/////////////////////////////////////////// // Running convGAN-majority-5 on folding_hypothyroid /////////////////////////////////////////// Load 'data_input/folding_hypothyroid' from pickle file non empty cut in data_input/folding_hypothyroid! (1 points) 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 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 579, 24 GAN fn, tp: 5, 26 GAN f1 score: 0.642 GAN cohens kappa score: 0.619 -> test with 'LR' LR tn, fp: 542, 61 LR fn, tp: 5, 26 LR f1 score: 0.441 LR cohens kappa score: 0.397 LR average precision score: 0.462 -> test with 'GB' GB tn, fp: 595, 8 GB fn, tp: 4, 27 GB f1 score: 0.818 GB cohens kappa score: 0.808 -> test with 'KNN' KNN tn, fp: 581, 22 KNN fn, tp: 6, 25 KNN f1 score: 0.641 KNN cohens kappa score: 0.619 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 549, 54 GAN fn, tp: 4, 27 GAN f1 score: 0.482 GAN cohens kappa score: 0.443 -> test with 'LR' LR tn, fp: 515, 88 LR fn, tp: 3, 28 LR f1 score: 0.381 LR cohens kappa score: 0.329 LR average precision score: 0.460 -> test with 'GB' GB tn, fp: 591, 12 GB fn, tp: 2, 29 GB f1 score: 0.806 GB cohens kappa score: 0.794 -> test with 'KNN' KNN tn, fp: 571, 32 KNN fn, tp: 7, 24 KNN f1 score: 0.552 KNN cohens kappa score: 0.522 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 544, 59 GAN fn, tp: 6, 25 GAN f1 score: 0.435 GAN cohens kappa score: 0.391 -> test with 'LR' LR tn, fp: 510, 93 LR fn, tp: 6, 25 LR f1 score: 0.336 LR cohens kappa score: 0.280 LR average precision score: 0.327 -> test with 'GB' GB tn, fp: 589, 14 GB fn, tp: 2, 29 GB f1 score: 0.784 GB cohens kappa score: 0.771 -> test with 'KNN' KNN tn, fp: 574, 29 KNN fn, tp: 6, 25 KNN f1 score: 0.588 KNN cohens kappa score: 0.561 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 550, 53 GAN fn, tp: 7, 24 GAN f1 score: 0.444 GAN cohens kappa score: 0.403 -> test with 'LR' LR tn, fp: 505, 98 LR fn, tp: 4, 27 LR f1 score: 0.346 LR cohens kappa score: 0.291 LR average precision score: 0.398 -> test with 'GB' GB tn, fp: 594, 9 GB fn, tp: 6, 25 GB f1 score: 0.769 GB cohens kappa score: 0.757 -> test with 'KNN' KNN tn, fp: 569, 34 KNN fn, tp: 11, 20 KNN f1 score: 0.471 KNN cohens kappa score: 0.436 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2288 synthetic samples -> test with GAN.predict GAN tn, fp: 591, 9 GAN fn, tp: 6, 21 GAN f1 score: 0.737 GAN cohens kappa score: 0.724 -> test with 'LR' LR tn, fp: 527, 73 LR fn, tp: 2, 25 LR f1 score: 0.400 LR cohens kappa score: 0.357 LR average precision score: 0.552 -> test with 'GB' GB tn, fp: 593, 7 GB fn, tp: 3, 24 GB f1 score: 0.828 GB cohens kappa score: 0.819 -> test with 'KNN' KNN tn, fp: 568, 32 KNN fn, tp: 3, 24 KNN f1 score: 0.578 KNN cohens kappa score: 0.552 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 581, 22 GAN fn, tp: 5, 26 GAN f1 score: 0.658 GAN cohens kappa score: 0.637 -> test with 'LR' LR tn, fp: 529, 74 LR fn, tp: 6, 25 LR f1 score: 0.385 LR cohens kappa score: 0.335 LR average precision score: 0.418 -> test with 'GB' GB tn, fp: 591, 12 GB fn, tp: 5, 26 GB f1 score: 0.754 GB cohens kappa score: 0.740 -> test with 'KNN' KNN tn, fp: 583, 20 KNN fn, tp: 5, 26 KNN f1 score: 0.675 KNN cohens kappa score: 0.655 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 584, 19 GAN fn, tp: 9, 22 GAN f1 score: 0.611 GAN cohens kappa score: 0.588 -> test with 'LR' LR tn, fp: 536, 67 LR fn, tp: 6, 25 LR f1 score: 0.407 LR cohens kappa score: 0.360 LR average precision score: 0.437 -> test with 'GB' GB tn, fp: 597, 6 GB fn, tp: 4, 27 GB f1 score: 0.844 GB cohens kappa score: 0.835 -> test with 'KNN' KNN tn, fp: 570, 33 KNN fn, tp: 4, 27 KNN f1 score: 0.593 KNN cohens kappa score: 0.565 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 579, 24 GAN fn, tp: 8, 23 GAN f1 score: 0.590 GAN cohens kappa score: 0.564 -> test with 'LR' LR tn, fp: 514, 89 LR fn, tp: 5, 26 LR f1 score: 0.356 LR cohens kappa score: 0.302 LR average precision score: 0.569 -> test with 'GB' GB tn, fp: 594, 9 GB fn, tp: 6, 25 GB f1 score: 0.769 GB cohens kappa score: 0.757 -> test with 'KNN' KNN tn, fp: 578, 25 KNN fn, tp: 8, 23 KNN f1 score: 0.582 KNN cohens kappa score: 0.556 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 580, 23 GAN fn, tp: 10, 21 GAN f1 score: 0.560 GAN cohens kappa score: 0.533 -> test with 'LR' LR tn, fp: 509, 94 LR fn, tp: 6, 25 LR f1 score: 0.333 LR cohens kappa score: 0.277 LR average precision score: 0.281 -> test with 'GB' GB tn, fp: 588, 15 GB fn, tp: 5, 26 GB f1 score: 0.722 GB cohens kappa score: 0.706 -> test with 'KNN' KNN tn, fp: 575, 28 KNN fn, tp: 7, 24 KNN f1 score: 0.578 KNN cohens kappa score: 0.551 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2288 synthetic samples -> test with GAN.predict GAN tn, fp: 565, 35 GAN fn, tp: 1, 26 GAN f1 score: 0.591 GAN cohens kappa score: 0.565 -> test with 'LR' LR tn, fp: 500, 100 LR fn, tp: 1, 26 LR f1 score: 0.340 LR cohens kappa score: 0.289 LR average precision score: 0.483 -> test with 'GB' GB tn, fp: 590, 10 GB fn, tp: 2, 25 GB f1 score: 0.806 GB cohens kappa score: 0.797 -> test with 'KNN' KNN tn, fp: 567, 33 KNN fn, tp: 5, 22 KNN f1 score: 0.537 KNN cohens kappa score: 0.508 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 576, 27 GAN fn, tp: 5, 26 GAN f1 score: 0.619 GAN cohens kappa score: 0.594 -> test with 'LR' LR tn, fp: 509, 94 LR fn, tp: 3, 28 LR f1 score: 0.366 LR cohens kappa score: 0.312 LR average precision score: 0.477 -> test with 'GB' GB tn, fp: 600, 3 GB fn, tp: 7, 24 GB f1 score: 0.828 GB cohens kappa score: 0.819 -> test with 'KNN' KNN tn, fp: 579, 24 KNN fn, tp: 9, 22 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 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 549, 54 GAN fn, tp: 3, 28 GAN f1 score: 0.496 GAN cohens kappa score: 0.457 -> test with 'LR' LR tn, fp: 528, 75 LR fn, tp: 10, 21 LR f1 score: 0.331 LR cohens kappa score: 0.277 LR average precision score: 0.288 -> test with 'GB' GB tn, fp: 586, 17 GB fn, tp: 3, 28 GB f1 score: 0.737 GB cohens kappa score: 0.721 -> test with 'KNN' KNN tn, fp: 571, 32 KNN fn, tp: 6, 25 KNN f1 score: 0.568 KNN cohens kappa score: 0.539 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 583, 20 GAN fn, tp: 6, 25 GAN f1 score: 0.658 GAN cohens kappa score: 0.637 -> test with 'LR' LR tn, fp: 517, 86 LR fn, tp: 1, 30 LR f1 score: 0.408 LR cohens kappa score: 0.359 LR average precision score: 0.546 -> test with 'GB' GB tn, fp: 588, 15 GB fn, tp: 2, 29 GB f1 score: 0.773 GB cohens kappa score: 0.760 -> test with 'KNN' KNN tn, fp: 576, 27 KNN fn, tp: 7, 24 KNN f1 score: 0.585 KNN cohens kappa score: 0.559 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 572, 31 GAN fn, tp: 8, 23 GAN f1 score: 0.541 GAN cohens kappa score: 0.511 -> test with 'LR' LR tn, fp: 507, 96 LR fn, tp: 2, 29 LR f1 score: 0.372 LR cohens kappa score: 0.318 LR average precision score: 0.452 -> test with 'GB' GB tn, fp: 593, 10 GB fn, tp: 6, 25 GB f1 score: 0.758 GB cohens kappa score: 0.744 -> test with 'KNN' KNN tn, fp: 569, 34 KNN fn, tp: 6, 25 KNN f1 score: 0.556 KNN cohens kappa score: 0.525 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2288 synthetic samples -> test with GAN.predict GAN tn, fp: 587, 13 GAN fn, tp: 5, 22 GAN f1 score: 0.710 GAN cohens kappa score: 0.695 -> test with 'LR' LR tn, fp: 512, 88 LR fn, tp: 5, 22 LR f1 score: 0.321 LR cohens kappa score: 0.271 LR average precision score: 0.308 -> test with 'GB' GB tn, fp: 594, 6 GB fn, tp: 1, 26 GB f1 score: 0.881 GB cohens kappa score: 0.876 -> test with 'KNN' KNN tn, fp: 578, 22 KNN fn, tp: 2, 25 KNN f1 score: 0.676 KNN cohens kappa score: 0.657 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 587, 16 GAN fn, tp: 6, 25 GAN f1 score: 0.694 GAN cohens kappa score: 0.676 -> test with 'LR' LR tn, fp: 522, 81 LR fn, tp: 4, 27 LR f1 score: 0.388 LR cohens kappa score: 0.338 LR average precision score: 0.395 -> test with 'GB' GB tn, fp: 591, 12 GB fn, tp: 4, 27 GB f1 score: 0.771 GB cohens kappa score: 0.758 -> test with 'KNN' KNN tn, fp: 571, 32 KNN fn, tp: 4, 27 KNN f1 score: 0.600 KNN cohens kappa score: 0.573 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 547, 56 GAN fn, tp: 5, 26 GAN f1 score: 0.460 GAN cohens kappa score: 0.419 -> test with 'LR' LR tn, fp: 523, 80 LR fn, tp: 5, 26 LR f1 score: 0.380 LR cohens kappa score: 0.329 LR average precision score: 0.436 -> test with 'GB' GB tn, fp: 594, 9 GB fn, tp: 3, 28 GB f1 score: 0.824 GB cohens kappa score: 0.814 -> test with 'KNN' KNN tn, fp: 571, 32 KNN fn, tp: 3, 28 KNN f1 score: 0.615 KNN cohens kappa score: 0.589 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 583, 20 GAN fn, tp: 6, 25 GAN f1 score: 0.658 GAN cohens kappa score: 0.637 -> test with 'LR' LR tn, fp: 501, 102 LR fn, tp: 2, 29 LR f1 score: 0.358 LR cohens kappa score: 0.303 LR average precision score: 0.576 -> test with 'GB' GB tn, fp: 595, 8 GB fn, tp: 6, 25 GB f1 score: 0.781 GB cohens kappa score: 0.770 -> test with 'KNN' KNN tn, fp: 579, 24 KNN fn, tp: 5, 26 KNN f1 score: 0.642 KNN cohens kappa score: 0.619 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 558, 45 GAN fn, tp: 6, 25 GAN f1 score: 0.495 GAN cohens kappa score: 0.458 -> test with 'LR' LR tn, fp: 497, 106 LR fn, tp: 2, 29 LR f1 score: 0.349 LR cohens kappa score: 0.293 LR average precision score: 0.434 -> test with 'GB' GB tn, fp: 592, 11 GB fn, tp: 2, 29 GB f1 score: 0.817 GB cohens kappa score: 0.806 -> test with 'KNN' KNN tn, fp: 576, 27 KNN fn, tp: 7, 24 KNN f1 score: 0.585 KNN cohens kappa score: 0.559 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2288 synthetic samples -> test with GAN.predict GAN tn, fp: 569, 31 GAN fn, tp: 6, 21 GAN f1 score: 0.532 GAN cohens kappa score: 0.503 -> test with 'LR' LR tn, fp: 512, 88 LR fn, tp: 6, 21 LR f1 score: 0.309 LR cohens kappa score: 0.258 LR average precision score: 0.410 -> test with 'GB' GB tn, fp: 588, 12 GB fn, tp: 5, 22 GB f1 score: 0.721 GB cohens kappa score: 0.707 -> test with 'KNN' KNN tn, fp: 565, 35 KNN fn, tp: 6, 21 KNN f1 score: 0.506 KNN cohens kappa score: 0.476 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 574, 29 GAN fn, tp: 6, 25 GAN f1 score: 0.588 GAN cohens kappa score: 0.561 -> test with 'LR' LR tn, fp: 514, 89 LR fn, tp: 4, 27 LR f1 score: 0.367 LR cohens kappa score: 0.314 LR average precision score: 0.431 -> test with 'GB' GB tn, fp: 591, 12 GB fn, tp: 4, 27 GB f1 score: 0.771 GB cohens kappa score: 0.758 -> test with 'KNN' KNN tn, fp: 575, 28 KNN fn, tp: 7, 24 KNN f1 score: 0.578 KNN cohens kappa score: 0.551 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 586, 17 GAN fn, tp: 10, 21 GAN f1 score: 0.609 GAN cohens kappa score: 0.586 -> test with 'LR' LR tn, fp: 516, 87 LR fn, tp: 5, 26 LR f1 score: 0.361 LR cohens kappa score: 0.308 LR average precision score: 0.512 -> test with 'GB' GB tn, fp: 595, 8 GB fn, tp: 2, 29 GB f1 score: 0.853 GB cohens kappa score: 0.845 -> test with 'KNN' KNN tn, fp: 566, 37 KNN fn, tp: 7, 24 KNN f1 score: 0.522 KNN cohens kappa score: 0.489 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 582, 21 GAN fn, tp: 12, 19 GAN f1 score: 0.535 GAN cohens kappa score: 0.508 -> test with 'LR' LR tn, fp: 506, 97 LR fn, tp: 2, 29 LR f1 score: 0.369 LR cohens kappa score: 0.316 LR average precision score: 0.499 -> test with 'GB' GB tn, fp: 594, 9 GB fn, tp: 8, 23 GB f1 score: 0.730 GB cohens kappa score: 0.716 -> test with 'KNN' KNN tn, fp: 572, 31 KNN fn, tp: 8, 23 KNN f1 score: 0.541 KNN cohens kappa score: 0.511 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2289 synthetic samples -> test with GAN.predict GAN tn, fp: 580, 23 GAN fn, tp: 5, 26 GAN f1 score: 0.650 GAN cohens kappa score: 0.628 -> test with 'LR' LR tn, fp: 515, 88 LR fn, tp: 4, 27 LR f1 score: 0.370 LR cohens kappa score: 0.317 LR average precision score: 0.570 -> test with 'GB' GB tn, fp: 593, 10 GB fn, tp: 2, 29 GB f1 score: 0.829 GB cohens kappa score: 0.819 -> test with 'KNN' KNN tn, fp: 570, 33 KNN fn, tp: 6, 25 KNN f1 score: 0.562 KNN cohens kappa score: 0.532 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2288 synthetic samples -> test with GAN.predict GAN tn, fp: 594, 6 GAN fn, tp: 12, 15 GAN f1 score: 0.625 GAN cohens kappa score: 0.610 -> test with 'LR' LR tn, fp: 523, 77 LR fn, tp: 3, 24 LR f1 score: 0.375 LR cohens kappa score: 0.329 LR average precision score: 0.309 -> test with 'GB' GB tn, fp: 593, 7 GB fn, tp: 6, 21 GB f1 score: 0.764 GB cohens kappa score: 0.753 -> test with 'KNN' KNN tn, fp: 574, 26 KNN fn, tp: 5, 22 KNN f1 score: 0.587 KNN cohens kappa score: 0.563 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 542, 106 LR fn, tp: 10, 30 LR f1 score: 0.441 LR cohens kappa score: 0.397 LR average precision score: 0.576 average: LR tn, fp: 515.56, 86.84 LR fn, tp: 4.08, 26.12 LR f1 score: 0.366 LR cohens kappa score: 0.314 LR average precision score: 0.441 minimum: LR tn, fp: 497, 61 LR fn, tp: 1, 21 LR f1 score: 0.309 LR cohens kappa score: 0.258 LR average precision score: 0.281 -----[ GB ]----- maximum: GB tn, fp: 600, 17 GB fn, tp: 8, 29 GB f1 score: 0.881 GB cohens kappa score: 0.876 average: GB tn, fp: 592.36, 10.04 GB fn, tp: 4.0, 26.2 GB f1 score: 0.789 GB cohens kappa score: 0.778 minimum: GB tn, fp: 586, 3 GB fn, tp: 1, 21 GB f1 score: 0.721 GB cohens kappa score: 0.706 -----[ KNN ]----- maximum: KNN tn, fp: 583, 37 KNN fn, tp: 11, 28 KNN f1 score: 0.676 KNN cohens kappa score: 0.657 average: KNN tn, fp: 573.12, 29.28 KNN fn, tp: 6.0, 24.2 KNN f1 score: 0.580 KNN cohens kappa score: 0.552 minimum: KNN tn, fp: 565, 20 KNN fn, tp: 2, 20 KNN f1 score: 0.471 KNN cohens kappa score: 0.436 -----[ GAN ]----- maximum: GAN tn, fp: 594, 59 GAN fn, tp: 12, 28 GAN f1 score: 0.737 GAN cohens kappa score: 0.724 average: GAN tn, fp: 573.16, 29.24 GAN fn, tp: 6.48, 23.72 GAN f1 score: 0.585 GAN cohens kappa score: 0.558 minimum: GAN tn, fp: 544, 6 GAN fn, tp: 1, 15 GAN f1 score: 0.435 GAN cohens kappa score: 0.391