/////////////////////////////////////////// // Running convGAN-proximary-full on folding_car_good /////////////////////////////////////////// Load 'data_input/folding_car_good' from pickle file Data loaded. -> Shuffling data ### Start exercise for synthetic point generator ====== Step 1/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 1/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 18s - loss: 0.4841 49/133 [==========>...................] - ETA: 0s - loss: 0.1871  98/133 [=====================>........] - ETA: 0s - loss: 0.1986 133/133 [==============================] - 0s 1ms/step - loss: 0.1851 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.5218 48/133 [=========>....................] - ETA: 0s - loss: 0.1139 67/133 [==============>...............] - ETA: 0s - loss: 0.1206 112/133 [========================>.....] - ETA: 0s - loss: 0.1123 133/133 [==============================] - 0s 2ms/step - loss: 0.1083 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0385 50/133 [==========>...................] - ETA: 0s - loss: 0.0839 99/133 [=====================>........] - ETA: 0s - loss: 0.0821 133/133 [==============================] - 0s 1ms/step - loss: 0.0761 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0042 50/133 [==========>...................] - ETA: 0s - loss: 0.0808 99/133 [=====================>........] - ETA: 0s - loss: 0.0687 133/133 [==============================] - 0s 1ms/step - loss: 0.0691 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0203 50/133 [==========>...................] - ETA: 0s - loss: 0.0669 98/133 [=====================>........] - ETA: 0s - loss: 0.0665 133/133 [==============================] - 0s 1ms/step - loss: 0.0629 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0455 49/133 [==========>...................] - ETA: 0s - loss: 0.0631 95/133 [====================>.........] - ETA: 0s - loss: 0.0606 133/133 [==============================] - 0s 1ms/step - loss: 0.0617 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0066 50/133 [==========>...................] - ETA: 0s - loss: 0.0628 98/133 [=====================>........] - ETA: 0s - loss: 0.0571 133/133 [==============================] - 0s 1ms/step - loss: 0.0565 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0107 50/133 [==========>...................] - ETA: 0s - loss: 0.0492 98/133 [=====================>........] - ETA: 0s - loss: 0.0482 133/133 [==============================] - 0s 1ms/step - loss: 0.0531 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0090 44/133 [========>.....................] - ETA: 0s - loss: 0.0523 89/133 [===================>..........] - ETA: 0s - loss: 0.0516 133/133 [==============================] - 0s 1ms/step - loss: 0.0495 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.1847 50/133 [==========>...................] - ETA: 0s - loss: 0.0480 99/133 [=====================>........] - ETA: 0s - loss: 0.0499 133/133 [==============================] - 0s 1ms/step - loss: 0.0500 -> test with GAN.predict GAN tn, fp: 321, 11 GAN fn, tp: 4, 10 GAN f1 score: 0.571 GAN cohens kappa score: 0.550 -> test with 'LR' LR tn, fp: 187, 145 LR fn, tp: 5, 9 LR f1 score: 0.107 LR cohens kappa score: 0.036 LR average precision score: 0.070 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 5, 9 RF f1 score: 0.783 RF cohens kappa score: 0.775 -> test with 'GB' GB tn, fp: 330, 2 GB fn, tp: 1, 13 GB f1 score: 0.897 GB cohens kappa score: 0.892 -> test with 'KNN' KNN tn, fp: 327, 5 KNN fn, tp: 0, 14 KNN f1 score: 0.848 KNN cohens kappa score: 0.841 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 18s - loss: 1.5112 48/133 [=========>....................] - ETA: 0s - loss: 0.1730  96/133 [====================>.........] - ETA: 0s - loss: 0.1378 133/133 [==============================] - 0s 1ms/step - loss: 0.1336 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 5.9895e-05 49/133 [==========>...................] - ETA: 0s - loss: 0.0639  98/133 [=====================>........] - ETA: 0s - loss: 0.0782 133/133 [==============================] - 0s 1ms/step - loss: 0.0695 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0279 49/133 [==========>...................] - ETA: 0s - loss: 0.0555 97/133 [====================>.........] - ETA: 0s - loss: 0.0547 133/133 [==============================] - 0s 1ms/step - loss: 0.0584 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 5.5007e-04 49/133 [==========>...................] - ETA: 0s - loss: 0.0489  97/133 [====================>.........] - ETA: 0s - loss: 0.0552 133/133 [==============================] - 0s 1ms/step - loss: 0.0519 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0111 44/133 [========>.....................] - ETA: 0s - loss: 0.0388 77/133 [================>.............] - ETA: 0s - loss: 0.0425 110/133 [=======================>......] - ETA: 0s - loss: 0.0431 133/133 [==============================] - 0s 1ms/step - loss: 0.0488 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0014 47/133 [=========>....................] - ETA: 0s - loss: 0.0466 93/133 [===================>..........] - ETA: 0s - loss: 0.0461 133/133 [==============================] - 0s 1ms/step - loss: 0.0425 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0067 49/133 [==========>...................] - ETA: 0s - loss: 0.0539 96/133 [====================>.........] - ETA: 0s - loss: 0.0502 133/133 [==============================] - 0s 1ms/step - loss: 0.0404 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0100 49/133 [==========>...................] - ETA: 0s - loss: 0.0461 97/133 [====================>.........] - ETA: 0s - loss: 0.0380 133/133 [==============================] - 0s 1ms/step - loss: 0.0397 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.2082 49/133 [==========>...................] - ETA: 0s - loss: 0.0370 93/133 [===================>..........] - ETA: 0s - loss: 0.0401 132/133 [============================>.] - ETA: 0s - loss: 0.0374 133/133 [==============================] - 0s 1ms/step - loss: 0.0372 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0046 48/133 [=========>....................] - ETA: 0s - loss: 0.0354 96/133 [====================>.........] - ETA: 0s - loss: 0.0387 133/133 [==============================] - 0s 1ms/step - loss: 0.0332 -> test with GAN.predict GAN tn, fp: 324, 8 GAN fn, tp: 7, 7 GAN f1 score: 0.483 GAN cohens kappa score: 0.460 -> test with 'LR' LR tn, fp: 194, 138 LR fn, tp: 3, 11 LR f1 score: 0.135 LR cohens kappa score: 0.066 LR average precision score: 0.088 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 6, 8 RF f1 score: 0.727 RF cohens kappa score: 0.719 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 3, 11 GB f1 score: 0.846 GB cohens kappa score: 0.840 -> test with 'KNN' KNN tn, fp: 319, 13 KNN fn, tp: 1, 13 KNN f1 score: 0.650 KNN cohens kappa score: 0.631 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 22s - loss: 0.7339 46/133 [=========>....................] - ETA: 0s - loss: 0.1968  91/133 [===================>..........] - ETA: 0s - loss: 0.1627 133/133 [==============================] - 0s 1ms/step - loss: 0.1338 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0048 44/133 [========>.....................] - ETA: 0s - loss: 0.0980 86/133 [==================>...........] - ETA: 0s - loss: 0.0874 128/133 [===========================>..] - ETA: 0s - loss: 0.0795 133/133 [==============================] - 0s 1ms/step - loss: 0.0802 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0639 44/133 [========>.....................] - ETA: 0s - loss: 0.0513 87/133 [==================>...........] - ETA: 0s - loss: 0.0613 128/133 [===========================>..] - ETA: 0s - loss: 0.0631 133/133 [==============================] - 0s 1ms/step - loss: 0.0631 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 8.4800e-04 46/133 [=========>....................] - ETA: 0s - loss: 0.0596  90/133 [===================>..........] - ETA: 0s - loss: 0.0491 133/133 [==============================] - ETA: 0s - loss: 0.0540 133/133 [==============================] - 0s 1ms/step - loss: 0.0540 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0015 45/133 [=========>....................] - ETA: 0s - loss: 0.0507 87/133 [==================>...........] - ETA: 0s - loss: 0.0530 125/133 [===========================>..] - ETA: 0s - loss: 0.0501 133/133 [==============================] - 0s 1ms/step - loss: 0.0512 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 8.9486e-04 38/133 [=======>......................] - ETA: 0s - loss: 0.0603  80/133 [=================>............] - ETA: 0s - loss: 0.0501 123/133 [==========================>...] - ETA: 0s - loss: 0.0497 133/133 [==============================] - 0s 1ms/step - loss: 0.0504 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.2683 45/133 [=========>....................] - ETA: 0s - loss: 0.0426 89/133 [===================>..........] - ETA: 0s - loss: 0.0510 133/133 [==============================] - 0s 1ms/step - loss: 0.0449 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.1522 43/133 [========>.....................] - ETA: 0s - loss: 0.0429 86/133 [==================>...........] - ETA: 0s - loss: 0.0430 129/133 [============================>.] - ETA: 0s - loss: 0.0421 133/133 [==============================] - 0s 1ms/step - loss: 0.0441 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.1315 45/133 [=========>....................] - ETA: 0s - loss: 0.0483 88/133 [==================>...........] - ETA: 0s - loss: 0.0357 131/133 [============================>.] - ETA: 0s - loss: 0.0394 133/133 [==============================] - 0s 1ms/step - loss: 0.0415 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0207 44/133 [========>.....................] - ETA: 0s - loss: 0.0454 83/133 [=================>............] - ETA: 0s - loss: 0.0400 126/133 [===========================>..] - ETA: 0s - loss: 0.0401 133/133 [==============================] - 0s 1ms/step - loss: 0.0403 -> test with GAN.predict GAN tn, fp: 328, 4 GAN fn, tp: 5, 9 GAN f1 score: 0.667 GAN cohens kappa score: 0.653 -> test with 'LR' LR tn, fp: 190, 142 LR fn, tp: 5, 9 LR f1 score: 0.109 LR cohens kappa score: 0.038 LR average precision score: 0.065 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 6, 8 RF f1 score: 0.727 RF cohens kappa score: 0.719 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 3, 11 GB f1 score: 0.846 GB cohens kappa score: 0.840 -> test with 'KNN' KNN tn, fp: 319, 13 KNN fn, tp: 3, 11 KNN f1 score: 0.579 KNN cohens kappa score: 0.556 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 19s - loss: 0.9594 49/133 [==========>...................] - ETA: 0s - loss: 0.2316  98/133 [=====================>........] - ETA: 0s - loss: 0.1714 133/133 [==============================] - 0s 1ms/step - loss: 0.1453 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0116 50/133 [==========>...................] - ETA: 0s - loss: 0.0724 97/133 [====================>.........] - ETA: 0s - loss: 0.0724 133/133 [==============================] - 0s 1ms/step - loss: 0.0777 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 7.3064e-04 49/133 [==========>...................] - ETA: 0s - loss: 0.0456  97/133 [====================>.........] - ETA: 0s - loss: 0.0496 133/133 [==============================] - 0s 1ms/step - loss: 0.0559 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0565 50/133 [==========>...................] - ETA: 0s - loss: 0.0616 99/133 [=====================>........] - ETA: 0s - loss: 0.0539 133/133 [==============================] - 0s 1ms/step - loss: 0.0496 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 3.7389e-05 48/133 [=========>....................] - ETA: 0s - loss: 0.0322  97/133 [====================>.........] - ETA: 0s - loss: 0.0418 133/133 [==============================] - 0s 1ms/step - loss: 0.0413 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 6.8958e-05 45/133 [=========>....................] - ETA: 0s - loss: 0.0283  88/133 [==================>...........] - ETA: 0s - loss: 0.0382 131/133 [============================>.] - ETA: 0s - loss: 0.0382 133/133 [==============================] - 0s 1ms/step - loss: 0.0380 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0114 50/133 [==========>...................] - ETA: 0s - loss: 0.0306 99/133 [=====================>........] - ETA: 0s - loss: 0.0283 133/133 [==============================] - 0s 1ms/step - loss: 0.0305 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0580 49/133 [==========>...................] - ETA: 0s - loss: 0.0370 97/133 [====================>.........] - ETA: 0s - loss: 0.0373 133/133 [==============================] - 0s 1ms/step - loss: 0.0332 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 9.6436e-04 49/133 [==========>...................] - ETA: 0s - loss: 0.0211  97/133 [====================>.........] - ETA: 0s - loss: 0.0262 133/133 [==============================] - 0s 1ms/step - loss: 0.0312 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0018 49/133 [==========>...................] - ETA: 0s - loss: 0.0319 98/133 [=====================>........] - ETA: 0s - loss: 0.0323 133/133 [==============================] - 0s 1ms/step - loss: 0.0281 -> test with GAN.predict GAN tn, fp: 324, 8 GAN fn, tp: 4, 10 GAN f1 score: 0.625 GAN cohens kappa score: 0.607 -> test with 'LR' LR tn, fp: 202, 130 LR fn, tp: 5, 9 LR f1 score: 0.118 LR cohens kappa score: 0.048 LR average precision score: 0.078 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 9, 5 RF f1 score: 0.526 RF cohens kappa score: 0.516 -> test with 'GB' GB tn, fp: 332, 0 GB fn, tp: 3, 11 GB f1 score: 0.880 GB cohens kappa score: 0.876 -> test with 'KNN' KNN tn, fp: 315, 17 KNN fn, tp: 2, 12 KNN f1 score: 0.558 KNN cohens kappa score: 0.533 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 18s - loss: 2.5542e-07 50/133 [==========>...................] - ETA: 0s - loss: 0.1740  99/133 [=====================>........] - ETA: 0s - loss: 0.1276 133/133 [==============================] - 0s 1ms/step - loss: 0.1222 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 9.6421e-05 50/133 [==========>...................] - ETA: 0s - loss: 0.0617  98/133 [=====================>........] - ETA: 0s - loss: 0.0682 133/133 [==============================] - 0s 1ms/step - loss: 0.0726 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0618 50/133 [==========>...................] - ETA: 0s - loss: 0.0614 99/133 [=====================>........] - ETA: 0s - loss: 0.0657 133/133 [==============================] - 0s 1ms/step - loss: 0.0632 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0102 50/133 [==========>...................] - ETA: 0s - loss: 0.0493 99/133 [=====================>........] - ETA: 0s - loss: 0.0526 133/133 [==============================] - 0s 1ms/step - loss: 0.0527 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0043 48/133 [=========>....................] - ETA: 0s - loss: 0.0490 97/133 [====================>.........] - ETA: 0s - loss: 0.0418 133/133 [==============================] - 0s 1ms/step - loss: 0.0496 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0136 50/133 [==========>...................] - ETA: 0s - loss: 0.0274 99/133 [=====================>........] - ETA: 0s - loss: 0.0453 133/133 [==============================] - 0s 1ms/step - loss: 0.0475 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0979 48/133 [=========>....................] - ETA: 0s - loss: 0.0454 97/133 [====================>.........] - ETA: 0s - loss: 0.0465 133/133 [==============================] - 0s 1ms/step - loss: 0.0466 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0127 50/133 [==========>...................] - ETA: 0s - loss: 0.0441 94/133 [====================>.........] - ETA: 0s - loss: 0.0436 133/133 [==============================] - 0s 1ms/step - loss: 0.0416 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0083 45/133 [=========>....................] - ETA: 0s - loss: 0.0361 93/133 [===================>..........] - ETA: 0s - loss: 0.0383 133/133 [==============================] - 0s 1ms/step - loss: 0.0399 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0054 49/133 [==========>...................] - ETA: 0s - loss: 0.0414 98/133 [=====================>........] - ETA: 0s - loss: 0.0378 133/133 [==============================] - 0s 1ms/step - loss: 0.0389 -> test with GAN.predict GAN tn, fp: 323, 8 GAN fn, tp: 7, 6 GAN f1 score: 0.444 GAN cohens kappa score: 0.422 -> test with 'LR' LR tn, fp: 187, 144 LR fn, tp: 5, 8 LR f1 score: 0.097 LR cohens kappa score: 0.029 LR average precision score: 0.053 -> test with 'RF' RF tn, fp: 331, 0 RF fn, tp: 9, 4 RF f1 score: 0.471 RF cohens kappa score: 0.461 -> test with 'GB' GB tn, fp: 329, 2 GB fn, tp: 2, 11 GB f1 score: 0.846 GB cohens kappa score: 0.840 -> test with 'KNN' KNN tn, fp: 321, 10 KNN fn, tp: 3, 10 KNN f1 score: 0.606 KNN cohens kappa score: 0.587 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 23s - loss: 3.4202e-08 50/133 [==========>...................] - ETA: 0s - loss: 0.4144  99/133 [=====================>........] - ETA: 0s - loss: 0.3423 133/133 [==============================] - 0s 1ms/step - loss: 0.3054 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 5.4349e-04 49/133 [==========>...................] - ETA: 0s - loss: 0.1734  97/133 [====================>.........] - ETA: 0s - loss: 0.1657 133/133 [==============================] - 0s 1ms/step - loss: 0.1613 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0044 49/133 [==========>...................] - ETA: 0s - loss: 0.1437 97/133 [====================>.........] - ETA: 0s - loss: 0.1202 133/133 [==============================] - 0s 1ms/step - loss: 0.1088 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.1586 49/133 [==========>...................] - ETA: 0s - loss: 0.1245 98/133 [=====================>........] - ETA: 0s - loss: 0.0922 133/133 [==============================] - 0s 1ms/step - loss: 0.0887 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0128 47/133 [=========>....................] - ETA: 0s - loss: 0.0778 95/133 [====================>.........] - ETA: 0s - loss: 0.0862 133/133 [==============================] - 0s 1ms/step - loss: 0.0808 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0852 50/133 [==========>...................] - ETA: 0s - loss: 0.0656 98/133 [=====================>........] - ETA: 0s - loss: 0.0627 133/133 [==============================] - 0s 1ms/step - loss: 0.0762 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0782 50/133 [==========>...................] - ETA: 0s - loss: 0.0640 95/133 [====================>.........] - ETA: 0s - loss: 0.0703 133/133 [==============================] - 0s 1ms/step - loss: 0.0724 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0050 48/133 [=========>....................] - ETA: 0s - loss: 0.0742 96/133 [====================>.........] - ETA: 0s - loss: 0.0716 133/133 [==============================] - 0s 1ms/step - loss: 0.0703 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0226 48/133 [=========>....................] - ETA: 0s - loss: 0.0702 96/133 [====================>.........] - ETA: 0s - loss: 0.0733 133/133 [==============================] - 0s 1ms/step - loss: 0.0681 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.1735 49/133 [==========>...................] - ETA: 0s - loss: 0.0537 97/133 [====================>.........] - ETA: 0s - loss: 0.0634 133/133 [==============================] - 0s 1ms/step - loss: 0.0657 -> test with GAN.predict GAN tn, fp: 327, 5 GAN fn, tp: 10, 4 GAN f1 score: 0.348 GAN cohens kappa score: 0.326 -> test with 'LR' LR tn, fp: 173, 159 LR fn, tp: 4, 10 LR f1 score: 0.109 LR cohens kappa score: 0.037 LR average precision score: 0.078 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 9, 5 RF f1 score: 0.526 RF cohens kappa score: 0.516 -> test with 'GB' GB tn, fp: 332, 0 GB fn, tp: 1, 13 GB f1 score: 0.963 GB cohens kappa score: 0.961 -> test with 'KNN' KNN tn, fp: 329, 3 KNN fn, tp: 4, 10 KNN f1 score: 0.741 KNN cohens kappa score: 0.730 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 23s - loss: 0.6038 45/133 [=========>....................] - ETA: 0s - loss: 0.2671  89/133 [===================>..........] - ETA: 0s - loss: 0.2144 132/133 [============================>.] - ETA: 0s - loss: 0.1797 133/133 [==============================] - 0s 1ms/step - loss: 0.1790 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.1222 43/133 [========>.....................] - ETA: 0s - loss: 0.0773 83/133 [=================>............] - ETA: 0s - loss: 0.0735 125/133 [===========================>..] - ETA: 0s - loss: 0.0857 133/133 [==============================] - 0s 1ms/step - loss: 0.0885 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 2.6313e-05 43/133 [========>.....................] - ETA: 0s - loss: 0.0501  85/133 [==================>...........] - ETA: 0s - loss: 0.0677 120/133 [==========================>...] - ETA: 0s - loss: 0.0654 133/133 [==============================] - 0s 1ms/step - loss: 0.0660 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0780 36/133 [=======>......................] - ETA: 0s - loss: 0.0460 74/133 [===============>..............] - ETA: 0s - loss: 0.0616 118/133 [=========================>....] - ETA: 0s - loss: 0.0551 133/133 [==============================] - 0s 1ms/step - loss: 0.0554 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0281 45/133 [=========>....................] - ETA: 0s - loss: 0.0412 88/133 [==================>...........] - ETA: 0s - loss: 0.0511 133/133 [==============================] - ETA: 0s - loss: 0.0509 133/133 [==============================] - 0s 1ms/step - loss: 0.0509 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0129 45/133 [=========>....................] - ETA: 0s - loss: 0.0482 90/133 [===================>..........] - ETA: 0s - loss: 0.0459 133/133 [==============================] - ETA: 0s - loss: 0.0467 133/133 [==============================] - 0s 1ms/step - loss: 0.0467 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0144 42/133 [========>.....................] - ETA: 0s - loss: 0.0357 85/133 [==================>...........] - ETA: 0s - loss: 0.0353 129/133 [============================>.] - ETA: 0s - loss: 0.0411 133/133 [==============================] - 0s 1ms/step - loss: 0.0408 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0558 42/133 [========>.....................] - ETA: 0s - loss: 0.0349 86/133 [==================>...........] - ETA: 0s - loss: 0.0337 128/133 [===========================>..] - ETA: 0s - loss: 0.0400 133/133 [==============================] - 0s 1ms/step - loss: 0.0395 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0048 45/133 [=========>....................] - ETA: 0s - loss: 0.0265 84/133 [=================>............] - ETA: 0s - loss: 0.0291 126/133 [===========================>..] - ETA: 0s - loss: 0.0331 133/133 [==============================] - 0s 1ms/step - loss: 0.0376 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0147 44/133 [========>.....................] - ETA: 0s - loss: 0.0346 86/133 [==================>...........] - ETA: 0s - loss: 0.0410 131/133 [============================>.] - ETA: 0s - loss: 0.0372 133/133 [==============================] - 0s 1ms/step - loss: 0.0369 -> test with GAN.predict GAN tn, fp: 321, 11 GAN fn, tp: 1, 13 GAN f1 score: 0.684 GAN cohens kappa score: 0.667 -> test with 'LR' LR tn, fp: 178, 154 LR fn, tp: 3, 11 LR f1 score: 0.123 LR cohens kappa score: 0.052 LR average precision score: 0.080 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 8, 6 RF f1 score: 0.600 RF cohens kappa score: 0.590 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 1, 13 GB f1 score: 0.929 GB cohens kappa score: 0.926 -> test with 'KNN' KNN tn, fp: 317, 15 KNN fn, tp: 3, 11 KNN f1 score: 0.550 KNN cohens kappa score: 0.525 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 22s - loss: 0.0563 49/133 [==========>...................] - ETA: 0s - loss: 0.2455  98/133 [=====================>........] - ETA: 0s - loss: 0.1936 133/133 [==============================] - 0s 1ms/step - loss: 0.1675 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.1368 49/133 [==========>...................] - ETA: 0s - loss: 0.0930 97/133 [====================>.........] - ETA: 0s - loss: 0.0976 133/133 [==============================] - 0s 1ms/step - loss: 0.0915 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.1159 49/133 [==========>...................] - ETA: 0s - loss: 0.0692 92/133 [===================>..........] - ETA: 0s - loss: 0.0778 133/133 [==============================] - 0s 1ms/step - loss: 0.0699 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0167 49/133 [==========>...................] - ETA: 0s - loss: 0.0581 94/133 [====================>.........] - ETA: 0s - loss: 0.0637 133/133 [==============================] - 0s 1ms/step - loss: 0.0600 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 5.3481e-04 50/133 [==========>...................] - ETA: 0s - loss: 0.0543  98/133 [=====================>........] - ETA: 0s - loss: 0.0629 133/133 [==============================] - 0s 1ms/step - loss: 0.0555 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0306 49/133 [==========>...................] - ETA: 0s - loss: 0.0491 96/133 [====================>.........] - ETA: 0s - loss: 0.0510 133/133 [==============================] - 0s 1ms/step - loss: 0.0534 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0276 50/133 [==========>...................] - ETA: 0s - loss: 0.0493 96/133 [====================>.........] - ETA: 0s - loss: 0.0495 133/133 [==============================] - 0s 1ms/step - loss: 0.0488 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0155 50/133 [==========>...................] - ETA: 0s - loss: 0.0450 99/133 [=====================>........] - ETA: 0s - loss: 0.0541 133/133 [==============================] - 0s 1ms/step - loss: 0.0473 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0110 49/133 [==========>...................] - ETA: 0s - loss: 0.0490 97/133 [====================>.........] - ETA: 0s - loss: 0.0467 133/133 [==============================] - 0s 1ms/step - loss: 0.0457 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0026 49/133 [==========>...................] - ETA: 0s - loss: 0.0439 98/133 [=====================>........] - ETA: 0s - loss: 0.0403 133/133 [==============================] - 0s 1ms/step - loss: 0.0419 -> test with GAN.predict GAN tn, fp: 322, 10 GAN fn, tp: 7, 7 GAN f1 score: 0.452 GAN cohens kappa score: 0.426 -> test with 'LR' LR tn, fp: 197, 135 LR fn, tp: 5, 9 LR f1 score: 0.114 LR cohens kappa score: 0.043 LR average precision score: 0.074 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 7, 7 RF f1 score: 0.667 RF cohens kappa score: 0.657 -> test with 'GB' GB tn, fp: 332, 0 GB fn, tp: 4, 10 GB f1 score: 0.833 GB cohens kappa score: 0.828 -> test with 'KNN' KNN tn, fp: 319, 13 KNN fn, tp: 4, 10 KNN f1 score: 0.541 KNN cohens kappa score: 0.516 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 23s - loss: 0.0104 49/133 [==========>...................] - ETA: 0s - loss: 0.1889  98/133 [=====================>........] - ETA: 0s - loss: 0.1359 133/133 [==============================] - 0s 1ms/step - loss: 0.1349 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0063 47/133 [=========>....................] - ETA: 0s - loss: 0.0789 96/133 [====================>.........] - ETA: 0s - loss: 0.0812 133/133 [==============================] - 0s 1ms/step - loss: 0.0813 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 6.2214e-04 50/133 [==========>...................] - ETA: 0s - loss: 0.0643  99/133 [=====================>........] - ETA: 0s - loss: 0.0585 133/133 [==============================] - 0s 1ms/step - loss: 0.0610 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 2.5847e-04 50/133 [==========>...................] - ETA: 0s - loss: 0.0683  98/133 [=====================>........] - ETA: 0s - loss: 0.0478 133/133 [==============================] - 0s 1ms/step - loss: 0.0511 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0090 48/133 [=========>....................] - ETA: 0s - loss: 0.0276 97/133 [====================>.........] - ETA: 0s - loss: 0.0374 133/133 [==============================] - 0s 1ms/step - loss: 0.0438 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0908 50/133 [==========>...................] - ETA: 0s - loss: 0.0302 99/133 [=====================>........] - ETA: 0s - loss: 0.0317 133/133 [==============================] - 0s 1ms/step - loss: 0.0406 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.1748 50/133 [==========>...................] - ETA: 0s - loss: 0.0449 98/133 [=====================>........] - ETA: 0s - loss: 0.0383 133/133 [==============================] - 0s 1ms/step - loss: 0.0416 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0395 49/133 [==========>...................] - ETA: 0s - loss: 0.0467 97/133 [====================>.........] - ETA: 0s - loss: 0.0389 133/133 [==============================] - 0s 1ms/step - loss: 0.0393 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0082 50/133 [==========>...................] - ETA: 0s - loss: 0.0332 98/133 [=====================>........] - ETA: 0s - loss: 0.0345 133/133 [==============================] - 0s 1ms/step - loss: 0.0342 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0403 50/133 [==========>...................] - ETA: 0s - loss: 0.0385 98/133 [=====================>........] - ETA: 0s - loss: 0.0319 133/133 [==============================] - 0s 1ms/step - loss: 0.0334 -> test with GAN.predict GAN tn, fp: 320, 12 GAN fn, tp: 6, 8 GAN f1 score: 0.471 GAN cohens kappa score: 0.444 -> test with 'LR' LR tn, fp: 202, 130 LR fn, tp: 7, 7 LR f1 score: 0.093 LR cohens kappa score: 0.021 LR average precision score: 0.050 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 11, 3 RF f1 score: 0.353 RF cohens kappa score: 0.344 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 3, 11 GB f1 score: 0.846 GB cohens kappa score: 0.840 -> test with 'KNN' KNN tn, fp: 312, 20 KNN fn, tp: 3, 11 KNN f1 score: 0.489 KNN cohens kappa score: 0.459 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 18s - loss: 0.0366 48/133 [=========>....................] - ETA: 0s - loss: 0.2778  96/133 [====================>.........] - ETA: 0s - loss: 0.2484 133/133 [==============================] - 0s 1ms/step - loss: 0.2121 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.1049 49/133 [==========>...................] - ETA: 0s - loss: 0.0908 98/133 [=====================>........] - ETA: 0s - loss: 0.1017 133/133 [==============================] - 0s 1ms/step - loss: 0.0985 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0057 48/133 [=========>....................] - ETA: 0s - loss: 0.0641 94/133 [====================>.........] - ETA: 0s - loss: 0.0701 133/133 [==============================] - 0s 1ms/step - loss: 0.0754 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.1557 50/133 [==========>...................] - ETA: 0s - loss: 0.0556 98/133 [=====================>........] - ETA: 0s - loss: 0.0606 133/133 [==============================] - 0s 1ms/step - loss: 0.0620 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0014 49/133 [==========>...................] - ETA: 0s - loss: 0.0673 92/133 [===================>..........] - ETA: 0s - loss: 0.0572 133/133 [==============================] - 0s 1ms/step - loss: 0.0557 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0303 46/133 [=========>....................] - ETA: 0s - loss: 0.0610 95/133 [====================>.........] - ETA: 0s - loss: 0.0610 133/133 [==============================] - 0s 1ms/step - loss: 0.0529 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0334 50/133 [==========>...................] - ETA: 0s - loss: 0.0581 98/133 [=====================>........] - ETA: 0s - loss: 0.0529 133/133 [==============================] - 0s 1ms/step - loss: 0.0480 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.1067 49/133 [==========>...................] - ETA: 0s - loss: 0.0548 98/133 [=====================>........] - ETA: 0s - loss: 0.0457 133/133 [==============================] - 0s 1ms/step - loss: 0.0457 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0614 49/133 [==========>...................] - ETA: 0s - loss: 0.0548 97/133 [====================>.........] - ETA: 0s - loss: 0.0466 133/133 [==============================] - 0s 1ms/step - loss: 0.0439 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0558 49/133 [==========>...................] - ETA: 0s - loss: 0.0362 97/133 [====================>.........] - ETA: 0s - loss: 0.0329 133/133 [==============================] - 0s 1ms/step - loss: 0.0407 -> test with GAN.predict GAN tn, fp: 326, 5 GAN fn, tp: 7, 6 GAN f1 score: 0.500 GAN cohens kappa score: 0.482 -> test with 'LR' LR tn, fp: 193, 138 LR fn, tp: 6, 7 LR f1 score: 0.089 LR cohens kappa score: 0.021 LR average precision score: 0.078 -> test with 'RF' RF tn, fp: 331, 0 RF fn, tp: 9, 4 RF f1 score: 0.471 RF cohens kappa score: 0.461 -> test with 'GB' GB tn, fp: 328, 3 GB fn, tp: 2, 11 GB f1 score: 0.815 GB cohens kappa score: 0.807 -> test with 'KNN' KNN tn, fp: 315, 16 KNN fn, tp: 0, 13 KNN f1 score: 0.619 KNN cohens kappa score: 0.598 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 20s - loss: 0.0015 46/133 [=========>....................] - ETA: 0s - loss: 0.1630  88/133 [==================>...........] - ETA: 0s - loss: 0.1534 132/133 [============================>.] - ETA: 0s - loss: 0.1403 133/133 [==============================] - 0s 1ms/step - loss: 0.1396 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0045 43/133 [========>.....................] - ETA: 0s - loss: 0.0835 87/133 [==================>...........] - ETA: 0s - loss: 0.0735 130/133 [============================>.] - ETA: 0s - loss: 0.0666 133/133 [==============================] - 0s 1ms/step - loss: 0.0654 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0184 45/133 [=========>....................] - ETA: 0s - loss: 0.0539 89/133 [===================>..........] - ETA: 0s - loss: 0.0555 133/133 [==============================] - ETA: 0s - loss: 0.0515 133/133 [==============================] - 0s 1ms/step - loss: 0.0515 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.5939 44/133 [========>.....................] - ETA: 0s - loss: 0.0459 88/133 [==================>...........] - ETA: 0s - loss: 0.0446 128/133 [===========================>..] - ETA: 0s - loss: 0.0459 133/133 [==============================] - 0s 1ms/step - loss: 0.0451 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0340 45/133 [=========>....................] - ETA: 0s - loss: 0.0465 87/133 [==================>...........] - ETA: 0s - loss: 0.0414 129/133 [============================>.] - ETA: 0s - loss: 0.0372 133/133 [==============================] - 0s 1ms/step - loss: 0.0398 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0222 44/133 [========>.....................] - ETA: 0s - loss: 0.0700 88/133 [==================>...........] - ETA: 0s - loss: 0.0446 131/133 [============================>.] - ETA: 0s - loss: 0.0358 133/133 [==============================] - 0s 1ms/step - loss: 0.0354 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 9.9051e-04 44/133 [========>.....................] - ETA: 0s - loss: 0.0205  79/133 [================>.............] - ETA: 0s - loss: 0.0279 116/133 [=========================>....] - ETA: 0s - loss: 0.0325 133/133 [==============================] - 0s 1ms/step - loss: 0.0357 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.1150 43/133 [========>.....................] - ETA: 0s - loss: 0.0461 85/133 [==================>...........] - ETA: 0s - loss: 0.0432 128/133 [===========================>..] - ETA: 0s - loss: 0.0356 133/133 [==============================] - 0s 1ms/step - loss: 0.0347 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0021 39/133 [=======>......................] - ETA: 0s - loss: 0.0404 82/133 [=================>............] - ETA: 0s - loss: 0.0405 125/133 [===========================>..] - ETA: 0s - loss: 0.0316 133/133 [==============================] - 0s 1ms/step - loss: 0.0302 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 8.2462e-04 44/133 [========>.....................] - ETA: 0s - loss: 0.0267  89/133 [===================>..........] - ETA: 0s - loss: 0.0341 132/133 [============================>.] - ETA: 0s - loss: 0.0281 133/133 [==============================] - 0s 1ms/step - loss: 0.0280 -> test with GAN.predict GAN tn, fp: 321, 11 GAN fn, tp: 4, 10 GAN f1 score: 0.571 GAN cohens kappa score: 0.550 -> test with 'LR' LR tn, fp: 175, 157 LR fn, tp: 2, 12 LR f1 score: 0.131 LR cohens kappa score: 0.061 LR average precision score: 0.080 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 5, 9 RF f1 score: 0.783 RF cohens kappa score: 0.775 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 3, 11 GB f1 score: 0.846 GB cohens kappa score: 0.840 -> test with 'KNN' KNN tn, fp: 311, 21 KNN fn, tp: 0, 14 KNN f1 score: 0.571 KNN cohens kappa score: 0.545 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 22s - loss: 3.9398e-05 49/133 [==========>...................] - ETA: 0s - loss: 0.1789  96/133 [====================>.........] - ETA: 0s - loss: 0.1649 133/133 [==============================] - 0s 1ms/step - loss: 0.1565 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.2061 47/133 [=========>....................] - ETA: 0s - loss: 0.0817 95/133 [====================>.........] - ETA: 0s - loss: 0.1113 133/133 [==============================] - 0s 1ms/step - loss: 0.0992 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0861 49/133 [==========>...................] - ETA: 0s - loss: 0.0810 96/133 [====================>.........] - ETA: 0s - loss: 0.0665 133/133 [==============================] - 0s 1ms/step - loss: 0.0683 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0074 49/133 [==========>...................] - ETA: 0s - loss: 0.0544 98/133 [=====================>........] - ETA: 0s - loss: 0.0574 133/133 [==============================] - 0s 1ms/step - loss: 0.0636 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0119 49/133 [==========>...................] - ETA: 0s - loss: 0.0574 92/133 [===================>..........] - ETA: 0s - loss: 0.0513 133/133 [==============================] - 0s 1ms/step - loss: 0.0566 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.3540 44/133 [========>.....................] - ETA: 0s - loss: 0.0633 92/133 [===================>..........] - ETA: 0s - loss: 0.0592 133/133 [==============================] - 0s 1ms/step - loss: 0.0522 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0255 49/133 [==========>...................] - ETA: 0s - loss: 0.0545 97/133 [====================>.........] - ETA: 0s - loss: 0.0500 133/133 [==============================] - 0s 1ms/step - loss: 0.0506 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0427 50/133 [==========>...................] - ETA: 0s - loss: 0.0589 98/133 [=====================>........] - ETA: 0s - loss: 0.0468 133/133 [==============================] - 0s 1ms/step - loss: 0.0467 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0095 46/133 [=========>....................] - ETA: 0s - loss: 0.0600 95/133 [====================>.........] - ETA: 0s - loss: 0.0476 133/133 [==============================] - 0s 1ms/step - loss: 0.0459 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0098 50/133 [==========>...................] - ETA: 0s - loss: 0.0347 97/133 [====================>.........] - ETA: 0s - loss: 0.0394 133/133 [==============================] - 0s 1ms/step - loss: 0.0420 -> test with GAN.predict GAN tn, fp: 324, 8 GAN fn, tp: 3, 11 GAN f1 score: 0.667 GAN cohens kappa score: 0.650 -> test with 'LR' LR tn, fp: 200, 132 LR fn, tp: 5, 9 LR f1 score: 0.116 LR cohens kappa score: 0.046 LR average precision score: 0.072 -> test with 'RF' RF tn, fp: 331, 1 RF fn, tp: 7, 7 RF f1 score: 0.636 RF cohens kappa score: 0.625 -> test with 'GB' GB tn, fp: 330, 2 GB fn, tp: 1, 13 GB f1 score: 0.897 GB cohens kappa score: 0.892 -> test with 'KNN' KNN tn, fp: 314, 18 KNN fn, tp: 2, 12 KNN f1 score: 0.545 KNN cohens kappa score: 0.519 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 18s - loss: 1.2133e-04 50/133 [==========>...................] - ETA: 0s - loss: 0.1948  99/133 [=====================>........] - ETA: 0s - loss: 0.1483 133/133 [==============================] - 0s 1ms/step - loss: 0.1243 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 1.1816e-04 48/133 [=========>....................] - ETA: 0s - loss: 0.0810  97/133 [====================>.........] - ETA: 0s - loss: 0.0756 133/133 [==============================] - 0s 1ms/step - loss: 0.0638 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0249 46/133 [=========>....................] - ETA: 0s - loss: 0.0490 94/133 [====================>.........] - ETA: 0s - loss: 0.0453 133/133 [==============================] - 0s 1ms/step - loss: 0.0497 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.1349 50/133 [==========>...................] - ETA: 0s - loss: 0.0377 99/133 [=====================>........] - ETA: 0s - loss: 0.0371 133/133 [==============================] - 0s 1ms/step - loss: 0.0427 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0986 50/133 [==========>...................] - ETA: 0s - loss: 0.0391 99/133 [=====================>........] - ETA: 0s - loss: 0.0441 133/133 [==============================] - 0s 1ms/step - loss: 0.0399 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0151 50/133 [==========>...................] - ETA: 0s - loss: 0.0324 99/133 [=====================>........] - ETA: 0s - loss: 0.0327 133/133 [==============================] - 0s 1ms/step - loss: 0.0356 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0053 50/133 [==========>...................] - ETA: 0s - loss: 0.0264 98/133 [=====================>........] - ETA: 0s - loss: 0.0305 133/133 [==============================] - 0s 1ms/step - loss: 0.0313 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0069 50/133 [==========>...................] - ETA: 0s - loss: 0.0322 99/133 [=====================>........] - ETA: 0s - loss: 0.0319 133/133 [==============================] - 0s 1ms/step - loss: 0.0330 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0289 47/133 [=========>....................] - ETA: 0s - loss: 0.0314 90/133 [===================>..........] - ETA: 0s - loss: 0.0320 133/133 [==============================] - ETA: 0s - loss: 0.0286 133/133 [==============================] - 0s 1ms/step - loss: 0.0286 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0018 47/133 [=========>....................] - ETA: 0s - loss: 0.0486 96/133 [====================>.........] - ETA: 0s - loss: 0.0304 133/133 [==============================] - 0s 1ms/step - loss: 0.0299 -> test with GAN.predict GAN tn, fp: 325, 7 GAN fn, tp: 7, 7 GAN f1 score: 0.500 GAN cohens kappa score: 0.479 -> test with 'LR' LR tn, fp: 189, 143 LR fn, tp: 6, 8 LR f1 score: 0.097 LR cohens kappa score: 0.025 LR average precision score: 0.058 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 7, 7 RF f1 score: 0.667 RF cohens kappa score: 0.657 -> test with 'GB' GB tn, fp: 330, 2 GB fn, tp: 7, 7 GB f1 score: 0.609 GB cohens kappa score: 0.596 -> test with 'KNN' KNN tn, fp: 319, 13 KNN fn, tp: 4, 10 KNN f1 score: 0.541 KNN cohens kappa score: 0.516 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 22s - loss: 1.6422e-06 49/133 [==========>...................] - ETA: 0s - loss: 0.1687  97/133 [====================>.........] - ETA: 0s - loss: 0.1337 133/133 [==============================] - 0s 1ms/step - loss: 0.1409 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0113 49/133 [==========>...................] - ETA: 0s - loss: 0.0695 97/133 [====================>.........] - ETA: 0s - loss: 0.0808 133/133 [==============================] - 0s 1ms/step - loss: 0.0728 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0326 49/133 [==========>...................] - ETA: 0s - loss: 0.0955 98/133 [=====================>........] - ETA: 0s - loss: 0.0753 133/133 [==============================] - 0s 1ms/step - loss: 0.0649 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0058 49/133 [==========>...................] - ETA: 0s - loss: 0.0476 97/133 [====================>.........] - ETA: 0s - loss: 0.0510 133/133 [==============================] - 0s 1ms/step - loss: 0.0533 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0298 49/133 [==========>...................] - ETA: 0s - loss: 0.0434 97/133 [====================>.........] - ETA: 0s - loss: 0.0465 133/133 [==============================] - 0s 1ms/step - loss: 0.0486 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0013 50/133 [==========>...................] - ETA: 0s - loss: 0.0579 98/133 [=====================>........] - ETA: 0s - loss: 0.0555 133/133 [==============================] - 0s 1ms/step - loss: 0.0467 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0051 46/133 [=========>....................] - ETA: 0s - loss: 0.0488 94/133 [====================>.........] - ETA: 0s - loss: 0.0455 133/133 [==============================] - 0s 1ms/step - loss: 0.0469 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0844 49/133 [==========>...................] - ETA: 0s - loss: 0.0361 97/133 [====================>.........] - ETA: 0s - loss: 0.0373 133/133 [==============================] - 0s 1ms/step - loss: 0.0400 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0020 49/133 [==========>...................] - ETA: 0s - loss: 0.0385 97/133 [====================>.........] - ETA: 0s - loss: 0.0397 133/133 [==============================] - 0s 1ms/step - loss: 0.0396 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0112 48/133 [=========>....................] - ETA: 0s - loss: 0.0248 96/133 [====================>.........] - ETA: 0s - loss: 0.0350 133/133 [==============================] - 0s 1ms/step - loss: 0.0374 -> test with GAN.predict GAN tn, fp: 323, 9 GAN fn, tp: 4, 10 GAN f1 score: 0.606 GAN cohens kappa score: 0.587 -> test with 'LR' LR tn, fp: 180, 152 LR fn, tp: 3, 11 LR f1 score: 0.124 LR cohens kappa score: 0.054 LR average precision score: 0.082 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 7, 7 RF f1 score: 0.667 RF cohens kappa score: 0.657 -> test with 'GB' GB tn, fp: 332, 0 GB fn, tp: 2, 12 GB f1 score: 0.923 GB cohens kappa score: 0.920 -> test with 'KNN' KNN tn, fp: 315, 17 KNN fn, tp: 2, 12 KNN f1 score: 0.558 KNN cohens kappa score: 0.533 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 19s - loss: 0.0859 49/133 [==========>...................] - ETA: 0s - loss: 0.1414  96/133 [====================>.........] - ETA: 0s - loss: 0.1303 133/133 [==============================] - 0s 1ms/step - loss: 0.1139 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 2.3851e-05 50/133 [==========>...................] - ETA: 0s - loss: 0.1183  98/133 [=====================>........] - ETA: 0s - loss: 0.0802 133/133 [==============================] - 0s 1ms/step - loss: 0.0756 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.1560 50/133 [==========>...................] - ETA: 0s - loss: 0.0623 99/133 [=====================>........] - ETA: 0s - loss: 0.0629 133/133 [==============================] - 0s 1ms/step - loss: 0.0571 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 3.1118e-04 50/133 [==========>...................] - ETA: 0s - loss: 0.0469  99/133 [=====================>........] - ETA: 0s - loss: 0.0434 133/133 [==============================] - 0s 1ms/step - loss: 0.0485 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0444 49/133 [==========>...................] - ETA: 0s - loss: 0.0399 95/133 [====================>.........] - ETA: 0s - loss: 0.0432 133/133 [==============================] - 0s 1ms/step - loss: 0.0442 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0011 48/133 [=========>....................] - ETA: 0s - loss: 0.0370 96/133 [====================>.........] - ETA: 0s - loss: 0.0380 133/133 [==============================] - 0s 1ms/step - loss: 0.0400 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.3417 50/133 [==========>...................] - ETA: 0s - loss: 0.0534 98/133 [=====================>........] - ETA: 0s - loss: 0.0405 133/133 [==============================] - 0s 1ms/step - loss: 0.0355 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0091 50/133 [==========>...................] - ETA: 0s - loss: 0.0241 99/133 [=====================>........] - ETA: 0s - loss: 0.0357 133/133 [==============================] - 0s 1ms/step - loss: 0.0363 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.3245 47/133 [=========>....................] - ETA: 0s - loss: 0.0323 90/133 [===================>..........] - ETA: 0s - loss: 0.0315 133/133 [==============================] - ETA: 0s - loss: 0.0303 133/133 [==============================] - 0s 1ms/step - loss: 0.0303 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0503 49/133 [==========>...................] - ETA: 0s - loss: 0.0285 95/133 [====================>.........] - ETA: 0s - loss: 0.0300 133/133 [==============================] - 0s 1ms/step - loss: 0.0298 -> test with GAN.predict GAN tn, fp: 325, 6 GAN fn, tp: 4, 9 GAN f1 score: 0.643 GAN cohens kappa score: 0.628 -> test with 'LR' LR tn, fp: 184, 147 LR fn, tp: 5, 8 LR f1 score: 0.095 LR cohens kappa score: 0.027 LR average precision score: 0.062 -> test with 'RF' RF tn, fp: 331, 0 RF fn, tp: 5, 8 RF f1 score: 0.762 RF cohens kappa score: 0.755 -> test with 'GB' GB tn, fp: 329, 2 GB fn, tp: 2, 11 GB f1 score: 0.846 GB cohens kappa score: 0.840 -> test with 'KNN' KNN tn, fp: 316, 15 KNN fn, tp: 1, 12 KNN f1 score: 0.600 KNN cohens kappa score: 0.578 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 23s - loss: 0.5007 46/133 [=========>....................] - ETA: 0s - loss: 0.2074  94/133 [====================>.........] - ETA: 0s - loss: 0.1734 133/133 [==============================] - 0s 1ms/step - loss: 0.1460 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.5667 46/133 [=========>....................] - ETA: 0s - loss: 0.0892 92/133 [===================>..........] - ETA: 0s - loss: 0.0867 133/133 [==============================] - 0s 1ms/step - loss: 0.0800 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.2209 47/133 [=========>....................] - ETA: 0s - loss: 0.0591 89/133 [===================>..........] - ETA: 0s - loss: 0.0666 129/133 [============================>.] - ETA: 0s - loss: 0.0632 133/133 [==============================] - 0s 1ms/step - loss: 0.0621 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0197 36/133 [=======>......................] - ETA: 0s - loss: 0.0458 73/133 [===============>..............] - ETA: 0s - loss: 0.0547 116/133 [=========================>....] - ETA: 0s - loss: 0.0595 133/133 [==============================] - 0s 1ms/step - loss: 0.0555 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.2076 46/133 [=========>....................] - ETA: 0s - loss: 0.0407 91/133 [===================>..........] - ETA: 0s - loss: 0.0485 133/133 [==============================] - 0s 1ms/step - loss: 0.0476 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0338 44/133 [========>.....................] - ETA: 0s - loss: 0.0524 87/133 [==================>...........] - ETA: 0s - loss: 0.0503 132/133 [============================>.] - ETA: 0s - loss: 0.0441 133/133 [==============================] - 0s 1ms/step - loss: 0.0446 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0525 43/133 [========>.....................] - ETA: 0s - loss: 0.0416 87/133 [==================>...........] - ETA: 0s - loss: 0.0423 132/133 [============================>.] - ETA: 0s - loss: 0.0392 133/133 [==============================] - 0s 1ms/step - loss: 0.0397 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0029 44/133 [========>.....................] - ETA: 0s - loss: 0.0362 86/133 [==================>...........] - ETA: 0s - loss: 0.0325 127/133 [===========================>..] - ETA: 0s - loss: 0.0376 133/133 [==============================] - 0s 1ms/step - loss: 0.0375 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0035 43/133 [========>.....................] - ETA: 0s - loss: 0.0347 84/133 [=================>............] - ETA: 0s - loss: 0.0410 126/133 [===========================>..] - ETA: 0s - loss: 0.0369 133/133 [==============================] - 0s 1ms/step - loss: 0.0362 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0409 42/133 [========>.....................] - ETA: 0s - loss: 0.0312 82/133 [=================>............] - ETA: 0s - loss: 0.0314 123/133 [==========================>...] - ETA: 0s - loss: 0.0330 133/133 [==============================] - 0s 1ms/step - loss: 0.0326 -> test with GAN.predict GAN tn, fp: 329, 3 GAN fn, tp: 4, 10 GAN f1 score: 0.741 GAN cohens kappa score: 0.730 -> test with 'LR' LR tn, fp: 181, 151 LR fn, tp: 5, 9 LR f1 score: 0.103 LR cohens kappa score: 0.031 LR average precision score: 0.071 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 5, 9 RF f1 score: 0.783 RF cohens kappa score: 0.775 -> test with 'GB' GB tn, fp: 332, 0 GB fn, tp: 0, 14 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 328, 4 KNN fn, tp: 2, 12 KNN f1 score: 0.800 KNN cohens kappa score: 0.791 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 22s - loss: 2.3938e-06 48/133 [=========>....................] - ETA: 0s - loss: 0.0885  96/133 [====================>.........] - ETA: 0s - loss: 0.1035 133/133 [==============================] - 0s 1ms/step - loss: 0.1003 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 1.4861e-05 49/133 [==========>...................] - ETA: 0s - loss: 0.0625  97/133 [====================>.........] - ETA: 0s - loss: 0.0632 133/133 [==============================] - 0s 1ms/step - loss: 0.0631 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.2432 49/133 [==========>...................] - ETA: 0s - loss: 0.0689 97/133 [====================>.........] - ETA: 0s - loss: 0.0523 133/133 [==============================] - 0s 1ms/step - loss: 0.0512 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0576 49/133 [==========>...................] - ETA: 0s - loss: 0.0474 97/133 [====================>.........] - ETA: 0s - loss: 0.0475 133/133 [==============================] - 0s 1ms/step - loss: 0.0425 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0011 49/133 [==========>...................] - ETA: 0s - loss: 0.0387 97/133 [====================>.........] - ETA: 0s - loss: 0.0376 133/133 [==============================] - 0s 1ms/step - loss: 0.0393 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0022 45/133 [=========>....................] - ETA: 0s - loss: 0.0297 87/133 [==================>...........] - ETA: 0s - loss: 0.0400 128/133 [===========================>..] - ETA: 0s - loss: 0.0371 133/133 [==============================] - 0s 1ms/step - loss: 0.0366 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.1226 49/133 [==========>...................] - ETA: 0s - loss: 0.0448 97/133 [====================>.........] - ETA: 0s - loss: 0.0362 133/133 [==============================] - 0s 1ms/step - loss: 0.0369 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0024 49/133 [==========>...................] - ETA: 0s - loss: 0.0459 97/133 [====================>.........] - ETA: 0s - loss: 0.0374 133/133 [==============================] - 0s 1ms/step - loss: 0.0355 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0181 49/133 [==========>...................] - ETA: 0s - loss: 0.0316 97/133 [====================>.........] - ETA: 0s - loss: 0.0333 133/133 [==============================] - 0s 1ms/step - loss: 0.0313 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0023 49/133 [==========>...................] - ETA: 0s - loss: 0.0322 97/133 [====================>.........] - ETA: 0s - loss: 0.0317 133/133 [==============================] - 0s 1ms/step - loss: 0.0296 -> test with GAN.predict GAN tn, fp: 324, 8 GAN fn, tp: 7, 7 GAN f1 score: 0.483 GAN cohens kappa score: 0.460 -> test with 'LR' LR tn, fp: 184, 148 LR fn, tp: 5, 9 LR f1 score: 0.105 LR cohens kappa score: 0.033 LR average precision score: 0.070 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 10, 4 RF f1 score: 0.444 RF cohens kappa score: 0.434 -> test with 'GB' GB tn, fp: 330, 2 GB fn, tp: 2, 12 GB f1 score: 0.857 GB cohens kappa score: 0.851 -> test with 'KNN' KNN tn, fp: 306, 26 KNN fn, tp: 3, 11 KNN f1 score: 0.431 KNN cohens kappa score: 0.396 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 21s - loss: 0.3390 49/133 [==========>...................] - ETA: 0s - loss: 0.2199  94/133 [====================>.........] - ETA: 0s - loss: 0.1599 133/133 [==============================] - 0s 1ms/step - loss: 0.1304 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0101 46/133 [=========>....................] - ETA: 0s - loss: 0.0681 94/133 [====================>.........] - ETA: 0s - loss: 0.0551 133/133 [==============================] - 0s 1ms/step - loss: 0.0644 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 1.7111e-04 49/133 [==========>...................] - ETA: 0s - loss: 0.0511  97/133 [====================>.........] - ETA: 0s - loss: 0.0542 133/133 [==============================] - 0s 1ms/step - loss: 0.0532 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0502 49/133 [==========>...................] - ETA: 0s - loss: 0.0414 97/133 [====================>.........] - ETA: 0s - loss: 0.0465 133/133 [==============================] - 0s 1ms/step - loss: 0.0487 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 8.5658e-05 45/133 [=========>....................] - ETA: 0s - loss: 0.0443  86/133 [==================>...........] - ETA: 0s - loss: 0.0508 129/133 [============================>.] - ETA: 0s - loss: 0.0449 133/133 [==============================] - 0s 1ms/step - loss: 0.0440 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0018 41/133 [========>.....................] - ETA: 0s - loss: 0.0253 89/133 [===================>..........] - ETA: 0s - loss: 0.0339 133/133 [==============================] - 0s 1ms/step - loss: 0.0408 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0255 49/133 [==========>...................] - ETA: 0s - loss: 0.0295 97/133 [====================>.........] - ETA: 0s - loss: 0.0411 133/133 [==============================] - 0s 1ms/step - loss: 0.0370 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0523 49/133 [==========>...................] - ETA: 0s - loss: 0.0420 97/133 [====================>.........] - ETA: 0s - loss: 0.0381 133/133 [==============================] - 0s 1ms/step - loss: 0.0344 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0120 49/133 [==========>...................] - ETA: 0s - loss: 0.0428 97/133 [====================>.........] - ETA: 0s - loss: 0.0349 133/133 [==============================] - 0s 1ms/step - loss: 0.0323 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0013 49/133 [==========>...................] - ETA: 0s - loss: 0.0344 95/133 [====================>.........] - ETA: 0s - loss: 0.0275 133/133 [==============================] - 0s 1ms/step - loss: 0.0315 -> test with GAN.predict GAN tn, fp: 321, 11 GAN fn, tp: 5, 9 GAN f1 score: 0.529 GAN cohens kappa score: 0.506 -> test with 'LR' LR tn, fp: 181, 151 LR fn, tp: 5, 9 LR f1 score: 0.103 LR cohens kappa score: 0.031 LR average precision score: 0.070 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 7, 7 RF f1 score: 0.667 RF cohens kappa score: 0.657 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 1, 13 GB f1 score: 0.929 GB cohens kappa score: 0.926 -> test with 'KNN' KNN tn, fp: 307, 25 KNN fn, tp: 0, 14 KNN f1 score: 0.528 KNN cohens kappa score: 0.498 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 20s - loss: 0.8599 49/133 [==========>...................] - ETA: 0s - loss: 0.1261  97/133 [====================>.........] - ETA: 0s - loss: 0.1222 133/133 [==============================] - 0s 1ms/step - loss: 0.1177 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 5.8927e-05 50/133 [==========>...................] - ETA: 0s - loss: 0.0575  99/133 [=====================>........] - ETA: 0s - loss: 0.0722 133/133 [==============================] - 0s 1ms/step - loss: 0.0690 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.3585 49/133 [==========>...................] - ETA: 0s - loss: 0.0592 97/133 [====================>.........] - ETA: 0s - loss: 0.0598 133/133 [==============================] - 0s 1ms/step - loss: 0.0554 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0040 50/133 [==========>...................] - ETA: 0s - loss: 0.0482 99/133 [=====================>........] - ETA: 0s - loss: 0.0566 133/133 [==============================] - 0s 1ms/step - loss: 0.0509 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0480 50/133 [==========>...................] - ETA: 0s - loss: 0.0464 99/133 [=====================>........] - ETA: 0s - loss: 0.0465 133/133 [==============================] - 0s 1ms/step - loss: 0.0464 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.1608 50/133 [==========>...................] - ETA: 0s - loss: 0.0480 99/133 [=====================>........] - ETA: 0s - loss: 0.0449 133/133 [==============================] - 0s 1ms/step - loss: 0.0442 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0124 50/133 [==========>...................] - ETA: 0s - loss: 0.0456 96/133 [====================>.........] - ETA: 0s - loss: 0.0434 133/133 [==============================] - 0s 1ms/step - loss: 0.0419 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0033 49/133 [==========>...................] - ETA: 0s - loss: 0.0350 98/133 [=====================>........] - ETA: 0s - loss: 0.0401 133/133 [==============================] - 0s 1ms/step - loss: 0.0400 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0208 50/133 [==========>...................] - ETA: 0s - loss: 0.0290 99/133 [=====================>........] - ETA: 0s - loss: 0.0301 133/133 [==============================] - 0s 1ms/step - loss: 0.0411 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.2005 49/133 [==========>...................] - ETA: 0s - loss: 0.0562 98/133 [=====================>........] - ETA: 0s - loss: 0.0427 133/133 [==============================] - 0s 1ms/step - loss: 0.0372 -> test with GAN.predict GAN tn, fp: 323, 9 GAN fn, tp: 4, 10 GAN f1 score: 0.606 GAN cohens kappa score: 0.587 -> test with 'LR' LR tn, fp: 203, 129 LR fn, tp: 6, 8 LR f1 score: 0.106 LR cohens kappa score: 0.035 LR average precision score: 0.057 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 4, 10 RF f1 score: 0.833 RF cohens kappa score: 0.828 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 1, 13 GB f1 score: 0.929 GB cohens kappa score: 0.926 -> test with 'KNN' KNN tn, fp: 318, 14 KNN fn, tp: 2, 12 KNN f1 score: 0.600 KNN cohens kappa score: 0.578 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 19s - loss: 0.0537 49/133 [==========>...................] - ETA: 0s - loss: 0.1042  98/133 [=====================>........] - ETA: 0s - loss: 0.1186 133/133 [==============================] - 0s 1ms/step - loss: 0.1266 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0372 49/133 [==========>...................] - ETA: 0s - loss: 0.0827 98/133 [=====================>........] - ETA: 0s - loss: 0.0674 133/133 [==============================] - 0s 1ms/step - loss: 0.0786 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0147 49/133 [==========>...................] - ETA: 0s - loss: 0.0818 98/133 [=====================>........] - ETA: 0s - loss: 0.0636 133/133 [==============================] - 0s 1ms/step - loss: 0.0613 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0772 49/133 [==========>...................] - ETA: 0s - loss: 0.0460 98/133 [=====================>........] - ETA: 0s - loss: 0.0508 133/133 [==============================] - 0s 1ms/step - loss: 0.0550 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0014 49/133 [==========>...................] - ETA: 0s - loss: 0.0590 98/133 [=====================>........] - ETA: 0s - loss: 0.0534 133/133 [==============================] - 0s 1ms/step - loss: 0.0492 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0511 50/133 [==========>...................] - ETA: 0s - loss: 0.0499 99/133 [=====================>........] - ETA: 0s - loss: 0.0469 133/133 [==============================] - 0s 1ms/step - loss: 0.0452 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0233 50/133 [==========>...................] - ETA: 0s - loss: 0.0275 99/133 [=====================>........] - ETA: 0s - loss: 0.0378 133/133 [==============================] - 0s 1ms/step - loss: 0.0451 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0050 50/133 [==========>...................] - ETA: 0s - loss: 0.0385 98/133 [=====================>........] - ETA: 0s - loss: 0.0422 133/133 [==============================] - 0s 1ms/step - loss: 0.0418 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0120 50/133 [==========>...................] - ETA: 0s - loss: 0.0292 97/133 [====================>.........] - ETA: 0s - loss: 0.0399 133/133 [==============================] - 0s 1ms/step - loss: 0.0401 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0197 49/133 [==========>...................] - ETA: 0s - loss: 0.0358 93/133 [===================>..........] - ETA: 0s - loss: 0.0342 133/133 [==============================] - 0s 1ms/step - loss: 0.0389 -> test with GAN.predict GAN tn, fp: 324, 7 GAN fn, tp: 4, 9 GAN f1 score: 0.621 GAN cohens kappa score: 0.604 -> test with 'LR' LR tn, fp: 194, 137 LR fn, tp: 1, 12 LR f1 score: 0.148 LR cohens kappa score: 0.085 LR average precision score: 0.078 -> test with 'RF' RF tn, fp: 331, 0 RF fn, tp: 6, 7 RF f1 score: 0.700 RF cohens kappa score: 0.692 -> test with 'GB' GB tn, fp: 329, 2 GB fn, tp: 2, 11 GB f1 score: 0.846 GB cohens kappa score: 0.840 -> test with 'KNN' KNN tn, fp: 320, 11 KNN fn, tp: 6, 7 KNN f1 score: 0.452 KNN cohens kappa score: 0.426 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 21s - loss: 1.8190 47/133 [=========>....................] - ETA: 0s - loss: 0.1834  96/133 [====================>.........] - ETA: 0s - loss: 0.1310 133/133 [==============================] - 0s 1ms/step - loss: 0.1180 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 3.8675e-04 50/133 [==========>...................] - ETA: 0s - loss: 0.0720  99/133 [=====================>........] - ETA: 0s - loss: 0.0647 133/133 [==============================] - 0s 1ms/step - loss: 0.0660 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0791 50/133 [==========>...................] - ETA: 0s - loss: 0.0714 99/133 [=====================>........] - ETA: 0s - loss: 0.0535 133/133 [==============================] - 0s 1ms/step - loss: 0.0490 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0091 49/133 [==========>...................] - ETA: 0s - loss: 0.0387 97/133 [====================>.........] - ETA: 0s - loss: 0.0446 133/133 [==============================] - 0s 1ms/step - loss: 0.0458 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.1979 50/133 [==========>...................] - ETA: 0s - loss: 0.0409 99/133 [=====================>........] - ETA: 0s - loss: 0.0419 133/133 [==============================] - 0s 1ms/step - loss: 0.0393 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0159 50/133 [==========>...................] - ETA: 0s - loss: 0.0302 99/133 [=====================>........] - ETA: 0s - loss: 0.0293 133/133 [==============================] - 0s 1ms/step - loss: 0.0353 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0067 49/133 [==========>...................] - ETA: 0s - loss: 0.0338 97/133 [====================>.........] - ETA: 0s - loss: 0.0336 133/133 [==============================] - 0s 1ms/step - loss: 0.0349 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0022 49/133 [==========>...................] - ETA: 0s - loss: 0.0404 98/133 [=====================>........] - ETA: 0s - loss: 0.0278 133/133 [==============================] - 0s 1ms/step - loss: 0.0308 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0649 50/133 [==========>...................] - ETA: 0s - loss: 0.0503 99/133 [=====================>........] - ETA: 0s - loss: 0.0335 133/133 [==============================] - 0s 1ms/step - loss: 0.0299 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.3272 50/133 [==========>...................] - ETA: 0s - loss: 0.0335 99/133 [=====================>........] - ETA: 0s - loss: 0.0311 133/133 [==============================] - 0s 1ms/step - loss: 0.0287 -> test with GAN.predict GAN tn, fp: 321, 11 GAN fn, tp: 5, 9 GAN f1 score: 0.529 GAN cohens kappa score: 0.506 -> test with 'LR' LR tn, fp: 187, 145 LR fn, tp: 8, 6 LR f1 score: 0.073 LR cohens kappa score: -0.001 LR average precision score: 0.055 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 8, 6 RF f1 score: 0.600 RF cohens kappa score: 0.590 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 0, 14 GB f1 score: 0.966 GB cohens kappa score: 0.964 -> test with 'KNN' KNN tn, fp: 308, 24 KNN fn, tp: 1, 13 KNN f1 score: 0.510 KNN cohens kappa score: 0.479 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 21s - loss: 0.0245 49/133 [==========>...................] - ETA: 0s - loss: 0.1153  98/133 [=====================>........] - ETA: 0s - loss: 0.0925 133/133 [==============================] - 0s 1ms/step - loss: 0.0904 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.2831 48/133 [=========>....................] - ETA: 0s - loss: 0.0399 96/133 [====================>.........] - ETA: 0s - loss: 0.0521 133/133 [==============================] - 0s 1ms/step - loss: 0.0516 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0183 46/133 [=========>....................] - ETA: 0s - loss: 0.0580 94/133 [====================>.........] - ETA: 0s - loss: 0.0461 133/133 [==============================] - 0s 1ms/step - loss: 0.0430 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0100 49/133 [==========>...................] - ETA: 0s - loss: 0.0305 97/133 [====================>.........] - ETA: 0s - loss: 0.0369 133/133 [==============================] - 0s 1ms/step - loss: 0.0369 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0091 49/133 [==========>...................] - ETA: 0s - loss: 0.0396 97/133 [====================>.........] - ETA: 0s - loss: 0.0355 133/133 [==============================] - 0s 1ms/step - loss: 0.0351 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0478 49/133 [==========>...................] - ETA: 0s - loss: 0.0214 94/133 [====================>.........] - ETA: 0s - loss: 0.0308 133/133 [==============================] - 0s 1ms/step - loss: 0.0306 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0014 44/133 [========>.....................] - ETA: 0s - loss: 0.0176 90/133 [===================>..........] - ETA: 0s - loss: 0.0248 133/133 [==============================] - 0s 1ms/step - loss: 0.0313 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0022 49/133 [==========>...................] - ETA: 0s - loss: 0.0362 97/133 [====================>.........] - ETA: 0s - loss: 0.0289 133/133 [==============================] - 0s 1ms/step - loss: 0.0267 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0057 49/133 [==========>...................] - ETA: 0s - loss: 0.0355 95/133 [====================>.........] - ETA: 0s - loss: 0.0325 133/133 [==============================] - 0s 1ms/step - loss: 0.0270 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0697 49/133 [==========>...................] - ETA: 0s - loss: 0.0283 97/133 [====================>.........] - ETA: 0s - loss: 0.0294 133/133 [==============================] - 0s 1ms/step - loss: 0.0286 -> test with GAN.predict GAN tn, fp: 330, 2 GAN fn, tp: 4, 10 GAN f1 score: 0.769 GAN cohens kappa score: 0.760 -> test with 'LR' LR tn, fp: 195, 137 LR fn, tp: 6, 8 LR f1 score: 0.101 LR cohens kappa score: 0.029 LR average precision score: 0.078 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 6, 8 RF f1 score: 0.727 RF cohens kappa score: 0.719 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 2, 12 GB f1 score: 0.889 GB cohens kappa score: 0.884 -> test with 'KNN' KNN tn, fp: 311, 21 KNN fn, tp: 0, 14 KNN f1 score: 0.571 KNN cohens kappa score: 0.545 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 23s - loss: 2.6944e-04 49/133 [==========>...................] - ETA: 0s - loss: 0.0733  97/133 [====================>.........] - ETA: 0s - loss: 0.0721 133/133 [==============================] - 0s 1ms/step - loss: 0.0640 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.1594 49/133 [==========>...................] - ETA: 0s - loss: 0.0557 97/133 [====================>.........] - ETA: 0s - loss: 0.0507 133/133 [==============================] - 0s 1ms/step - loss: 0.0438 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0025 49/133 [==========>...................] - ETA: 0s - loss: 0.0326 98/133 [=====================>........] - ETA: 0s - loss: 0.0365 133/133 [==============================] - 0s 1ms/step - loss: 0.0401 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0147 49/133 [==========>...................] - ETA: 0s - loss: 0.0352 96/133 [====================>.........] - ETA: 0s - loss: 0.0356 133/133 [==============================] - 0s 1ms/step - loss: 0.0375 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0046 46/133 [=========>....................] - ETA: 0s - loss: 0.0399 94/133 [====================>.........] - ETA: 0s - loss: 0.0389 133/133 [==============================] - 0s 1ms/step - loss: 0.0344 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0049 49/133 [==========>...................] - ETA: 0s - loss: 0.0223 97/133 [====================>.........] - ETA: 0s - loss: 0.0335 133/133 [==============================] - 0s 1ms/step - loss: 0.0333 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0339 49/133 [==========>...................] - ETA: 0s - loss: 0.0303 97/133 [====================>.........] - ETA: 0s - loss: 0.0359 133/133 [==============================] - 0s 1ms/step - loss: 0.0324 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0069 49/133 [==========>...................] - ETA: 0s - loss: 0.0314 97/133 [====================>.........] - ETA: 0s - loss: 0.0301 133/133 [==============================] - 0s 1ms/step - loss: 0.0323 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0390 50/133 [==========>...................] - ETA: 0s - loss: 0.0379 98/133 [=====================>........] - ETA: 0s - loss: 0.0327 133/133 [==============================] - 0s 1ms/step - loss: 0.0281 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0029 46/133 [=========>....................] - ETA: 0s - loss: 0.0240 88/133 [==================>...........] - ETA: 0s - loss: 0.0229 129/133 [============================>.] - ETA: 0s - loss: 0.0281 133/133 [==============================] - 0s 1ms/step - loss: 0.0275 -> test with GAN.predict GAN tn, fp: 321, 11 GAN fn, tp: 3, 11 GAN f1 score: 0.611 GAN cohens kappa score: 0.591 -> test with 'LR' LR tn, fp: 172, 160 LR fn, tp: 4, 10 LR f1 score: 0.109 LR cohens kappa score: 0.037 LR average precision score: 0.093 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 5, 9 RF f1 score: 0.783 RF cohens kappa score: 0.775 -> test with 'GB' GB tn, fp: 329, 3 GB fn, tp: 1, 13 GB f1 score: 0.867 GB cohens kappa score: 0.861 -> test with 'KNN' KNN tn, fp: 312, 20 KNN fn, tp: 1, 13 KNN f1 score: 0.553 KNN cohens kappa score: 0.526 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 23s - loss: 0.0120 49/133 [==========>...................] - ETA: 0s - loss: 0.0994  97/133 [====================>.........] - ETA: 0s - loss: 0.1113 133/133 [==============================] - 0s 1ms/step - loss: 0.0983 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0041 50/133 [==========>...................] - ETA: 0s - loss: 0.0495 98/133 [=====================>........] - ETA: 0s - loss: 0.0464 133/133 [==============================] - 0s 1ms/step - loss: 0.0511 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0112 49/133 [==========>...................] - ETA: 0s - loss: 0.0413 96/133 [====================>.........] - ETA: 0s - loss: 0.0454 133/133 [==============================] - 0s 1ms/step - loss: 0.0455 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0459 48/133 [=========>....................] - ETA: 0s - loss: 0.0375 96/133 [====================>.........] - ETA: 0s - loss: 0.0307 133/133 [==============================] - 0s 1ms/step - loss: 0.0380 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0447 50/133 [==========>...................] - ETA: 0s - loss: 0.0245 95/133 [====================>.........] - ETA: 0s - loss: 0.0306 133/133 [==============================] - 0s 1ms/step - loss: 0.0348 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0246 48/133 [=========>....................] - ETA: 0s - loss: 0.0484 96/133 [====================>.........] - ETA: 0s - loss: 0.0339 133/133 [==============================] - 0s 1ms/step - loss: 0.0320 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 8.0547e-04 49/133 [==========>...................] - ETA: 0s - loss: 0.0363  97/133 [====================>.........] - ETA: 0s - loss: 0.0335 133/133 [==============================] - 0s 1ms/step - loss: 0.0299 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0175 49/133 [==========>...................] - ETA: 0s - loss: 0.0271 97/133 [====================>.........] - ETA: 0s - loss: 0.0279 133/133 [==============================] - 0s 1ms/step - loss: 0.0293 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0600 46/133 [=========>....................] - ETA: 0s - loss: 0.0261 88/133 [==================>...........] - ETA: 0s - loss: 0.0304 131/133 [============================>.] - ETA: 0s - loss: 0.0266 133/133 [==============================] - 0s 1ms/step - loss: 0.0263 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 0.0062 49/133 [==========>...................] - ETA: 0s - loss: 0.0265 96/133 [====================>.........] - ETA: 0s - loss: 0.0316 133/133 [==============================] - 0s 1ms/step - loss: 0.0266 -> test with GAN.predict GAN tn, fp: 328, 4 GAN fn, tp: 7, 7 GAN f1 score: 0.560 GAN cohens kappa score: 0.544 -> test with 'LR' LR tn, fp: 179, 153 LR fn, tp: 4, 10 LR f1 score: 0.113 LR cohens kappa score: 0.042 LR average precision score: 0.084 -> test with 'RF' RF tn, fp: 332, 0 RF fn, tp: 8, 6 RF f1 score: 0.600 RF cohens kappa score: 0.590 -> test with 'GB' GB tn, fp: 331, 1 GB fn, tp: 6, 8 GB f1 score: 0.696 GB cohens kappa score: 0.686 -> test with 'KNN' KNN tn, fp: 314, 18 KNN fn, tp: 0, 14 KNN f1 score: 0.609 KNN cohens kappa score: 0.585 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1272 synthetic samples -> retrain GAN for predict Epoch 1/10 1/133 [..............................] - ETA: 20s - loss: 3.1405e-06 44/133 [========>.....................] - ETA: 0s - loss: 0.1328  93/133 [===================>..........] - ETA: 0s - loss: 0.1178 133/133 [==============================] - 0s 1ms/step - loss: 0.0992 Epoch 2/10 1/133 [..............................] - ETA: 0s - loss: 0.0012 48/133 [=========>....................] - ETA: 0s - loss: 0.0411 96/133 [====================>.........] - ETA: 0s - loss: 0.0471 133/133 [==============================] - 0s 1ms/step - loss: 0.0452 Epoch 3/10 1/133 [..............................] - ETA: 0s - loss: 0.0049 48/133 [=========>....................] - ETA: 0s - loss: 0.0215 96/133 [====================>.........] - ETA: 0s - loss: 0.0278 133/133 [==============================] - 0s 1ms/step - loss: 0.0357 Epoch 4/10 1/133 [..............................] - ETA: 0s - loss: 0.0936 45/133 [=========>....................] - ETA: 0s - loss: 0.0623 85/133 [==================>...........] - ETA: 0s - loss: 0.0468 128/133 [===========================>..] - ETA: 0s - loss: 0.0406 133/133 [==============================] - 0s 1ms/step - loss: 0.0413 Epoch 5/10 1/133 [..............................] - ETA: 0s - loss: 0.0215 49/133 [==========>...................] - ETA: 0s - loss: 0.0318 97/133 [====================>.........] - ETA: 0s - loss: 0.0292 133/133 [==============================] - 0s 1ms/step - loss: 0.0251 Epoch 6/10 1/133 [..............................] - ETA: 0s - loss: 0.0065 49/133 [==========>...................] - ETA: 0s - loss: 0.0297 97/133 [====================>.........] - ETA: 0s - loss: 0.0254 133/133 [==============================] - 0s 1ms/step - loss: 0.0238 Epoch 7/10 1/133 [..............................] - ETA: 0s - loss: 0.0016 49/133 [==========>...................] - ETA: 0s - loss: 0.0182 97/133 [====================>.........] - ETA: 0s - loss: 0.0275 133/133 [==============================] - 0s 1ms/step - loss: 0.0246 Epoch 8/10 1/133 [..............................] - ETA: 0s - loss: 0.0591 49/133 [==========>...................] - ETA: 0s - loss: 0.0176 97/133 [====================>.........] - ETA: 0s - loss: 0.0223 133/133 [==============================] - 0s 1ms/step - loss: 0.0211 Epoch 9/10 1/133 [..............................] - ETA: 0s - loss: 0.0110 49/133 [==========>...................] - ETA: 0s - loss: 0.0164 97/133 [====================>.........] - ETA: 0s - loss: 0.0214 133/133 [==============================] - 0s 1ms/step - loss: 0.0222 Epoch 10/10 1/133 [..............................] - ETA: 0s - loss: 1.4743e-04 49/133 [==========>...................] - ETA: 0s - loss: 0.0103  97/133 [====================>.........] - ETA: 0s - loss: 0.0183 133/133 [==============================] - 0s 1ms/step - loss: 0.0227 -> test with GAN.predict GAN tn, fp: 322, 9 GAN fn, tp: 0, 13 GAN f1 score: 0.743 GAN cohens kappa score: 0.730 -> test with 'LR' LR tn, fp: 187, 144 LR fn, tp: 3, 10 LR f1 score: 0.120 LR cohens kappa score: 0.054 LR average precision score: 0.066 -> test with 'RF' RF tn, fp: 330, 1 RF fn, tp: 6, 7 RF f1 score: 0.667 RF cohens kappa score: 0.657 -> test with 'GB' GB tn, fp: 330, 1 GB fn, tp: 1, 12 GB f1 score: 0.923 GB cohens kappa score: 0.920 -> test with 'KNN' KNN tn, fp: 311, 20 KNN fn, tp: 0, 13 KNN f1 score: 0.565 KNN cohens kappa score: 0.540 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 203, 160 LR fn, tp: 8, 12 LR f1 score: 0.148 LR cohens kappa score: 0.085 LR average precision score: 0.093 average: LR tn, fp: 187.76, 144.04 LR fn, tp: 4.64, 9.16 LR f1 score: 0.110 LR cohens kappa score: 0.039 LR average precision score: 0.072 minimum: LR tn, fp: 172, 129 LR fn, tp: 1, 6 LR f1 score: 0.073 LR cohens kappa score: -0.001 LR average precision score: 0.050 -----[ RF ]----- maximum: RF tn, fp: 332, 1 RF fn, tp: 11, 10 RF f1 score: 0.833 RF cohens kappa score: 0.828 average: RF tn, fp: 331.72, 0.08 RF fn, tp: 7.0, 6.8 RF f1 score: 0.647 RF cohens kappa score: 0.638 minimum: RF tn, fp: 330, 0 RF fn, tp: 4, 3 RF f1 score: 0.353 RF cohens kappa score: 0.344 -----[ GB ]----- maximum: GB tn, fp: 332, 3 GB fn, tp: 7, 14 GB f1 score: 1.000 GB cohens kappa score: 1.000 average: GB tn, fp: 330.56, 1.24 GB fn, tp: 2.16, 11.64 GB f1 score: 0.869 GB cohens kappa score: 0.864 minimum: GB tn, fp: 328, 0 GB fn, tp: 0, 7 GB f1 score: 0.609 GB cohens kappa score: 0.596 -----[ KNN ]----- maximum: KNN tn, fp: 329, 26 KNN fn, tp: 6, 14 KNN f1 score: 0.848 KNN cohens kappa score: 0.841 average: KNN tn, fp: 316.12, 15.68 KNN fn, tp: 1.88, 11.92 KNN f1 score: 0.585 KNN cohens kappa score: 0.561 minimum: KNN tn, fp: 306, 3 KNN fn, tp: 0, 7 KNN f1 score: 0.431 KNN cohens kappa score: 0.396 -----[ GAN ]----- maximum: GAN tn, fp: 330, 12 GAN fn, tp: 10, 13 GAN f1 score: 0.769 GAN cohens kappa score: 0.760 average: GAN tn, fp: 323.88, 7.92 GAN fn, tp: 4.92, 8.88 GAN f1 score: 0.577 GAN cohens kappa score: 0.558 minimum: GAN tn, fp: 320, 2 GAN fn, tp: 0, 4 GAN f1 score: 0.348 GAN cohens kappa score: 0.326