcrfCzXrljwUbGZmViIHVjMzsxI5sJqZmZXI11jNzGrxNWzrIAdWs27K3ws1a04eCjYzMyuRe6xm1tw8ZGvdjAOrrTPaHlr1DzkA3S5QeUjcmo0Dq9kaarYDf7PV16zZOLCu67pZ78TMrNk5sK4D/L+pZrV537BG8F3BZmZmJXJgNTMzK5EDq5mZWYl8jbVZ+Cajlfz/pmbWjTV9YJU0Hvgx0AO4KCLO7OIqdYi/ArGS28LMmllTDwVL6gH8FPggsD3wSUnbd22tzMxsXdbUgRUYC8yNiMcj4h/AlcCBXVwnMzNbhykiuroOHSbpEGB8RByTXx8B7BYRn69abhIrRxDfBcxpYLU2A55zuQ0rt5Flu1yX63I7v2yArSNiUAPL71TNfo1VNdJWO1OIiAug3g/BlkfSrIgY43IbU24jy3a5Ltfldn7Za6NmHwpuAYYVXg8FFnRRXczMzJo+sN4LjJC0jaT1gQnAtC6uk5mZrcOaeig4IpZL+jxwC+nrNr+IiNldXK1GDTm73MaX7XJdrsvt/LLXOk1985KZmVl30+xDwWZmZt2KA6uZmVmJHFhLJKmHpAck3djdy5XUW9I9kv4kabak75RU7jBJt0t6NJd7YhnlFsovtS0aVd8Gtm9TlZvLbtg2IWmepIckPShpVkllNts2/K68/pXHS5JO6q7lrgt8jbVEkk4GxgD9ImL/7lyuJAEbRcTLknoBdwEnRsTda1juYGBwRNwvqS9wH3BQRDyy5rUuvy0aVd8Gtm9TlZvLbtg2IWkeMCYiSvvxgmbbhqvK7gH8nfRDOU9293LXVu6xlkTSUODDwEXNUG4kL+eXvfJjjc+yImJhRNyfp5cCjwJD1rRcaExbNKq+DWzfpio3l92wbaIRmm0brjIO+FsDgl+jyl0rObCW5xzgq8A/m6TcypDUg8CzwPSImFly+cOBXYCyyj2HBrUFlF/fRrVvs5Vb9R7DKXebCOB/Jd2n9NOlpWq2bZj0Xf4rmqjctZIDawkk7Q88GxH3NUO5FRGxIiLeQ/rFqrGSRpZVtqSNgV8DJ0XESyWU19C2KLu+0Lj2bbZyKxrRxsAeETGK9A9XJ0jau6Rym3EbXh84APifZih3bebAWo49gAPy9Z4rgfdJuqwbl7uKiHgBuAMYX0Z5+Rrdr4FfRcQ1ZZRJA9uiQfV9U9nt24zlNqqNI2JBfn4WuJb0j1drrNm24eyDwP0R8UyJZTay3LVXRPhR4gPYB7ixu5cLDAI2ydN9gDuB/UsoV8AlwDnN0MaNqm8D27epym1wG28E9C1M/5H0b1fdsr5V71H6cYIUrD/dgLo2pNy1+dHUP2loa2QwMCXf7bceMDUiyrj9fw/gCOChfL0O4OsRcVMJZTdCo+rbqPZttnKhcW28BXBtuqGZnsDlEfHbNSwTmm8bRtKGwL7Asc1Q7trOX7cxMzMrka+xmpmZlciB1czMrEQOrGZmZiVyYDUzMyuRA6uZmVmJHFjNzMxK5MBq1oUknSbpyw1+j5fz83BJr+W/LHs0/03cxEa+t9m6yD8QYbZu+VtE7AIg6e3ANZLWi4hfdnG9zNYa7rGadSJJR0r6s9Ifil9aNe+zku7N836df/UGSYdKejin/z6n7ZB7nA/m8ka81bpExOPAycAXy1g3M0scWM06iaQdgG8A74uInYETqxa5JiJ2zfMeBY7O6d8C9svpB+S044AfR/o3mjFASwerdT/w7g7mNbMaHFjNOs/7gKsj4jmAiHi+av5ISXdKegg4DNghp/8BmCzps0CPnDYD+LqkrwFbR8RrHayTOpjPzFrhwGrWeUT6Y+7WTAY+HxE7At8BegNExHHAN4FhwIOSBkbE5aTe62vALZLe18E67ULqHZtZSRxYzTrPrcDHJQ0EkDSgan5fYGH+L9DDKomS3hERMyPiW8BzwLB849HjEXEuMA3Y6a1WRtJw4AfATzqyMmZWm+8KNuskETFb0hnA7yStAB4A5hUW+XdgJvAk8BAp0AL8V745SaTg/CfgFOBwSW8ATwPfbWc13iHpAVJveCnwE98RbFYu/22cmZlZiTwUbGZmViIPBZutBfJ121trzBoXEYs7uz5m6zIPBZuZmZXIQ8FmZmYlcmA1MzMrkQOrmZlZiRxYzczMSvT/Aa3b/aVMvfIFAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.ticker import MaxNLocator\n",
    "from collections import namedtuple\n",
    "for target in gt_targets:\n",
    "    gt_boxes = [tuple(x['bbox']) for x in target['gts']]\n",
    "    gt_boxes = np.array(gt_boxes)\n",
    "    obj_boxes = [tuple(x['bbox']) for x in target['objs']]\n",
    "    obj_boxes = np.array(obj_boxes)\n",
    "    overlaps = utils.compute_overlaps(obj_boxes, gt_boxes)\n",
    "    #print(overlaps)\n",
    "    obj_bars = []\n",
    "    gt_bars = []\n",
    "    index_bars = []\n",
    "    n_groups = len(obj_boxes)\n",
    "    for i in range(0, n_groups):\n",
    "        w = np.where(overlaps == max(overlaps[i,:]))\n",
    "        #print(w)\n",
    "        # print(target['objs'][w[0][0]]['class_id'])\n",
    "        # print(target['gts'][w[1][0]]['class_id'])\n",
    "        obj = target['objs'][w[0][0]]\n",
    "        gt = target['gts'][w[1][0]]\n",
    "        obj_bars.append(obj['mask_area'])\n",
    "        gt_bars.append(gt['mask_area'])\n",
    "        index_bars.append(obj['class_id'])\n",
    "\n",
    "    print(obj_bars)\n",
    "    print(gt_bars)\n",
    "    print(index_bars)\n",
    "    fig, ax = plt.subplots()\n",
    "\n",
    "    index = np.arange(n_groups)\n",
    "    bar_width = 0.35\n",
    "\n",
    "    opacity = 0.4\n",
    "    rects1 = ax.bar(index, obj_bars, bar_width,\n",
    "                    alpha=opacity, color='b')\n",
    "\n",
    "    rects2 = ax.bar(index + bar_width, gt_bars, bar_width,\n",
    "                    alpha=opacity, color='r')\n",
    "\n",
    "    ax.set_xlabel('class_ID')\n",
    "    ax.set_ylabel('Pixels')\n",
    "    ax.set_title(target['file'])\n",
    "    ax.set_xticks(index + bar_width / 2)\n",
    "    ax.set_xticklabels(index_bars)\n",
    "    ax.legend(labels=['obj', 'gt'])\n",
    "\n",
    "    fig.tight_layout()\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 覆盖测试\n",
    "\n",
    "人工不可能把所有目标的尺寸和宽高比逐个设置。现实做法是设置几个代表性的尺寸，如32、64、128。代表性的宽高比，如0.5、1、2\n",
    "\n",
    "代表性数据不可能把所有目标都覆盖（IOU）\n",
    "\n",
    "根据上述坐标图的数据分布情况，设置代表性尺寸、宽高比，在每个目标上标识最大的交并比（IOU）。\n",
    "\n",
    "最大交并比（IOU）：理想值就是目标本身，但是以理想值为代表值，必然造成其它目标的交并比（IOU）下降。\n",
    "\n",
    "三维坐标图：尺寸、宽高比、IOU\n",
    "\n",
    "尺寸：锚框的正方形尺寸，应该从GT目标中挑选，把尺寸差值小于阈值的目标归为一组，将平均值作为锚框尺寸。目前网络设计，最多只有5个尺寸设置\n",
    "\n",
    "宽高比：锚框的，也应该从GT目标中挑选。IoU：对所有目标的平均IoU，\n",
    "\n",
    "理想状态是IOU值足够高（0~1），足以用于目标的过滤。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 从小到大排序GT目标框尺寸\n",
    "\n",
    "# 查找GT目标框尺寸的分组位置，即组内尺寸接近"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 代表性的尺寸、宽高比挑选（实验性）\n",
    "\n",
    "人工设置代表值达到IOU理想状态也不现实。。。\n",
    "\n",
    "怎么让计算机挑选代表值？"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}