ntb.ipynb 242 KB
Newer Older
Ondřej Švejstil's avatar
Ondřej Švejstil committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "3c7ecc59",
   "metadata": {},
   "source": [
    "## Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "83413fc2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "import tensorflow as tf\n",
    "import os\n",
    "from tensorflow.keras import datasets, layers, models\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import precision_recall_fscore_support\n",
    "from sklearn.metrics import classification_report\n",
    "from imblearn.under_sampling import RandomUnderSampler\n",
    "import random"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "59c22538",
   "metadata": {},
   "outputs": [],
   "source": [
    "CLEAR_CLASS = 0\n",
    "FIELD_CLASS = 1\n",
    "MASK_CLASS = 2\n",
    "PROBE_CLASS = 3\n",
    "SCRATCH_CLASS = 4"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f6e611a",
   "metadata": {},
   "source": [
    "## Helper functions\n",
    "Here we declare some functions that will help us later with loading the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "58489208",
   "metadata": {},
   "outputs": [],
   "source": [
    "# function to quickly convert from classification outputs to strings\n",
    "def defect_class_to_str(defect_type):\n",
    "    type_str = \"\"\n",
    "    if defect_type == CLEAR_CLASS:\n",
    "        type_str = 'clear'\n",
    "    if defect_type == FIELD_CLASS:\n",
    "        type_str = 'field'\n",
    "    if defect_type == MASK_CLASS:\n",
    "        type_str = 'mask'\n",
    "    if defect_type == PROBE_CLASS:\n",
    "        type_str = 'probe'\n",
    "    if defect_type == SCRATCH_CLASS:\n",
    "        type_str = 'scratch'\n",
    "    if type_str == \"\":\n",
    "        raise ValueError(defect_type)\n",
    "    else:\n",
    "        return type_str\n",
    "    \n",
    "# function that loads single map from the folder of one class\n",
    "def load_map(index, defect_type):\n",
    "    type_str = defect_class_to_str(defect_type)\n",
    "    df = pd.read_csv('data/'+ type_str + '/' + str(index) + '_' + type_str + '.csv')\n",
    "    wafer_map = df.iloc[:,1:].to_numpy() # data have unnecessary column names which are not relevant\n",
    "    return wafer_map\n",
    "\n",
    "def visualize_map(wafer_map):\n",
    "    plt.figure(figsize=(5,5))\n",
    "    plt.imshow(wafer_map, cmap='jet', aspect=0.60747) # this aspect value = 65/107 - so it stretches the map to square\n",
    "    plt.show()\n",
    "    \n",
    "    \n",
    "# loads all of the data\n",
    "def load_maps():\n",
    "    n_maps = []\n",
    "    \n",
    "    for class_type in range(5):\n",
    "        n_samples = len(os.listdir('./data/' + defect_class_to_str(class_type)))\n",
    "        n_maps.append(n_samples)\n",
    "    \n",
    "    maps = np.zeros((sum(n_maps), 107, 65))\n",
    "    y = np.zeros(sum(n_maps))\n",
    "    \n",
    "    class_type_i = 0\n",
    "    \n",
    "    # basically goes through every folder and loads every CSV file there is in that folder\n",
    "    for n in n_maps:\n",
    "        for file_i in range(1,n+1):\n",
    "            i = file_i-1 + sum(n_maps[0:class_type_i])\n",
    "            maps[i] = load_map(file_i, class_type_i) \n",
    "            y[i] = class_type_i\n",
    "        class_type_i += 1\n",
    "    \n",
    "    return (maps, y)\n",
    "            \n",
    "                 "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1f54ab2",
   "metadata": {},
   "source": [
    "## Data exploration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3b450183",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting each class to see how it looks\n",
    "\n",
    "fig, axes = plt.subplots(2,2, figsize=(10,10))\n",
    "axes[0,0].imshow(load_map(9,FIELD_CLASS), cmap='jet', aspect=0.60747)\n",
    "axes[0,0].set_title('Field')\n",
    "axes[0,1].imshow(load_map(7, MASK_CLASS), cmap='jet', aspect=0.60747)\n",
    "axes[0,1].set_title('Mask')\n",
    "axes[1,0].imshow(load_map(11, PROBE_CLASS), cmap='jet', aspect=0.60747)\n",
    "axes[1,0].set_title('Probe')\n",
    "axes[1,1].imshow(load_map(7, SCRATCH_CLASS), cmap='jet', aspect=0.60747)\n",
    "axes[1,1].set_title('Scratch')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9b7408dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAExCAYAAADocL6iAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAAyv0lEQVR4nO2dfXxV1ZX3v7vcqwmSkAYkEYJGDBUQFJFHUHmmjKKDFotWq7a1gxbGOh9ttdanovWpL6NT7WNVOnZsFVud0lasWrVMfQNLp0wrGhALChRGQwma8GYMMYnc0P38sc7xJnhDknPPzdn33PX9fM4nue/r3L3v7+y91t5rGWstiqIoceQTURugKIqSK1TgFEWJLSpwiqLEFhU4RVFiiwqcoiixRQVOUZTYkhOBM8bMNMZsNMZsNsbMz8VnKIqi9IQJex2cMWYA8BfgdKAeeBX4grX2zVA/SFEUpQcSOXjPE4HN1tq3AIwxjwKzgW4FzpiBFspyYEqhMQAoBg7x/t+PBDAcyg/dyTAaSbCPXZTz7gdVUAe0A/wNaPWOjoB2FHk2HBzw9bmiHTmv9mAvTwzn0OMaOHzzNt55H4YCnzjB8PoHk+SS/rfevEmHZ0MbsC+YHcp+vLvTWntopkdyIXAjgK2dbtcDU/Z/kjHmMuAyuTU4/a+SBeXAscAkoPTjD5cB34K/v2IRV3MPJexhMRfx3f++FS4H1oH8+FZ7x+6AdtQgTT464OtzxZvIeW0O9vJBN/PZ2vtY+OWvcf8imHMIvFf7Sar+uxZmAi29eZNmYCWwnuDfr9KVW7Z090guBK5XWGsfAB4AMGa47hfrFaXAMKACGantT7H3WEnml7cAS+GJqotZOnMGxQe30rByFCwCdubG4ljR9DYPPXQlB/1sL+f+7Nc8wlHcyXz5/no9KCwBxiIXozZgD9DoHW25sLqgyYXAbQNGdrpd5d2nZI3/4zgWSPb95e3AEmApvJ+o5H2QGVM7wWejBcVzMO/T3H/5NdxfeY3c1eIdvf7+DPKTqPJuNwN/9v6qwIVNLgTuVWC0MeZIRNguAr6Yg88pAJLIqMwfrVUgo7gA4ubTQdep1FBgDPJ7SwB1A6H2FGRkURfgAxLAEDKPMKOmGLGtjWCKXi6v60hBfRZt8DGbSpG2TQIpz7Y2738lG0IXOGtthzHmSuB5xNP9E2vtG2F/TmEwDBmxVSM/hCTdTj+DkABmwCfu/YAvV/wHw3mXP3Iyv39kJsw/Exp2BXzjYmBgeHaGRjUiJEFHSv4FJyxx89/Tb+MUMpLbjProwiEnPjhr7W+B3+bivQuHJOkre1UPzw1IAqiGkyv+yJf4BUexmeG8w8pzTqT90XJYOkSeF5vpq0GE1zXxTZIWzXKCj56V/YksyKBkohwRtHKkw/u3c0QHUAsrnj6dr8yuZhiN7KGEgYPaaF/YDi1FEgN/DvHdbcidKYqPQdp8EjKq6wB2AduR0Z3SF1TgnCGJrLCZBBzZPx/ZASwHaqG+aDT1idFwDhzxbxu4Y8R8JrKG2qNP4KbTbmVH/eEqcP1GBekLWwqZrq5GBa7vqMBFiu/TSZB2Nvfz9KkDaOp02/u/jCaG0chItnI0G9kx/XB5rB1o8I7OrwsFS/fz4QQyuik0fL+rv6xEgxB9QQUuMoqRKUgN6QiaL3LuMJKtfJ0fMOOrS2n8agVbGcmSjZ+HG4GnCNE/V4841xszPJZA/JA1SBS00KhG+oUvatuQ72obKnIHRgUuMnwfWzUu/2jLaOJEVnIiK0mwj40czZ6jS/j9mJnSe0IROItEDDeReclk0vugYbj8XeUOQ9fzTpL5QqDsjwpcv+ELWjlyNfZCmI6N2NgMWxaPYf6Fd1DDZoaxneN5jRNZycguO/C6IQGMR7YuTSXzFuM6xPe3FBm4KX2kFNkG54/q2hDB24OO6LqiAtdvlCBTrEk4PQqpBa6Et+Yfw1tFx8B0eP7+v3I73+6dwBUBs2DSv6zgG9yb8TV//PTJ3DTrFlKXlKrABaIU2c0CMvrdhgQh1qMC1xUVuH6l864ER+lA9qX6e1OHwo43DmfjMUczkq1sZSRtFEMlMBnZFdHkPb8l/R77SLCXg9jLQRzEXsp4jxL2MIB9bGUkZUOa2FHk2Oh1ELKzo8y73Y6cVxMOrwU0iOD5R+fdECp2KnDKgdkA3Ajfn3Yj36++EcogObGZE6/4PSOv2MoeSnhhy1lwWxIe5aP9rq+3TGXu5KlQBoNm7OCSQ37KhSzu3SgwChLItHo+fPqE5xjKTmqZzJYfjoH7cHyJTCkyMzgCETZ/N0Qdhb6/VQVOOTA7kWjpU97t8ZBaWMoZQ57nH/kZeyhh2BHbWTT5n+Q5LUjapXXe8xPQMu9Qlt//98xgmdsCNx5OPWEJ3+FfqOZtnuJcrj5nATxe5LjAQXoEByJwfpYSFTglJ/hLPsqRr9nfduX4FLUnmoDlsHDKPP7C0exjABs5GmZYGGpE4DYAK4A19DKNkEG+Jz+N0P4kSScayCHr4KU3ZpE4Zh9D2EUtJ8CSIlnzF4QEkshgGjARmQJvRr6bWnKwjtCnGFk07u9t9dMy7abQpq0qcDmjFJhA1x9tDBaqNgB3QcPDo3isbJT4rK6E8/5hEecd9TgJ9rGYC3li8cUwnz4EEfwUQt2lBszxd9eObEmrhReGzpZfRgtd/ZF9pQg4RwIul/NjhrGd33A2Dy27Eq5FLgA5IYlEWWu827uRAMRaCm15iQpczui8OyEGwuazfxCiUv4fwi7G8SYHs5daJovwBepdEX5Xfm63uhDfswgG0kYZTZTRRAl7RPj6Bf+7zDLFVh6jAqcoucILuKzoOJ0V006XKeo6ZKSoy2P6BRU4RckVHcDL3qFEggqckh3twDr4NeeyjwEcxF7eZBzUpODGJLRA0cW7mcdCZr/9AtTCwM8/zwT+zEtlh0dtvRJzVOCU7GgBHoYdSw/nobIrZZHs+XDShS9x4ZzFDOcdprOcQ29toflfYfWHMP3zLXzpsV/wUs2saG1XYo8KnJIdHaTTJ4H4mcbDAPZRTR3VvM2hb7fAVtj1oaRu5AN5XHufkmu0iynh0g4shRWDTmfFjNNhaDuVR77D8Q+uYcKDaymjiQuZyGMr50iWYEXJISpwSrh0IAtZV+D1riIaGMWziVE8y+e6Ps/Z/Z1KXFCBU3JHRzf/K0o/8YmoDVAURckVKnA5oTBXjSsuk/COwuqbOkXNiF/hyi/hF4QS0rUWgpJClrxvJ9gm6WLE/qos7cgFu5BEjUEqRfmFWKpwLiMyrch5Bd3YXoz0mxGEt20tSbp4+J5ePD+F2J/NebiBClxG/GwMUWff3Y2k5lhPMCEoRTb7DySn9VX7TApJq7GaYJu/k0juM5Dzc2mvbyNyXpsJJgzlSLbezumPwmAIfevL/l6yZlTgYknnauNRkiJ4B8vmtS7jZ6yNY+TClXPyU32NIF33wT/yBxU4RVEyUApMQUaTKWRkuh4ZmeaPyKnAZSROX4sLo9BcEad28nHpnDrPYpKITy6JClze4Tt2/Yy7vg+uJEqj6JrlNsjUJUF2gZJc4Tu9E/TO6Z2JcuTcXPK/gfi5JiD9J0ibudL34oEKHCDDcb+k38CIbemMn5l1dNSG5AD/ghI3SoFxURuheKjAfYQrQQVFcRG/rohfYwTyIeigAqcoSi8YiMxwapCp9y4k4LCZYEuY+gcVOEVRekmS9Fq6csR/GnSxdv9QoALnBxXKkUbz1/voFFVR4kSBClw56ZJ+LgUVFEUJkwIVOD+gkOdFmBUlUhLIchY/0JDCtaBDgQqcoijZkURmQFVIgoE2YAuwCdlD7QYqcIqiBGQgaRePRaKrbhV8LRCB89fvlJAOKri4Cr6PVAHTvKMKqTb/MrAcid7nK4OAicAMYDzyu1mHnFctUvchX6kBpgNTgaGIHixHUrw3dPsqJSAFIHBJYBiyabgG8bvlubD5jIdP3PsB36m4lZP5I28yjjvmXkfDpaOgDncSU/SVMmAenDdnEV/kFwD89MJLWfLQ5+W83Bok9J4EMB2qHtzEddzJ0Wzkv/jf3DbrFpiXVIHLAQUgcJAOKMQvYjogsY8BpI8E+6I2KRy8njnAO58BdMSmt0pbdcSrvRzFWGujtgFjhlu4LEfvniS9zzRmezqrkGncdNJT1BXAUiRPZr4yCJgMzESmqh3AGuS8Xia/p6hjkDabhkxR65Ap6nLyd2QKiA9uLbASWfzbn9yyylo7OdMjga+JxpiRwH8gziwLPGCtXWCMKQcWI+ki6oALrLXvBf0c5QDUAw97R5xoIf2jjxsbvOO+qA0pDLIpOtMBfNNaOw5xmV5hjBkHzAeWWWtHA8u824qiKP1OYIGz1r5rrV3t/b8HSfc5ApgNPOI97RHgnCxtVBRFCUQobltjTDVwPDIBr7DWvus91EA3Sb+MMZfxkeNtcBhmKIqidCHruqjGmEHAE8DV1touaQWsRDAyRjGstQ9YayeLczB+0U1FUaInK4EzxiQRcfu5tfZJ7+5GY8xh3uOHIUU9FUVR+p3AAmeMMcBDwHpr7d2dHnoGmOP9Pwd4Orh5iqIowcnGB3cK8GVgrTFmjXffDcAdwGPGmLnI7tsLsrJQURQlIIEFzlq7gu73PJ0W9H0VRVHCIusgg6IoiquowCmKEltisn25GEmJlOl0kshSvNIcfG62GUyLcbMOhEXOK2g6ElfPCyQ5Y9DzSuBuNpqo+6JB0pFVee+TyvCcNvq71GAMBK4Y2Uw/FhGyTI2Uix/cLmRzcT3BGswvNu1aXYgUsiklaGbWJLKhxc/26hL1yLltI/MPsCfKkYQNY3FLwJtJt1mQvliMbB0fS7pqVhCORH6DmS4gKWRr+nok9hjk++87MRC4JCIWFWTXOH2lGRG5oI21B/nBtOGewDUjIhA09XQC+cG4RjPQiPzQgtCG9LMUbglcCmmrbQQTOD+dWHUIthyoL7chNiZQgVMUJWZUkJ5ptZG+4GwnV4KnAqcoSj+RpGtOxkZgNTL6VIHrRDFpv1oJMkXt7xKAA73PHUYwp7UfGHFpqgPp6Up3/syeSCCuAhdLMhYjtgUNoJTjbpuVkB4Z9ZWo2syXnyS5CjzkocD5zvkapFESRNPpKpDNHJMI9mOJyu6eSCL1K0YT/MfiX4BcoxrpMymCX5RcjBCXAlOQ6V/QvhjPlP55KHAJ5EpaTfQN0rlsWpzwC2PnYmlNlBjid04+Sfo3yJYf6EJfRVFiiwqcoiixJQ+nqBXANJiFVChqQYocryO8upKDvPceA1QiVZz8z8jrykeK4hJ+cKQEkaIOst+R0ZU8HMFdQHJnM+t/U42dZLDfMlz54vfgaqAopI+YBqNffZ31P6vGnmiwXzXc/Ox1cBsifoqihEApEqT7LPA5pJ5iDWH6SfNP4CoNVwz5IWM+s4WbvwiNo+A7/IvUmQxrPFoDl/IwYz63hdsvgD9PgJu2fg9mtIcnooqiIEG6CmRbX433f3gTy/wTuIZW7n39ep78zzO5+S34wFbyTzwAzxFeQeANcOeH1/H0k2fw7b/CMDuYS0b+OzxelN9FhxWlwMhDH9xjMPFszhvzW1kp0o5UdK8neJKI/VkB74+p5Jyq52VK2oH49+oRn5+iKHlBnghcEjHVX2BZBxtKYUOOFly2y0cE3pMNMpUt844i7z2bvCPKUWAlcmGoRL7Snch5NqCj07iSQPrhUKQvdiD9cCexb/M8mKKWAxOAM4GzEUeka+lqMjAVWAjHrX+Z815bxPj1r8J9wMQIbaoEfgQ3/+k67I8N9ibDE787C1a0w/gI7VJyyxjgDhi1/g1mv/ZLTlr7EjyK/JTyZIgTlDwQuGIkv9gEZPtQFXmxe6AaRn/mdb7DrXyfa7iB2znivA3RZhEaBEfM3sBNL32PZyvgBxPgc1c/y9wRC91L3aaERyUUnb+ba/l/LOAqbuNGpp3yolzUYi5wMT+9CKmDTcuO467TruVTbGQ949jyn2OiXUfXAlueH8Pd//DPXLP2fs7cDi+dehI/f/+Lur4vzuyE9qfK+fc5V/BHTmE7w1jx+umwgfD81o5ipPh8xEaY4RYu6+bREchamUm4mSq6G4qQKaHv92hBfB5R+j0Snk1jkBGb74PbgPjhYu6PKViKkH7Y2Qfn90WngmatSPokP4VSb7lllbV2cqZHdASXK8IIVIRNBzJS09FaYdFOHrW7n66rmPTwMniuOBU4RVEcYSASQByGbNfag6jyFiT7b99RgVMUxSFKSW/VSiEStZugApcHUVRFUZRgqMApihJbHJ6iJjv9dXxRr6IoTuKgwJUjS0P8dQx+zdM8WiKiKIoTOCpwY4FxURuiKEqeoz44RVFiiwqcoiixxWGBi34LmaIo+Y2DPrjdwFpgO7Jdw6/Y7VrNx1agEbHTJ0X3Ed/9H0sg5zUC9wIojd6xf/GP7s5v//v9QJFrWV+akfPqyz7HzhQjbVYRmkXhYIFtpNts/zY6UL/0GYYE9uK1YsFRgdsNrEe+bH+zfTluCcFuZFPwZoJVASpFUkAV45Z4twKbkItMY4DXJ5Hc+iDprVyiEViJbP0Jsr+xAmmzEtwSb//3spZgK/6LSedYjFfeLIenqCCd0D9cw7cprvlmgn7nrrYXpO3K5txcxS+519+vdRvHBU5RFCU4KnCBiZGvIoHkCSsCma5ks3skH76XuJ5bUPsc9FSFhINnVoo4PCtIV76uxi3/G6QXJFcQbJqaQPyL4RW5DWTCRVB0324uHfwwI9nKUk7jpR/OgiunAMsDvnE57jniQfxLHQT3M/np84tDsygcSpEspiUE74sVSLu5RGef4GjE170LCez1ztfooMCVA8cizlzXRK0zA8n73RZFwJWwefBoRozYzTPvwPXz7uW7D17NDZvvgXunR21hyJQifStuJIEjvSNudI5ap5Bgymp6K3AOTlET3uGyuMWInbCWCXA2fHYMcCrUUa3pyxUH8cuH9h4HR3BKv9EO3AZn1i2H/9vO4AVNvL+uEv4P8FzEtilKCGQtcMaYAUAtsM1aO8sYcyRSdXEIsAr4srV2b7afo+SADuBl77iyiPepjNggRQmXMKaoVyETY587gXustTXAe8Dcnt/iEGAKcAppx72iKEp2ZCVwxpgq4DPAQu+2AU4FHvee8ghwTs/vNBipXH86cAJurexXFCVfyXYEdy/wLeBv3u0hQJO11o9V1yNx9Y9hjLnMGFNrjKmFHVmaoSiK8nECC5wxZhaw3Vq7KsjrrbUPWGsnS8HWQ4OaoSiK0i3ZBBlOAT5rjDkLWVFVCiwAyowxCW8UV4WkOVAURel3AguctfZ64HoAY8x04Fpr7ZeMMb8CzkciqXOAp3t8s1HA95Co3gZkAX0t0BLUOgeoBqZ5x1CgAYlWLidPKox3QxkwGZiOLJ5vB9Yg57WG+OYeUPKSXKyDuw541BhzG/Aa8FBPLxj9yY384LzpfMjBLOZCfjn+K3At+S1wU2HwwgZuOPh2jmUttUzm9otvoP2S8vwWuErgapjzmfs5l6fYQwkLvzSP3985UzJHNUVrnqJ0JhSBs9Yux9u4aK19CzixL683/I1iL6faAPalNzPkOQMS+ziYvRTTxkF8SCKxL2qTwiEBCfZxMB+yl4M4mA9j02ZKvDDWRp8a3Bw92fJArUxv1gBLgRXk9wiuBpiBTOUqkVHbcuTc6qIyKgTKkGn3DGA8MkWtRXY+rEG3eCk55k3gFbr+iG5ZJcHKj+PGNfcviBDEic3e8aOoDQmZJmCJdyiK4zi42V5RFCUcVOAURYktbkxRFUVRekUSSThaTG/WJKnAKYqSR4xAZGtsp/tu7PbZKnCKouQRA+lL5mL1wSmKEltU4BRFiS0qcIqixBYVOEVRYosKnKIoscWRKOo+uq9zmEAiJy6SAi9JQN9JIGt5XCyPaJHzCpr7qBh3q8DHtc2yOS9wt82y64uOCNx7wOIM9yeRAjQ13uFSx9oFrEV20QfpWKXIOY3FLQH3i+tuAnYHeH0SWas0luAV5HNFPXJu25Dz7CvlyHmNxi0xaEbOq47eFkTuSjGSwHAsbtVDsciG7vVIX+x7mzkicHvpPvFvChjWj7b0lmZE5LYQ7MeyB/nBtOGewDUj7RFE4EC6VXVYBoVIM9BI8HQubcgFN4VbApdC2qqOYBdbf3dAdXgmhUIH6TYLlhhcfXCKosQWFThFUWKLI1NU39eWiXLEX+UaAxG7hhHMAVrsvd6lqQ6kpyvlBLdtiPcerlGM2LYn4Ov9vuhim5Ugv6EgU9QEbraZH9QpJ2iQwY2MvuY4Cy9082gx0nguBRh8Wgke4Ung5o8F0n64bKKoLraZRcQtmyiqq22WbV8sxi1fsE9v2qzS8Yy+BxzBeXS21JnKTQNxs1NkSxK3omlhYRCBcnFGkC1x7YvZtZn7PriZQC3MSv2Kq1Pf5dTUEqlrMC1qwxRFcR23BS4BXALLTjiZ30y7gLPMDSz73Nk8fNqFcE7EtimK4jyOTFEPQAu8w3CYCKdvAiZ6t3tVcau79WkJ3PMPKeJvyeR/0PZyF7fbzJEgw2QrtecyMBm4BJIXNTN8yDts2VYNi4pgEbCuu3fchayArifzl+/vjnBtpX0hU4+0WWOGxxJIW9UQT99gPuLvMqgj84LwYmRHy2hy7/M0rgcZDkCtHKkrS9nS6y/KX9W9mcyjuDYk9DwCF64yikXabBOZV6wnkQvVMFTgXKEDabPuLkr+kpMqogzquC9wSm4ZjxRxnooUda5DClQvBxoisklRQkIFrpApAqbD2HtWcy13UU0dS5nBd6ffCpejAqfkPSpwhc4gGMZ2qqnjKDazkU9RVLmb9kR5716fAIYi+7Qrvft2IiPBBhxas6gUIm4vE1HcpxJYBHf+6WvYHxrs3YaH/vBFeA4YE7VxSqGjAqdkRxkcd9rLfGvRfSweAQ+Ogq/86y8575hF6RGdokRETKeo5cicqbvTq/CeoxHUrGmB11dO5ZGLL2DOkY/Bh/D0qWfwxLbz++DDM0h7jPb+ZiLaaJyyPwkkql1D5nyNfuLTaDfwx1TghnjHiQd4jopbKNQDF8ElYxZzSbWXlfnfgA30Ma9kFfKD6A5tL7cwSAHm6h6eEy0xFTif6L/g2NOBCFldGG+m7ZV/uN1m6oNTFCW2qMApihJbVOAURYktKnCKosSWmAcZFAYhe0yLvNstnQ5FiTk6goszNcAimLv1Pv6w6QSWbTqZC959BO5DF+EqBYEKXJypgjNnP8nCBV9jm1nNh+ZPLP71JUyas0IFTikIdIoaFWXAdGAWkrKoBXgZ2cP5MuFsUm+Clfum8NZVlVz4TgMcAq+eO57V2ybrFFUpCFTgomIyTPr1Cla987/hWmA4vHzXcZx0zhrJzxZGqqLNsHvGCI6a/i5M8u67Hsn1Vh/C+yuK42QlcMaYMmAhMgaxwFeAjcBiZA9HHXCBtfa9bD4nlhTB8ayBH8LiX8JYYOrFr3PoxL+yo+jwcD6jhXTyym5syMS+joSmOVJiQbYjuAXAc9ba840xByGFGW8Alllr7zDGzAfmA9cd+G0+AFZ285hfsdu1VNUpZBi0PdjLl5zIQwuupOn2Mr54+y+opYzLmceO2w+HphDN7CNtDCRVVwproPs26Qk/HbxrdTqbkfTamWoI9IZipC/2UMO337FIqvdGghd+Lkf2A7tW1HoXcl57Ar06sMAZYwYDfwdcAmCt3QvsNcbMRrxLAI8g44ceBK4ZKXaaiRHI/Mq17B+NwGokJ32QTrUSrp7CE1dfzBOVF8td7Z2OiGhloIy7m+rpvk16osb7OzoUm8KjERHtLQFfXw5MQC66Lon3bmA9sJYDV4DvjgQyh0jiViGmFNIZVxN0IJHNCO5IYAfwU2PMccAq4Cqgwlr7rvecBrq53BljLgMuk1uD6b7EX+oAj0WJb1MHwezrQIR9FzS4Njr1Cfq9u9hekLYrbucF6X4Yx3ODoPZls0wkgQyt7rfWHo/MM+d3foKVmoQZ6xJaax+w1k6Wcl8uXQ0VRYkL2QhcPVBvrfUdNY8jgtdojDkMwPsb0EnlOq75KlwhH76XfLAxCEHPK76LKQKfmbW2wRiz1RhztLV2I3Aa8KZ3zAHu8P4+3fO7lZB222V6rBq3/G8g/pixBA9+JBD/oltZagfSKm6Yoiponx7wXcpxzxEPcmIdBPcz+cWMo81S+3FKkQIYJWQXZOhloaF+o7NPcDTiX9yFjJmae/UO2Ur314CfexHUt4BLkVHhY8aYuYg394Ke3+ZAAucqA4Fx3hEfimklWdNMamopLJ8etTkhUwocG7UROSCJuMSPjNqQHNA5ap1Cgimr6ReBs9auASZneOi0bN5X6Sc6gJ2wkU+xkim8w3DWcLwsE9F1cIpzJOmrZMV38q30TAewFBq+MIobxt8jmUfqkTVwGyK1TFFCQQWu0KkjpHoKiuIemk1EUZTYogKnKEpsUYFTFCW2qMApihJbVOAURYktjkRRPwTe9v4v9Y48305TDUwDpiLpwRuAFd6Rz8kmByHnNA1ZPN8OrENyxqxB188pTuGIwDUBvya9fWkSeb8qeyoMXtjAdQffyfG8Ri2Tuf3iG2i/pDy/Ba4KuBou/syDnM/j7KGEhczj93fOlMxRTdGapyidcUTg9pHeelFOsJxW7jEgsY+BtFJMGwfxIYnEvqhNCoeE7FkdSCv7GMDBfCg9yZHepCg+RjIaRWyEGW4/Sg1HDTKCy/M9njVIbYXpyKjHzx+5lPxeWFuGTE9nIonq24FaYAkyRY0wWadSCLwJvELXH9EtqyTt2sfRa26u2OwdP4rakJBpQsRsScR2KEov0CiqosSdGiQV7TqotG+R3NkspaKmEfshjgqcosSdGfDP370b+6rh3VFHsffawTw7d7qIXjeV1eKCCpyixJ1BMJKt8N/w7NvAMvgUG2Fo1Ib1lRR9XYcU8wGqoigshxv+8x6WPzidEx9cyRMM56Ft8+A+8iAo1IhE6HYj4raLvpR9VIFTlLhTC5wDLzCbFxKz04Mg5xdl+7WHVyN1X/uOCpyiFAJ5I2rhoj44RVFiiwqcoiixRQVOUZTYogKnKEpscTDI0IaEgRuRArsJpAZp2Fjvs4J6XYtxN6VTK8HPK4Gcm2uFtkGiakETMbh8XnHui9m0WTOiBcGTbzgocNuRE1qPNFoFstekhvA6Z6v3/pvpbQHZrhQjO+jH4l4F97eRmn+NSOfqC0kkm8tYwv2+w2AX0mZ1BOvwpUiSvrHe/66QAjYh59b79V1pkkiKsbGkq8C7Qj1yXtvoe1/02UPMBC5F14ZOAcNC/gx/lLiNYAKXRESuNUyjQsAiHaKeoOuG5LupQEYTLo0K2hDR3kKwH8tuRNjacE/g/L4YROB8RoRjTqj4bZaNwGWH+uAURYktKnCKosQWB6eo/UESmaaUI1PNvpLwXp+L4Ee2FCPnFdRhPQwowb2uUQwMQewLcm7FyHkFae9c4vfFCoK7BIL241zjt9mebh5vI7sgRM+41ov7iVIka3ANwYWgBPcEziDnVEHwTuMLgUsBBpAfyonABIK1mR9Fda3NkkiAYATZRVFLQrMoPEaQ/q1lopHsAkc9U6ACB9KxhkRtRA4wpCuTxY2BuCdQYVCofTGBBCCS5Erg1AenKEpsUYFTFCW25MEUtYP0urVSxOSQ/UNlSBHjGiTLaROyBngDsDPcj1IUpf/IgxHcbmAt8CzwJFJ3bxOyqDUkLoIz/vQ07z98EPZkw9++b/jCH34Cl5MXlwBFUTKTBwKXQqItm5GIy3rvdohOyfHwDe6hdHSKmy+AjlHwHW6FqeF9hKIo/U8eCFw/UAvf5jZef2s0N6+FuqYqrmKBDBYVRclbdAIG8DisfnkaE6v+AoOAFqAB2dJZYCmeFSVOqMCBCNoG7wD5VsqQgEMVInItSPChpf/NUwqcMu8Y5N1uQYJf2hd7RKeomagGboTKTW8xa+2vOGn9S/A4MJPYF8pVHKMSuByK1uzmjLVPc+raJbA8BRcjoqccEBW4TFQC56SYz50s4Ot8l+s5Y8rTMBEd8yr9yyBgBnx98L+xgKu4m2uYe8SPYRp6se0F+nPNxE5gaZIfz/0qrzGR7VTwwsbZsA71ySn9SwuwAn562qVsZST7GMAz758ttU6dL9ocPcba4OvJjDHfAOYhi9LWApcChwGPIpvrVgFfttbuPfD7DLdwWS8/tRzZvDuJnO1LTCCjuKHIFbQDEb2diB9OUfoL3x9cSXpK2oQEwfJ+Efou4BVk6VeQxLM+t6yy1k7O9EjgEZwxZgTwdWCctbbNGPMYcBFwFnCPtfZRY8yPgLnA/UE/JzP+7oYcCVwHEkGtz83bK0qv6XxxVfpMtj64BFBsjPErw7wLnIq45AEeAc7J8jP2Yw+yk2E58CKwwrsdTUpkRVHcJfAIzlq7zRhzF/BXZDj1AjIlbbLW+p6qerpJFm+MuYyP5qWD+/DJKSTFil9zoBTJEVZOPFPOKIoSlMAjOGPMJ4HZwJHAcOAQZCFFr7DWPmCtnSxz5zjm+FIU5eO0Ir43vxhNM7mcfWUTRZ0BvG2t3QFgjHkSOAUoM8YkvFFcFcHLOymKEit2IbHITaT96LlNWZ6ND+6vwFRjzEBjjAFOA94Efgec7z1nDvB0diYqihIP2hCR246M4JrJpbhBFgJnrV2JBBNWI7L8CeAB4DrgGmPMZsQp9lAIdiqKovSZrBb6WmtvAm7a7+63kOogiqIokRKDnQwpZKjb6N32q867VJVdUZQoiIHAtSHJMLeTLg1XjZRi02UjilLIxEDgQETOd1YWI2vjDuC8TCAblRPe0YHs69O9fYqSQ/p/MX5MBK4PJICLoXzhNr4x4B6qqeOPnMz9b1wjNRhW9PaNUsg65t0EL0RcgayDdq3IciPZpYUvR87LtfWNrciqpd0BX1+MtFlFaBaFg0XOq5HgfbEcWdUVpmun8/ed8v5upz+FrjAFbh68OmAyoyoaWLQd/v3sJzj+mde4bPrP+iBwjUgAeTPBOlUJMo0uxq2pdCuyTmktwYWgxvs7OhSLwmMbsBLYEvD15ciumRLcEu/dyIb1tQS7KCWQvphERC4stiG/kTrkN6IjuBDIVGYwSZcrUx384ZS/Y9QVj3HxS8ClUMvkPm5o9hsraMO5nncpRfAO6eq+4NR+f4O+3kX8fujauaWIStwglgKXQq4Yu5GrRwK5Ko0FKsTPdhdcsmExl8xYDBcig7B/BpZEY7GiKLkhhgIHXYMO/uhtBB/5TtZ4x23ZfIYuQ8lMPnwvSdwejQUlSfApai6J7ruOqcD1B+Wkl6IEacBiRHRLwzQqBPxlNkmCJSFMIt+Na454SFcQqiJYm5UibVYcplEhUIr4PUsIJnBJYBjSbtnQivim93h2NBLcjxsOKnCBGQiM8444YRABCNPZ7AqlwLFRG5EDkkhAJ8qgjkVcQ6uR4ILvY452pKwCpyhKSPgBvtxuoO8LWlVLUZTYogKnKEpsKYApqr+CejMydE4gvphy3FqsWei0Iu2UKbCRJN1m+RClVVyhAAQO0tGcP3u3q5EV6WNxb5tUIbK/g3p/kkiUcAKSIV9RekeBCBx0XeVdCoyTChJDkTqTdd7REoFpCgd2UHd0c7/iBpb00hC3KEAfXDHUnMkku4Ltz5Zg5xnW/eYojlv7ch9K5ihKoWOBt5HynU8CzyIzpGjXve1PAQpcEmbA3XyTQwa1cPN0OOZ/vcV13AHjo7ZNUfKFDiQziL/Jfz3iXnBrFFdAU1SfFDwFX7n/J/yq5fPc/M56Xhw+jeu5Q7ZvKYXLGGA6MBkYhLgsVgAvk9+V5YcCU4FpiPu5CahFzm1DZFb1CwUocG3Q8CBvmWM5gZeR7S2gwYYCpwiYCWPvWc3X+AEj2crz/AP3zfgWXEl+C9wY4A745jG3MZ3fUceR3PPVb/DW144REY9xotcCFDiQofQ2xG8Asm9yknfoMoRo8NPNZ9rnmezm/vBNOIi9lNBCCS0MpE2EL99JAEWWYlq9c9vDwXyY5a/fL/nndtICY62N2gaMGW7hsggtUIGLngOtgwNJapDDdXAJxAc7E5nKlSFLJ59D/OgNufnYfqESKdM+A1lt04Sc03PAut6+iZ+GzPez+e3Vvxl6M3PLKmvt5EyPFOgITnGPgUS68LqDdBqtuNEALPKOwLQhAreWYFlmokEFDkindtmErJFL4l5aakVR+ooKHCBXpPXInMTfyjUWWTmvIqco+YoK3Ed0XkXvr5x3vW6CovQXbq1v6y0qcIqiZMAPKvjlI9vIrpRkNKjAKYqSAb8U4XryTdQ6owKXkRSwCynsXEF6HZYuIVHiTCtp10w94pvObzeNClxG2pCAwzbS4laD5PN3rUiMooRBI7JZvo50HVP3F/L2hApct3QOOhQji0ybUYFT4olLC3fDQwWuV3SQTsRYR7o03gh0GYmiuIsKXK9IIUP4Ru+2P2UtRgVOUdxFBS4Q/jq5RnIbfHA5sOH7aILgb6x3MYOLJfgayATSXq62WSuZz8uvW5K/0dLuUIELRAqZsu5G6gjkqkOPQAIbrlWJryed4DAIJchOkdG4JQb+jpY6gv3YE0jCtbFIcgBXsEjQbD2ZM+76F6v8DyrsjwpcYHJd4DaJNE8z7glcMyJudQFfX46cUwq3BK4NEYA6grWtH3GvDs+kUOhAzmsbaTdLYVCAKcsVRSkUVOAURYktOkWNFD+IkKkZkogfpx8y2fYZf11gUJ9NBem0VC5RjPgHKwiW86wYd9usMFGBiwzfV1NDejtYpueUZLg/aqqRH3I2UVQXBa4Uyeo8luBblDSPoEuowEWGv1i4Grcibr3BIGIQx10dEWcWVkJFBU5RYsku0mvb2pClPfFb59YTPQYZjDE/McZsN8as63RfuTHmRWPMJu/vJ737jTHmB8aYzcaYPxtjJuXSeEVRMtGKrHlbjlSOW4oUQM2fWgph0Zso6sNIraHOzAeWWWtHA8u82wBnIqs3RyNlsu4Px0xFUXrG0nVXwh7SgaB4LeDtLT1OUa21/2WMqd7v7tlIDXCAR5BLxXXe/f9hpRbhy8aYMmPMYdbad0OzWFGU/WhFdirUkd6VEM+tV30lqA+uopNoNZBeaj8C2NrpefXefR8TOGPMZXxUDHVwQDMURREh20a+Z9/NBVkHGay11hjT5+rR1toHgAfAL/ysKErvsKSLZHcg2692ke/Zd3NBUIFr9KeexpjDkCx5IJeRkZ2eV0XwHdmKomRkO5LkYRNdK8EVpp/tQATdqvUMMMf7fw7wdKf7/9GLpk4F3lf/W3d0zuAQswFsEXJpGwOMR5b6lZH/i5ISwFBkbfZ45PwqkfPtN1LIyK2ZdBAhfllAwqLHLmeM+SUSUBhqjKkHbgLuAB4zxswFtgAXeE//LXAW4vFsBS7Ngc0xoYN01gp/y9YIRA3yeAFtETALuBZOnPJ7hrCLWiaz4/uHw4+QnpGvjAGuhiPmbmAcb9LIMFYvmwZ3ISsxcjZD3IX0FT+deDNxSy2eK3oTRf1CNw+dluG5FrgiW6MKAz+c7+fnSiLXhHLyWuASwEQ4c8qTXMedjOSvPMW5fPP8BfBUMr8FrgoGXbSDb3M7Z/A8f+FobjztNl557tMicDnDL+GXz19eNOT7pCFm7CEd6vdzi5XjZubbbugANsCz/3MuA47axxB28gpT4Lkk7IzauCxpgJYlh/KjC7/KSqawlZG8surTOdAdf5TmTz/rkL6h9BUjg66IjTDD7UcrRgoaP7OIL24jgAnAkVEa1XfKEN/UUOR0WpDFRA3kd6CvCDmnSmAQ0I6IdgNyjqFgkdHaWiQ66gcPNIjQPbesstZOzvSIjuCcwl9x3oZcwUtIO5Bdy7xxAJq8I260Iys763P5If4OhGYypxdX+oIKnNP4ywG2IE2Vr9lHlO6xyEqqbaSnpZ19s0o2qMA5jb8cwHfyVJNOs5RHfjnlAPg+Nn9KqoSJClxe4Zcq7C6poquJJJWu+EGEDrqmNVLCRgUur9iNXOnXk1nESpBELhPQpI2u0oq04SbE19Z5wbcSNipweUVPaW/akLwHbajAucoe5EK1nULMz9bfqMDFCt+f04ZMVYuBYci+KZ229g+tSBtkChJ0IKLWiI7Y+gcVuFiRIh2RAxG4Y5HRnGvFo+NKIzIF3YyuW4serYsae/RHphQujuxkMDuQxV5DyY8NPWpnuOSDnflgIxSmnUdYaw/N9IATAudjjKntbsuFS6id4ZIPduaDjaB27o9OURVFiS0qcIqixBbXBO6BqA3oJWpnuOSDnflgI6idXXDKB6coihImro3gFEVRQkMFTlGU2OKMwBljZhpjNhpjNhtj5kdtj48x5ifGmO3GmHWd7is3xrxojNnk/f1kxDaONMb8zhjzpjHmDWPMVY7aWWSMecUY87pn5y3e/UcaY1Z6bb/YGHNQlHb6GGMGGGNeM8Ys8W47Z6cxps4Ys9YYs8YYU+vd51q7lxljHjfGbDDGrDfGnNRfNjohcMaYAcAPgTOBccAXjDHjorXqIx4GZu5333xgmbV2NLDMux0lHcA3rbXjgKnAFd7355qdHwKnWmuPAyYCM73ykncC91hra4D3gLnRmdiFq5DULT6u2vn31tqJndaVudbuC4DnrLVjgOOQ77R/bLTWRn4AJwHPd7p9PXB91HZ1sqcaWNfp9kbgMO//w4CNUdu4n71PA6e7bCeyQXY1MAVZ0Z7I1BcitK/K++GdCixBMoy6aGcdMHS/+5xpd2Aw8DZeQLO/bXRiBIdUV9na6Xa9d5+rVNh0QesGHNrJboypBo4HVuKgnd60bw2SL+hF4H+AJmutX47Glba/F/gW8Dfv9hDctNMCLxhjVhlj/MpNLrX7kcAO4KfedH+hMeYQ+slGVwQub7FyCXJirY0xZhDwBHC1tbZLsjFX7LTW7rPWTkRGSCci5ZSdwhgzC9hurV0VtS29YJq1dhLi3rnCGPN3nR90oN0TwCTgfmvt8cAH7DcdzaWNrgjcNmBkp9tVpHP+uEijMeYwAO/v9ojtwRiTRMTt59baJ727nbPTx1rbBPwOmeqVGWP81F0utP0pwGeNMXXAo8g0dQHu2Ym1dpv3dzvwa+Si4VK71wP11tqV3u3HEcHrFxtdEbhXgdFelOog4CLgmYhtOhDPAHO8/+cgPq/IMMYY4CFgvbX27k4PuWbnocaYMu//YsRPuB4RuvO9p0Vup7X2emttlbW2GumLL1lrv4RjdhpjDjHGlPj/A2cA63Co3a21DcBWY8zR3l2nAW/SXzZG7STt5HQ8C/gL4pP5dtT2dLLrl8C7SGK1eiRyNgRxQG8ClgLlEds4DRni/xlY4x1nOWjnscBrnp3rgO94948CXkGyRP4KODjqdu9k83RgiYt2eva87h1v+L8bB9t9IlDrtftTwCf7y0bdqqUoSmxxZYqqKIoSOipwiqLEFhU4RVFiiwqcoiixRQVOUZTYogKnKEpsUYFTFCW2/H9e/El1I55/mwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# exploration of the scratch class\n",
    "\n",
    "for x in range(90,95):\n",
    "    wafer_map = load_map(x, SCRATCH_CLASS)\n",
    "    visualize_map(wafer_map)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7b3f44e",
   "metadata": {},
   "source": [
    "### Loading the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1874b4e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# loading the dataset\n",
    "(x_all, y_all) = load_maps()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc0a6014",
   "metadata": {},
   "source": [
    "Here I calculated how many multilabeled maps there is."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "6dc60561",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Multilabeled wafer bin maps: 214\n"
     ]
    }
   ],
   "source": [
    "arrs, idxs, counts = np.unique(x_all,axis=0, return_counts=True, return_index=True)\n",
    "\n",
    "y_multi = []\n",
    "for i in range(len(y_all)):\n",
    "    y_multi.append(set())\n",
    "    y_multi[i].add(y_all[i])\n",
    "\n",
    "for i in idxs[counts != 1]:\n",
    "    for x in range(x_all.shape[0]):\n",
    "        if (x_all[i] == x_all[x]).all(): # both maps equal\n",
    "           y_multi[i].add(y_all[x]) # therefore add that label to the label set\n",
    "\n",
    "total = 0\n",
    "for y in y_multi:\n",
    "    if len(y) > 1:\n",
    "        total += 1\n",
    "        \n",
    "print(\"Multilabeled wafer bin maps:\", end=' ')\n",
    "print(total)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "85ca10be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAD4CAYAAAAO9oqkAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAASn0lEQVR4nO3de5DdZX3H8fenRARECciWYoJu1KAigwoR8FoVDDenQQctViVQ2rSKt45jBXUm44UWqy2KVioDwaAWpIhDFCzGIKNVuSyCXEW2XCQRJBpE8QIEv/3jPFsPcRd292z2bMj7NXPm/H7P8/x+53nOnnM+v9s5m6pCkrR5+5N+d0CS1H+GgSTJMJAkGQaSJAwDSRIwq98dmKwdd9yxBgcH+90NSdpkXHHFFT+rqoHR6jbZMBgcHGRoaKjf3ZCkTUaS28aq8zCRJMkwkCQZBpIkDANJEoaBJAnDQJLEOMIgybIkdyW5tqtshyQrk9zU7rdv5UlyUpLhJFcn2bNrmcWt/U1JFneV75XkmrbMSUky1YOUJD288ewZfBY4cIOyY4FVVTUfWNXmAQ4C5rfbEuBk6IQHsBTYB9gbWDoSIK3N33Ytt+FjSZI2skcMg6r6FrBug+JFwPI2vRw4tKv8jOq4BJidZGfgAGBlVa2rqruBlcCBre4JVXVJdf6xwhld65IkTZPJfgN5p6q6o03fCezUpucAt3e1W93KHq589Sjlo0qyhM4eB09+8pMn2XUYPPb8SS8709x6wiH97oKkR4GeTyC3Lfpp+XdpVXVKVS2oqgUDA6P+vIYkaRImGwY/bYd4aPd3tfI1wC5d7ea2socrnztKuSRpGk02DFYAI1cELQbO6yo/ol1VtC9wTzucdCGwMMn27cTxQuDCVvfLJPu2q4iO6FqXJGmaPOI5gyRnAi8Ddkyyms5VQScAZyc5GrgNeF1rfgFwMDAM/AY4CqCq1iX5EHB5a/fBqho5Kf0WOlcsbQ18rd0kSdPoEcOgql4/RtV+o7Qt4Jgx1rMMWDZK+RCw+yP1Q5K08fgNZEmSYSBJMgwkSRgGkiQMA0kShoEkCcNAkoRhIEnCMJAkYRhIkjAMJEkYBpIkDANJEoaBJAnDQJKEYSBJwjCQJGEYSJIwDCRJGAaSJAwDSRKGgSQJw0CShGEgScIwkCRhGEiSMAwkSRgGkiQMA0kShoEkCcNAkoRhIEmixzBI8g9JrktybZIzk2yVZF6SS5MMJ/liki1b28e2+eFWP9i1nuNa+Y1JDuhxTJKkCZp0GCSZA7wdWFBVuwNbAIcDHwFOrKqnA3cDR7dFjgbubuUntnYk2a0t92zgQODTSbaYbL8kSRPX62GiWcDWSWYB2wB3AK8Azmn1y4FD2/SiNk+r3y9JWvlZVXVfVd0CDAN799gvSdIETDoMqmoN8DHgx3RC4B7gCuAXVbW+NVsNzGnTc4Db27LrW/sndpePssxDJFmSZCjJ0Nq1ayfbdUnSBno5TLQ9na36ecCTgMfROcyz0VTVKVW1oKoWDAwMbMyHkqTNSi+HifYHbqmqtVX1AHAu8CJgdjtsBDAXWNOm1wC7ALT67YCfd5ePsowkaRr0EgY/BvZNsk079r8fcD3wTeCw1mYxcF6bXtHmafUXVVW18sPb1UbzgPnAZT30S5I0QbMeucnoqurSJOcA3wfWA1cCpwDnA2cl+XArO60tchrwuSTDwDo6VxBRVdclOZtOkKwHjqmqByfbL0nSxE06DACqaimwdIPimxnlaqCq+h3w2jHWczxwfC99kSRNnt9AliQZBpIkw0CShGEgScIwkCRhGEiSMAwkSRgGkiQMA0kShoEkCcNAkoRhIEnCMJAkYRhIkjAMJEkYBpIkDANJEoaBJAnDQJKEYSBJwjCQJGEYSJIwDCRJGAaSJAwDSRKGgSQJw0CShGEgScIwkCRhGEiSMAwkSRgGkiR6DIMks5Ock+SHSW5I8oIkOyRZmeSmdr99a5skJyUZTnJ1kj271rO4tb8pyeJeByVJmphe9ww+Afx3VT0TeA5wA3AssKqq5gOr2jzAQcD8dlsCnAyQZAdgKbAPsDewdCRAJEnTY9JhkGQ74KXAaQBVdX9V/QJYBCxvzZYDh7bpRcAZ1XEJMDvJzsABwMqqWldVdwMrgQMn2y9J0sT1smcwD1gLnJ7kyiSnJnkcsFNV3dHa3Ans1KbnALd3Lb+6lY1VLkmaJr2EwSxgT+Dkqnoe8Gv+cEgIgKoqoHp4jIdIsiTJUJKhtWvXTtVqJWmz10sYrAZWV9Wlbf4cOuHw03b4h3Z/V6tfA+zStfzcVjZW+R+pqlOqakFVLRgYGOih65KkbpMOg6q6E7g9yTNa0X7A9cAKYOSKoMXAeW16BXBEu6poX+CedjjpQmBhku3bieOFrUySNE1m9bj824AvJNkSuBk4ik7AnJ3kaOA24HWt7QXAwcAw8JvWlqpal+RDwOWt3Qeral2P/ZIkTUBPYVBVVwELRqnab5S2BRwzxnqWAct66YskafL8BrIkyTCQJBkGkiQMA0kShoEkCcNAkoRhIEnCMJAkYRhIkjAMJEkYBpIkDANJEoaBJAnDQJKEYSBJwjCQJGEYSJIwDCRJGAaSJAwDSRKGgSQJw0CShGEgScIwkCRhGEiSMAwkSRgGkiQMA0kShoEkCcNAkoRhIEnCMJAkYRhIkpiCMEiyRZIrk3y1zc9LcmmS4SRfTLJlK39smx9u9YNd6ziuld+Y5IBe+yRJmpip2DN4B3BD1/xHgBOr6unA3cDRrfxo4O5WfmJrR5LdgMOBZwMHAp9OssUU9EuSNE49hUGSucAhwKltPsArgHNak+XAoW16UZun1e/X2i8Czqqq+6rqFmAY2LuXfkmSJqbXPYOPA/8I/L7NPxH4RVWtb/OrgTlteg5wO0Crv6e1///yUZZ5iCRLkgwlGVq7dm2PXZckjZh0GCR5FXBXVV0xhf15WFV1SlUtqKoFAwMD0/WwkvSoN6uHZV8E/EWSg4GtgCcAnwBmJ5nVtv7nAmta+zXALsDqJLOA7YCfd5WP6F5GkjQNJr1nUFXHVdXcqhqkcwL4oqp6A/BN4LDWbDFwXpte0eZp9RdVVbXyw9vVRvOA+cBlk+2XJGnietkzGMt7gLOSfBi4EjitlZ8GfC7JMLCOToBQVdclORu4HlgPHFNVD26EfkmSxjAlYVBVFwMXt+mbGeVqoKr6HfDaMZY/Hjh+KvoiSZo4v4EsSTIMJEmGgSQJw0CShGEgScIwkCRhGEiSMAwkSRgGkiQMA0kShoEkCcNAkoRhIEnCMJAkYRhIkjAMJEkYBpIkNs6/vZRmrMFjz+93F6bErScc0u8u6FHGPQNJkmEgSTIMJEkYBpIkDANJEoaBJAnDQJKEYSBJwjCQJGEYSJIwDCRJGAaSJAwDSRKGgSSJHsIgyS5Jvpnk+iTXJXlHK98hycokN7X77Vt5kpyUZDjJ1Un27FrX4tb+piSLex+WJGkietkzWA+8q6p2A/YFjkmyG3AssKqq5gOr2jzAQcD8dlsCnAyd8ACWAvsAewNLRwJEkjQ9Jh0GVXVHVX2/Tf8KuAGYAywClrdmy4FD2/Qi4IzquASYnWRn4ABgZVWtq6q7gZXAgZPtlyRp4qbknEGSQeB5wKXATlV1R6u6E9ipTc8Bbu9abHUrG6t8tMdZkmQoydDatWunouuSJKYgDJJsC3wJeGdV/bK7rqoKqF4fo2t9p1TVgqpaMDAwMFWrlaTNXk9hkOQxdILgC1V1biv+aTv8Q7u/q5WvAXbpWnxuKxurXJI0TXq5mijAacANVfVvXVUrgJErghYD53WVH9GuKtoXuKcdTroQWJhk+3bieGErkyRNk1k9LPsi4E3ANUmuamXvBU4Azk5yNHAb8LpWdwFwMDAM/AY4CqCq1iX5EHB5a/fBqlrXQ78kSRM06TCoqv8BMkb1fqO0L+CYMda1DFg22b5IknrjN5AlSYaBJMkwkCRhGEiSMAwkSRgGkiQMA0kShoEkCcNAkoRhIEnCMJAkYRhIkjAMJEn09hPWkjYhg8ee3+8uTJlbTzik31141HHPQJJkGEiSDANJEoaBJAnDQJKEYSBJwjCQJGEYSJIwDCRJGAaSJAwDSRKGgSQJw0CShGEgScKfsN7s+DPGkkbjnoEkyTCQJBkGkiQMA0kSM+gEcpIDgU8AWwCnVtUJfe6SpEcJL5x4ZDNizyDJFsC/AwcBuwGvT7Jbf3slSZuPGREGwN7AcFXdXFX3A2cBi/rcJ0nabKSq+t0HkhwGHFhVf9Pm3wTsU1Vv3aDdEmBJm30GcOO0dnRidgR+1u9O9NHmPH7Hvvma6eN/SlUNjFYxY84ZjEdVnQKc0u9+jEeSoapa0O9+9MvmPH7HvnmOHTbt8c+Uw0RrgF265ue2MknSNJgpYXA5MD/JvCRbAocDK/rcJ0nabMyIw0RVtT7JW4EL6Vxauqyqrutzt3q1SRzO2og25/E79s3XJjv+GXECWZLUXzPlMJEkqY8MA0mSYTARSf4syVlJ/jfJFUkuSLJrkmv73bfpkOTBJFd13QaTfHccy907Rvln23dMZrQkleTzXfOzkqxN8tVJrGtwU3m9dP29r03yX0m2mcCyRyb51Mbs31RL8r4k1yW5uo17nx7XNzvJW8bR7uIkfb8c1TAYpyQBvgxcXFVPq6q9gOOAnaZq/Ulm+t/jt1X13K7brVX1wn53ahr8Gtg9ydZt/pVsHpc+j/y9dwfuB/6+uzLJjLgAZSokeQHwKmDPqtoD2B+4fRzLPdxzMBt4xDCYKWb6h89M8nLggar6j5GCqvoBXS+YJFsk+WiSy9vWxd+18m2TrEry/STXJFnUygeT3JjkDOBaHvpdi01C91Z/knd3jf0Do7RNkk+1MX8D+NNp7WxvLgBGfiHs9cCZIxVJ9k7yvSRXJvlukme08mcnuaxtZV6dZH73CpM8tS3z/GkbxeR9G3h6kpcl+XaSFcD1SbZKcnp7XV+Z5OVdy+zStnpvSrJ0pDDJG7uel8+03ybrt52Bn1XVfQBV9bOq+kmS57e/6Q9anx/f9npWJLkIWDXW+xs4AXhaG+dHAZK8p7X5QZLuH+N8bVv/j5K8ZFpHPqKqvI3jBrwdOHGU8kHg2ja9BHh/m34sMATMo3MJ7xNa+Y7AMJC27O+Bffs9vnE+Bw8CV7Xbl1vZve1+IZ3L6kJnI+OrwEs3aPMaYCWdy4efBPwCOKzf4xrHuO8F9gDOAbZq438Z8NVW/wRgVpveH/hSm/4k8IY2vSWw9cjrhc7PqVwJPKff43u4cbf7WcB5wJvbuH8NzGt176JzKTjAM4Eft+foSOAO4Ilt3NcCC4BnAV8BHtOW+TRwxAwY67bt7/qj1qc/b3+zm4Hnd/+d29hWAzt0PT9jvb+v7XqMg4DvAtu0+ZHlLwb+tU0fDHyjH8/Bo2Y3b4ZYCOzRdRx8O2A+nRfOPyV5KZ0P/zn84fDSbVV1ybT3dHJ+W1XPHaNuYbtd2ea3pTP2b3W1eSlwZlU9CPykbVltEqrq6iSDdPYKLtigejtgedvyL+Axrfx7wPuSzAXOraqbOkcbGaDz4fqaqrp+Ovo/SVsnuapNfxs4DXghcFlV3dLKX0wn9KiqHya5Ddi11a2sqp8DJDm3tV0P7AVc3p6LrYG7Nv5QHl5V3ZtkL+AldI4CfBE4Hrijqi5vbX4J0Pq9sqrWtcXD2O/vbvsDp1fVb9r61nXVndvur6ATItPOMBi/64BHOtkZ4G1VdeFDCpMj6XwA7FVVDyS5lc7WE3S2sh4NAvxzVX2m3x3ZiFYAH6OzdfzErvIPAd+sqle3wLgYoKr+M8mldA4vXdAOG94M3ENnC/rFwEwOgz8K//ZBON7X7IZfYio6r5PlVXVcz72bYm0j5WLg4iTXAMc8TPPu5+ANjP3+Hq/72v2D9Olz2XMG43cR8Nh0fjkVgCR78NDj/BcCb07ymFa/a5LH0dlyvKu9UF4OPGUa+z1dLgT+Osm2AEnmJNnwnMC3gL9s51Z2prMFtilZBnygqq7ZoHw7/nBC+ciRwiRPBW6uqpPo7Ans0aruB14NHJHkrzZqjze+b9P5MCTJrsCT+cOvCb8yyQ7txPuhwHeAVcBhI6+NVt/390OSZ2xwTue5wA3AziPndNr5gtE+qMd6f/8KeHxXu5XAUWlXZSXZYYqH0RP3DMapqirJq4GPJ3kP8DvgVuCdXc1OpbOL9/10NqHW0nkTfAH4StvaGAJ+OG0dnyZV9fUkzwK+17Ye7wXeyEMPAXwZeAWdreEf0zmMssmoqtXASaNU/Qudw0TvB7r/pdbrgDcleQC4E/gnOsedqapfJ3kVsDLJvVW1qf4W16eBk9trez1wZFXd114DlwFfovPDk5+vqiGA9jx9PZ2r5x6gswV+Wz8632Vb4JNJZtMZxzCdc4Cnt/Ktgd/SOdSzoVHf31X18yTfSedS4q9V1buTPBcYSnI/ncON7924wxo/f45CkuRhIkmSYSBJwjCQJGEYSJIwDCRJGAaSJAwDSRLwf+31Y38nmPteAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Samples without defect:10000\n",
      "Samples with defect:4474\n"
     ]
    }
   ],
   "source": [
    "# data count for each class\n",
    "counts = []\n",
    "names = ['Clear', 'Field', 'Mask', 'Probe', 'Scratch']\n",
    "for defect_type in range(5):\n",
    "    count = len(os.listdir('./data/' + defect_class_to_str(defect_type)))\n",
    "    counts.append(count)\n",
    "\n",
    "plt.figure()\n",
    "plt.bar(names,counts)\n",
    "plt.show()\n",
    "\n",
    "print('Samples without defect:' + str(counts[0]))\n",
    "print('Samples with defect:' + str(sum(counts[1:])))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "66da24e4",
   "metadata": {},
   "source": [
    "## 1. CNN - Defect detection"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f6a7ac0",
   "metadata": {},
   "source": [
    "### 1.1 Data transformation and class balancing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "66a6cbe9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# converting labels into binary, 0 = clear, 1 = defect\n",
    "y_binary = (y_all != 0).astype(np.uint8)\n",
    "x_train, x_test, y_train, y_test = train_test_split(x_all, y_binary, test_size=.35, random_state=7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "c15545e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# class balancing\n",
    "under_sampler = RandomUnderSampler(random_state=10)\n",
    "\n",
    "n, x, y = x_train.shape\n",
    "x_train_d2 = x_train.reshape((n,x*y))\n",
    "x_resampled, y_resampled = under_sampler.fit_resample(x_train_d2, y_train)\n",
    "x_resampled = x_resampled.reshape((x_resampled.shape[0],x,y))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cee97962",
   "metadata": {},
   "source": [
    "### 1.2 Model training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "3a9a802b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_4\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d_12 (Conv2D)          (None, 105, 63, 32)       320       \n",
      "                                                                 \n",
      " max_pooling2d_12 (MaxPoolin  (None, 52, 31, 32)       0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " conv2d_13 (Conv2D)          (None, 50, 29, 64)        18496     \n",
      "                                                                 \n",
      " max_pooling2d_13 (MaxPoolin  (None, 25, 14, 64)       0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " conv2d_14 (Conv2D)          (None, 23, 12, 128)       73856     \n",
      "                                                                 \n",
      " max_pooling2d_14 (MaxPoolin  (None, 11, 6, 128)       0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " flatten_4 (Flatten)         (None, 8448)              0         \n",
      "                                                                 \n",
      " dense_8 (Dense)             (None, 256)               2162944   \n",
      "                                                                 \n",
      " dropout_4 (Dropout)         (None, 256)               0         \n",
      "                                                                 \n",
      " dense_9 (Dense)             (None, 1)                 257       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 2,255,873\n",
      "Trainable params: 2,255,873\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Epoch 1/15\n",
      "12/12 [==============================] - 2s 115ms/step - loss: 0.8553 - accuracy: 0.5335 - val_loss: 0.4756 - val_accuracy: 0.9092\n",
      "Epoch 2/15\n",
      "12/12 [==============================] - 1s 87ms/step - loss: 0.3177 - accuracy: 0.8827 - val_loss: 0.2140 - val_accuracy: 0.8995\n",
      "Epoch 3/15\n",
      "12/12 [==============================] - 1s 87ms/step - loss: 0.1647 - accuracy: 0.9331 - val_loss: 0.1468 - val_accuracy: 0.9356\n",
      "Epoch 4/15\n",
      "12/12 [==============================] - 1s 87ms/step - loss: 0.1069 - accuracy: 0.9573 - val_loss: 0.0768 - val_accuracy: 0.9751\n",
      "Epoch 5/15\n",
      "12/12 [==============================] - 1s 88ms/step - loss: 0.0672 - accuracy: 0.9770 - val_loss: 0.0565 - val_accuracy: 0.9836\n",
      "Epoch 6/15\n",
      "12/12 [==============================] - 1s 88ms/step - loss: 0.0597 - accuracy: 0.9790 - val_loss: 0.0520 - val_accuracy: 0.9812\n",
      "Epoch 7/15\n",
      "12/12 [==============================] - 1s 88ms/step - loss: 0.0494 - accuracy: 0.9819 - val_loss: 0.0428 - val_accuracy: 0.9866\n",
      "Epoch 8/15\n",
      "12/12 [==============================] - 1s 88ms/step - loss: 0.0359 - accuracy: 0.9868 - val_loss: 0.0470 - val_accuracy: 0.9842\n",
      "Epoch 9/15\n",
      "12/12 [==============================] - 1s 92ms/step - loss: 0.0288 - accuracy: 0.9899 - val_loss: 0.0402 - val_accuracy: 0.9868\n",
      "Epoch 10/15\n",
      "12/12 [==============================] - 1s 91ms/step - loss: 0.0293 - accuracy: 0.9892 - val_loss: 0.0395 - val_accuracy: 0.9862\n",
      "Epoch 11/15\n",
      "12/12 [==============================] - 1s 87ms/step - loss: 0.0248 - accuracy: 0.9924 - val_loss: 0.0426 - val_accuracy: 0.9872\n",
      "Epoch 12/15\n",
      "12/12 [==============================] - 1s 91ms/step - loss: 0.0246 - accuracy: 0.9911 - val_loss: 0.0530 - val_accuracy: 0.9840\n",
      "Epoch 13/15\n",
      "12/12 [==============================] - 1s 88ms/step - loss: 0.0186 - accuracy: 0.9943 - val_loss: 0.0391 - val_accuracy: 0.9872\n",
      "Epoch 14/15\n",
      "12/12 [==============================] - 1s 87ms/step - loss: 0.0171 - accuracy: 0.9952 - val_loss: 0.0545 - val_accuracy: 0.9818\n",
      "Epoch 15/15\n",
      "12/12 [==============================] - 1s 87ms/step - loss: 0.0162 - accuracy: 0.9952 - val_loss: 0.0408 - val_accuracy: 0.9886\n"
     ]
    }
   ],
   "source": [
    "model_binary = models.Sequential()\n",
    "\n",
    "model_binary.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(107, 65, 1)))\n",
    "model_binary.add(layers.MaxPooling2D())\n",
    "model_binary.add(layers.Conv2D(64, (3, 3), activation='relu', strides=(1, 1)))\n",
    "model_binary.add(layers.MaxPooling2D((2, 2)))\n",
    "model_binary.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "model_binary.add(layers.MaxPooling2D((2, 2)))\n",
    "model_binary.add(layers.Flatten())\n",
    "model_binary.add(layers.Dense(256, activation='sigmoid'))\n",
    "model_binary.add(layers.Dropout(0.3))\n",
    "model_binary.add(layers.Dense(1, activation='sigmoid'))\n",
    "\n",
    "model_binary.summary()\n",
    "\n",
    "model_binary.compile(optimizer='adam',\n",
    "              loss='binary_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "history = model_binary.fit(x_resampled, y_resampled,\n",
    "                    epochs=15,\n",
    "                    validation_data=(x_test, y_test),\n",
    "                    batch_size=500)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7133d0a9",
   "metadata": {},
   "source": [
    "### 1.3 Model evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "bfc7f982",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Clear class metrics:\n",
      "Precision =  0.9911403258073735\n",
      "Recall =  0.992274678111588\n",
      "Fscore=  0.9917071775807835\n",
      "=========\n",
      "Defect class metrics:\n",
      "Precision =  0.982769623484365\n",
      "Recall =  0.98026734563972\n",
      "Fscore=  0.9815168897386871\n"
     ]
    }
   ],
   "source": [
    "predictions = np.round(model_binary.predict(x_test))\n",
    "precision, recall, fscore, support = precision_recall_fscore_support(predictions,y_test)\n",
    "print(\"Clear class metrics:\")\n",
    "print(\"Precision = \", precision[0])\n",
    "print(\"Recall = \", recall[0])\n",
    "print(\"Fscore= \", fscore[0])\n",
    "print(\"=========\")\n",
    "print(\"Defect class metrics:\")\n",
    "print(\"Precision = \", precision[1])\n",
    "print(\"Recall = \", recall[1])\n",
    "print(\"Fscore= \", fscore[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0986f1fc",
   "metadata": {},
   "source": [
    "## 2. CNN pattern classification"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4410aaa4",
   "metadata": {},
   "source": [
    "### 2.1 Data preparation and balancing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "53e0921f",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_defects = np.zeros((sum(counts[1:]),107,65)) # counts refer to sample count for each class, so we start from first non-clear sample\n",
    "y_defects = np.zeros((sum(counts[1:])))\n",
    "x_defects = x_all[counts[0]:]\n",
    "y_defects = y_all[counts[0]:] - 1 # so we have classes starting from 0\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(x_defects, y_defects, test_size=.35, random_state=14)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 993,
   "id": "07075410",
   "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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 241 1889  700   78]\n"
     ]
    }
   ],
   "source": [
    "arr, cnts = np.unique(y_train, return_counts=True)\n",
    "\n",
    "plt.figure()\n",
    "plt.bar([\"field\",\"mask\",\"probe\",\"scratch\"],cnts)\n",
    "plt.show()\n",
    "print(cnts)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "988a6c04",
   "metadata": {},
   "source": [
    "Class balancing in that case was a bit more complicated, as I mentioned in the report I upscaled the scratch class samples by rotating and fliping existing simples.\n",
    "\n",
    "Rest of the classes were downsampled to the count of the newly upssampled scratch samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "40e98054",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_scratch = x_train[y_train==3.0]\n",
    "x_scratch_rotated = [np.rot90(np.rot90(wm)) for wm in x_scratch]\n",
    "x_scratch_flipped = [np.fliplr(wm) for wm in x_scratch]\n",
    "x_scratch_upsampled = np.concatenate((x_scratch, x_scratch_rotated, x_scratch_flipped)) # upsampled scratch samples\n",
    "\n",
    "cnt = x_scratch_upsampled.shape[0]\n",
    "\n",
    "x_train = np.concatenate((x_train[y_train != 3.0], x_scratch_upsampled)) # concat all non scratch samples with new upsampled samples\n",
    "y_train = np.concatenate((y_train[y_train != 3.0], [3.0] * cnt))\n",
    "\n",
    "under_sampler = RandomUnderSampler(sampling_strategy={0.0:cnt, 1.0:cnt, 2.0:cnt}, random_state=10) # downsample rest\n",
    "\n",
    "n, x, y = x_train.shape\n",
    "x_train_d2 = x_train.reshape((n,x*y))\n",
    "x_resampled, y_resampled = under_sampler.fit_resample(x_train_d2, y_train)\n",
    "x_resampled = x_resampled.reshape((x_resampled.shape[0],x,y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "ae115bfa",
   "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"
    }
   ],
   "source": [
    "arr, cnts = np.unique(y_resampled, return_counts=True)\n",
    "\n",
    "plt.figure()\n",
    "plt.bar([\"field\",\"mask\",\"probe\",\"scratch\"],cnts)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92082c6e",
   "metadata": {},
   "source": [
    "### 2.2 Model training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "05a23faf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d_6 (Conv2D)           (None, 105, 63, 32)       320       \n",
      "                                                                 \n",
      " max_pooling2d_6 (MaxPooling  (None, 52, 31, 32)       0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " conv2d_7 (Conv2D)           (None, 47, 26, 64)        73792     \n",
      "                                                                 \n",
      " max_pooling2d_7 (MaxPooling  (None, 23, 13, 64)       0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " conv2d_8 (Conv2D)           (None, 18, 8, 128)        295040    \n",
      "                                                                 \n",
      " max_pooling2d_8 (MaxPooling  (None, 9, 4, 128)        0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten_2 (Flatten)         (None, 4608)              0         \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 256)               1179904   \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 256)               0         \n",
      "                                                                 \n",
      " dense_5 (Dense)             (None, 4)                 1028      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 1,550,084\n",
      "Trainable params: 1,550,084\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Epoch 1/30\n",
      "13/13 [==============================] - 3s 97ms/step - loss: 1.2779 - accuracy: 0.5800 - val_loss: 0.7242 - val_accuracy: 0.7305\n",
      "Epoch 2/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.7139 - accuracy: 0.7585 - val_loss: 0.5969 - val_accuracy: 0.8148\n",
      "Epoch 3/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.5951 - accuracy: 0.7911 - val_loss: 0.4741 - val_accuracy: 0.8519\n",
      "Epoch 4/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.4752 - accuracy: 0.8293 - val_loss: 0.3880 - val_accuracy: 0.8851\n",
      "Epoch 5/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.4294 - accuracy: 0.8633 - val_loss: 0.4511 - val_accuracy: 0.8423\n",
      "Epoch 6/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.4015 - accuracy: 0.8743 - val_loss: 0.3460 - val_accuracy: 0.8997\n",
      "Epoch 7/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.3597 - accuracy: 0.8907 - val_loss: 0.3339 - val_accuracy: 0.9061\n",
      "Epoch 8/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.3139 - accuracy: 0.9044 - val_loss: 0.2990 - val_accuracy: 0.9138\n",
      "Epoch 9/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.2836 - accuracy: 0.9106 - val_loss: 0.3525 - val_accuracy: 0.8736\n",
      "Epoch 10/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.2406 - accuracy: 0.9233 - val_loss: 0.2992 - val_accuracy: 0.9061\n",
      "Epoch 11/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.2129 - accuracy: 0.9298 - val_loss: 0.3001 - val_accuracy: 0.9074\n",
      "Epoch 12/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.1882 - accuracy: 0.9435 - val_loss: 0.3189 - val_accuracy: 0.8908\n",
      "Epoch 13/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.1667 - accuracy: 0.9478 - val_loss: 0.2916 - val_accuracy: 0.9042\n",
      "Epoch 14/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.1589 - accuracy: 0.9510 - val_loss: 0.3363 - val_accuracy: 0.8972\n",
      "Epoch 15/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.1517 - accuracy: 0.9494 - val_loss: 0.3478 - val_accuracy: 0.8780\n",
      "Epoch 16/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.1432 - accuracy: 0.9533 - val_loss: 0.3283 - val_accuracy: 0.8972\n",
      "Epoch 17/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.1155 - accuracy: 0.9612 - val_loss: 0.3558 - val_accuracy: 0.8914\n",
      "Epoch 18/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.1085 - accuracy: 0.9586 - val_loss: 0.5567 - val_accuracy: 0.8519\n",
      "Epoch 19/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.1391 - accuracy: 0.9510 - val_loss: 0.3413 - val_accuracy: 0.9010\n",
      "Epoch 20/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.1086 - accuracy: 0.9595 - val_loss: 0.3381 - val_accuracy: 0.8966\n",
      "Epoch 21/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.0945 - accuracy: 0.9621 - val_loss: 0.3921 - val_accuracy: 0.8812\n",
      "Epoch 22/30\n",
      "13/13 [==============================] - 1s 48ms/step - loss: 0.1032 - accuracy: 0.9644 - val_loss: 0.4266 - val_accuracy: 0.8972\n",
      "Epoch 23/30\n",
      "13/13 [==============================] - 1s 49ms/step - loss: 0.1037 - accuracy: 0.9605 - val_loss: 0.3824 - val_accuracy: 0.8914\n"
     ]
    }
   ],
   "source": [
    "model_defects = models.Sequential()\n",
    "\n",
    "model_defects.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(107, 65, 1)))\n",
    "model_defects.add(layers.MaxPooling2D())\n",
    "model_defects.add(layers.Conv2D(64, (6, 6), activation='relu'))\n",
    "model_defects.add(layers.MaxPooling2D((2, 2)))\n",
    "model_defects.add(layers.Conv2D(128, (6, 6), activation='relu'))\n",
    "model_defects.add(layers.MaxPooling2D((2, 2)))\n",
    "model_defects.add(layers.Flatten())\n",
    "model_defects.add(layers.Dense(256, activation='relu'))\n",
    "model_defects.add(layers.Dropout(0.5))\n",
    "model_defects.add(layers.Dense(4, activation='softmax'))\n",
    "\n",
    "model_defects.summary()\n",
    "\n",
    "earlyStopping = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=10, verbose=0, mode='min', restore_best_weights=True)\n",
    "\n",
    "model_defects.compile(optimizer='adam',\n",
    "              loss='sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "history = model_defects.fit(x_train, y_train,\n",
    "                    epochs=30,\n",
    "                    validation_data=(x_test, y_test),\n",
    "                    batch_size=250,\n",
    "                    callbacks=[earlyStopping])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7dca9256",
   "metadata": {},
   "source": [
    "### 2.3 Model evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "76e3996f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.94      0.96      0.95       127\n",
      "         1.0       0.94      0.95      0.95      1008\n",
      "         2.0       0.83      0.82      0.82       380\n",
      "         3.0       0.51      0.45      0.48        51\n",
      "\n",
      "    accuracy                           0.90      1566\n",
      "   macro avg       0.81      0.80      0.80      1566\n",
      "weighted avg       0.90      0.90      0.90      1566\n",
      "\n"
     ]
    }
   ],
   "source": [
    "predictions = np.argmax(model_defects.predict(x_test), axis=1)\n",
    "print(classification_report(y_test, predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "3b13fd6f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted class: mask\n",
      "Actual class: mask\n",
      "Class probabilites for [field, mask, probe, scratch]: ['0.00', '1.00', '0.00', '0.00']\n"
     ]
    }
   ],
   "source": [
    "# random test of our model\n",
    "def random_map_test():\n",
    "    rnd_example = random.randint(0,x_defects.shape[0])\n",
    "    defect_type = y_defects[rnd_example]\n",
    "    wm = x_defects[rnd_example]\n",
    "    class_probs = model_defects.predict(np.expand_dims(wm,axis=0))\n",
    "    predicted_class = np.argmax(class_probs)\n",
    "\n",
    "    visualize_map(wm)\n",
    "    print(\"Predicted class: \" + defect_class_to_str(predicted_class + 1))\n",
    "    print(\"Actual class: \" + defect_class_to_str(defect_type + 1))\n",
    "    print(\"Class probabilites for [field, mask, probe, scratch]: \", end='')\n",
    "    print([\"{:.2f}\".format(float(x)) for x in class_probs[0]])\n",
    "    \n",
    "random_map_test()"
   ]
  }
 ],
 "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": "3.9.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}