/////////////////////////////////////////// // Running convGAN on imblearn_webpage /////////////////////////////////////////// Load 'data_input/imblearn_webpage' from imblearn non empty cut in data_input/imblearn_webpage! (76 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 26255 synthetic samples -> test with 'LR' LR tn, fp: 6350, 410 LR fn, tp: 24, 173 LR f1 score: 0.444 LR cohens kappa score: 0.419 LR average precision score: 0.765 -> test with 'GB' GB tn, fp: 6399, 361 GB fn, tp: 90, 107 GB f1 score: 0.322 GB cohens kappa score: 0.294 -> test with 'KNN' KNN tn, fp: 6265, 495 KNN fn, tp: 15, 182 KNN f1 score: 0.416 KNN cohens kappa score: 0.390 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6402, 358 LR fn, tp: 21, 176 LR f1 score: 0.482 LR cohens kappa score: 0.459 LR average precision score: 0.793 -> test with 'GB' GB tn, fp: 6324, 436 GB fn, tp: 87, 110 GB f1 score: 0.296 GB cohens kappa score: 0.266 -> test with 'KNN' KNN tn, fp: 6370, 390 KNN fn, tp: 33, 164 KNN f1 score: 0.437 KNN cohens kappa score: 0.412 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6386, 374 LR fn, tp: 15, 182 LR f1 score: 0.483 LR cohens kappa score: 0.461 LR average precision score: 0.841 -> test with 'GB' GB tn, fp: 6359, 401 GB fn, tp: 89, 108 GB f1 score: 0.306 GB cohens kappa score: 0.276 -> test with 'KNN' KNN tn, fp: 6226, 534 KNN fn, tp: 26, 171 KNN f1 score: 0.379 KNN cohens kappa score: 0.350 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6372, 388 LR fn, tp: 17, 180 LR f1 score: 0.471 LR cohens kappa score: 0.447 LR average precision score: 0.754 -> test with 'GB' GB tn, fp: 6348, 412 GB fn, tp: 94, 103 GB f1 score: 0.289 GB cohens kappa score: 0.259 -> test with 'KNN' KNN tn, fp: 6270, 490 KNN fn, tp: 27, 170 KNN f1 score: 0.397 KNN cohens kappa score: 0.369 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26252 synthetic samples -> test with 'LR' LR tn, fp: 6398, 361 LR fn, tp: 33, 160 LR f1 score: 0.448 LR cohens kappa score: 0.425 LR average precision score: 0.741 -> test with 'GB' GB tn, fp: 6373, 386 GB fn, tp: 92, 101 GB f1 score: 0.297 GB cohens kappa score: 0.268 -> test with 'KNN' KNN tn, fp: 6246, 513 KNN fn, tp: 30, 163 KNN f1 score: 0.375 KNN cohens kappa score: 0.347 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6374, 386 LR fn, tp: 23, 174 LR f1 score: 0.460 LR cohens kappa score: 0.436 LR average precision score: 0.793 -> test with 'GB' GB tn, fp: 6353, 407 GB fn, tp: 87, 110 GB f1 score: 0.308 GB cohens kappa score: 0.279 -> test with 'KNN' KNN tn, fp: 6243, 517 KNN fn, tp: 30, 167 KNN f1 score: 0.379 KNN cohens kappa score: 0.351 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6390, 370 LR fn, tp: 22, 175 LR f1 score: 0.472 LR cohens kappa score: 0.449 LR average precision score: 0.796 -> test with 'GB' GB tn, fp: 6383, 377 GB fn, tp: 93, 104 GB f1 score: 0.307 GB cohens kappa score: 0.278 -> test with 'KNN' KNN tn, fp: 6236, 524 KNN fn, tp: 27, 170 KNN f1 score: 0.382 KNN cohens kappa score: 0.353 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6400, 360 LR fn, tp: 28, 169 LR f1 score: 0.466 LR cohens kappa score: 0.443 LR average precision score: 0.758 -> test with 'GB' GB tn, fp: 6320, 440 GB fn, tp: 95, 102 GB f1 score: 0.276 GB cohens kappa score: 0.245 -> test with 'KNN' KNN tn, fp: 6260, 500 KNN fn, tp: 27, 170 KNN f1 score: 0.392 KNN cohens kappa score: 0.364 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6330, 430 LR fn, tp: 20, 177 LR f1 score: 0.440 LR cohens kappa score: 0.415 LR average precision score: 0.755 -> test with 'GB' GB tn, fp: 6422, 338 GB fn, tp: 91, 106 GB f1 score: 0.331 GB cohens kappa score: 0.303 -> test with 'KNN' KNN tn, fp: 6356, 404 KNN fn, tp: 20, 177 KNN f1 score: 0.455 KNN cohens kappa score: 0.431 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26252 synthetic samples -> test with 'LR' LR tn, fp: 6372, 387 LR fn, tp: 20, 173 LR f1 score: 0.459 LR cohens kappa score: 0.436 LR average precision score: 0.791 -> test with 'GB' GB tn, fp: 6341, 418 GB fn, tp: 99, 94 GB f1 score: 0.267 GB cohens kappa score: 0.236 -> test with 'KNN' KNN tn, fp: 6292, 467 KNN fn, tp: 32, 161 KNN f1 score: 0.392 KNN cohens kappa score: 0.365 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6360, 400 LR fn, tp: 30, 167 LR f1 score: 0.437 LR cohens kappa score: 0.412 LR average precision score: 0.733 -> test with 'GB' GB tn, fp: 6361, 399 GB fn, tp: 93, 104 GB f1 score: 0.297 GB cohens kappa score: 0.267 -> test with 'KNN' KNN tn, fp: 6339, 421 KNN fn, tp: 29, 168 KNN f1 score: 0.427 KNN cohens kappa score: 0.402 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6410, 350 LR fn, tp: 18, 179 LR f1 score: 0.493 LR cohens kappa score: 0.471 LR average precision score: 0.798 -> test with 'GB' GB tn, fp: 6342, 418 GB fn, tp: 94, 103 GB f1 score: 0.287 GB cohens kappa score: 0.256 -> test with 'KNN' KNN tn, fp: 6234, 526 KNN fn, tp: 23, 174 KNN f1 score: 0.388 KNN cohens kappa score: 0.360 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6400, 360 LR fn, tp: 32, 165 LR f1 score: 0.457 LR cohens kappa score: 0.434 LR average precision score: 0.706 -> test with 'GB' GB tn, fp: 6390, 370 GB fn, tp: 100, 97 GB f1 score: 0.292 GB cohens kappa score: 0.263 -> test with 'KNN' KNN tn, fp: 6282, 478 KNN fn, tp: 40, 157 KNN f1 score: 0.377 KNN cohens kappa score: 0.349 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6360, 400 LR fn, tp: 17, 180 LR f1 score: 0.463 LR cohens kappa score: 0.440 LR average precision score: 0.808 -> test with 'GB' GB tn, fp: 6320, 440 GB fn, tp: 86, 111 GB f1 score: 0.297 GB cohens kappa score: 0.266 -> test with 'KNN' KNN tn, fp: 6253, 507 KNN fn, tp: 21, 176 KNN f1 score: 0.400 KNN cohens kappa score: 0.372 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26252 synthetic samples -> test with 'LR' LR tn, fp: 6372, 387 LR fn, tp: 17, 176 LR f1 score: 0.466 LR cohens kappa score: 0.443 LR average precision score: 0.773 -> test with 'GB' GB tn, fp: 6391, 368 GB fn, tp: 91, 102 GB f1 score: 0.308 GB cohens kappa score: 0.279 -> test with 'KNN' KNN tn, fp: 6240, 519 KNN fn, tp: 16, 177 KNN f1 score: 0.398 KNN cohens kappa score: 0.371 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6398, 362 LR fn, tp: 27, 170 LR f1 score: 0.466 LR cohens kappa score: 0.443 LR average precision score: 0.745 -> test with 'GB' GB tn, fp: 6387, 373 GB fn, tp: 100, 97 GB f1 score: 0.291 GB cohens kappa score: 0.261 -> test with 'KNN' KNN tn, fp: 6270, 490 KNN fn, tp: 37, 160 KNN f1 score: 0.378 KNN cohens kappa score: 0.350 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6351, 409 LR fn, tp: 21, 176 LR f1 score: 0.450 LR cohens kappa score: 0.426 LR average precision score: 0.756 -> test with 'GB' GB tn, fp: 6331, 429 GB fn, tp: 101, 96 GB f1 score: 0.266 GB cohens kappa score: 0.234 -> test with 'KNN' KNN tn, fp: 6238, 522 KNN fn, tp: 26, 171 KNN f1 score: 0.384 KNN cohens kappa score: 0.356 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6377, 383 LR fn, tp: 17, 180 LR f1 score: 0.474 LR cohens kappa score: 0.451 LR average precision score: 0.807 -> test with 'GB' GB tn, fp: 6421, 339 GB fn, tp: 79, 118 GB f1 score: 0.361 GB cohens kappa score: 0.335 -> test with 'KNN' KNN tn, fp: 6346, 414 KNN fn, tp: 19, 178 KNN f1 score: 0.451 KNN cohens kappa score: 0.427 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6377, 383 LR fn, tp: 21, 176 LR f1 score: 0.466 LR cohens kappa score: 0.442 LR average precision score: 0.752 -> test with 'GB' GB tn, fp: 6317, 443 GB fn, tp: 92, 105 GB f1 score: 0.282 GB cohens kappa score: 0.251 -> test with 'KNN' KNN tn, fp: 6230, 530 KNN fn, tp: 29, 168 KNN f1 score: 0.375 KNN cohens kappa score: 0.347 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26252 synthetic samples -> test with 'LR' LR tn, fp: 6354, 405 LR fn, tp: 20, 173 LR f1 score: 0.449 LR cohens kappa score: 0.425 LR average precision score: 0.791 -> test with 'GB' GB tn, fp: 6351, 408 GB fn, tp: 87, 106 GB f1 score: 0.300 GB cohens kappa score: 0.270 -> test with 'KNN' KNN tn, fp: 6250, 509 KNN fn, tp: 21, 172 KNN f1 score: 0.394 KNN cohens kappa score: 0.366 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6411, 349 LR fn, tp: 22, 175 LR f1 score: 0.485 LR cohens kappa score: 0.463 LR average precision score: 0.769 -> test with 'GB' GB tn, fp: 6387, 373 GB fn, tp: 91, 106 GB f1 score: 0.314 GB cohens kappa score: 0.285 -> test with 'KNN' KNN tn, fp: 6320, 440 KNN fn, tp: 23, 174 KNN f1 score: 0.429 KNN cohens kappa score: 0.404 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6407, 353 LR fn, tp: 26, 171 LR f1 score: 0.474 LR cohens kappa score: 0.452 LR average precision score: 0.741 -> test with 'GB' GB tn, fp: 6355, 405 GB fn, tp: 94, 103 GB f1 score: 0.292 GB cohens kappa score: 0.262 -> test with 'KNN' KNN tn, fp: 6243, 517 KNN fn, tp: 31, 166 KNN f1 score: 0.377 KNN cohens kappa score: 0.349 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6311, 449 LR fn, tp: 25, 172 LR f1 score: 0.421 LR cohens kappa score: 0.395 LR average precision score: 0.746 -> test with 'GB' GB tn, fp: 6335, 425 GB fn, tp: 85, 112 GB f1 score: 0.305 GB cohens kappa score: 0.275 -> test with 'KNN' KNN tn, fp: 6301, 459 KNN fn, tp: 21, 176 KNN f1 score: 0.423 KNN cohens kappa score: 0.397 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26255 synthetic samples -> test with 'LR' LR tn, fp: 6375, 385 LR fn, tp: 17, 180 LR f1 score: 0.472 LR cohens kappa score: 0.449 LR average precision score: 0.824 -> test with 'GB' GB tn, fp: 6353, 407 GB fn, tp: 92, 105 GB f1 score: 0.296 GB cohens kappa score: 0.266 -> test with 'KNN' KNN tn, fp: 6276, 484 KNN fn, tp: 30, 167 KNN f1 score: 0.394 KNN cohens kappa score: 0.366 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 26252 synthetic samples -> test with 'LR' LR tn, fp: 6388, 371 LR fn, tp: 25, 168 LR f1 score: 0.459 LR cohens kappa score: 0.436 LR average precision score: 0.755 -> test with 'GB' GB tn, fp: 6372, 387 GB fn, tp: 100, 93 GB f1 score: 0.276 GB cohens kappa score: 0.247 -> test with 'KNN' KNN tn, fp: 6213, 546 KNN fn, tp: 31, 162 KNN f1 score: 0.360 KNN cohens kappa score: 0.330 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 6411, 449 LR fn, tp: 33, 182 LR f1 score: 0.493 LR cohens kappa score: 0.471 LR average precision score: 0.841 average: LR tn, fp: 6377.0, 382.8 LR fn, tp: 22.32, 173.88 LR f1 score: 0.462 LR cohens kappa score: 0.439 LR average precision score: 0.772 minimum: LR tn, fp: 6311, 349 LR fn, tp: 15, 160 LR f1 score: 0.421 LR cohens kappa score: 0.395 LR average precision score: 0.706 -----[ GB ]----- maximum: GB tn, fp: 6422, 443 GB fn, tp: 101, 118 GB f1 score: 0.361 GB cohens kappa score: 0.335 average: GB tn, fp: 6361.4, 398.4 GB fn, tp: 92.08, 104.12 GB f1 score: 0.298 GB cohens kappa score: 0.269 minimum: GB tn, fp: 6317, 338 GB fn, tp: 79, 93 GB f1 score: 0.266 GB cohens kappa score: 0.234 -----[ KNN ]----- maximum: KNN tn, fp: 6370, 546 KNN fn, tp: 40, 182 KNN f1 score: 0.455 KNN cohens kappa score: 0.431 average: KNN tn, fp: 6271.96, 487.84 KNN fn, tp: 26.56, 169.64 KNN f1 score: 0.398 KNN cohens kappa score: 0.371 minimum: KNN tn, fp: 6213, 390 KNN fn, tp: 15, 157 KNN f1 score: 0.360 KNN cohens kappa score: 0.330