/////////////////////////////////////////// // Running convGAN-majority-5 on folding_yeast6 /////////////////////////////////////////// Load 'data_input/folding_yeast6' 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 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 269, 21 GAN fn, tp: 1, 6 GAN f1 score: 0.353 GAN cohens kappa score: 0.328 -> test with 'LR' LR tn, fp: 262, 28 LR fn, tp: 0, 7 LR f1 score: 0.333 LR cohens kappa score: 0.306 LR average precision score: 0.691 -> test with 'GB' GB tn, fp: 287, 3 GB fn, tp: 4, 3 GB f1 score: 0.462 GB cohens kappa score: 0.450 -> test with 'KNN' KNN tn, fp: 265, 25 KNN fn, tp: 1, 6 KNN f1 score: 0.316 KNN cohens kappa score: 0.288 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 265, 25 GAN fn, tp: 2, 5 GAN f1 score: 0.270 GAN cohens kappa score: 0.241 -> test with 'LR' LR tn, fp: 264, 26 LR fn, tp: 2, 5 LR f1 score: 0.263 LR cohens kappa score: 0.234 LR average precision score: 0.432 -> test with 'GB' GB tn, fp: 286, 4 GB fn, tp: 4, 3 GB f1 score: 0.429 GB cohens kappa score: 0.415 -> test with 'KNN' KNN tn, fp: 266, 24 KNN fn, tp: 3, 4 KNN f1 score: 0.229 KNN cohens kappa score: 0.198 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 272, 18 GAN fn, tp: 1, 6 GAN f1 score: 0.387 GAN cohens kappa score: 0.364 -> test with 'LR' LR tn, fp: 261, 29 LR fn, tp: 1, 6 LR f1 score: 0.286 LR cohens kappa score: 0.257 LR average precision score: 0.290 -> test with 'GB' GB tn, fp: 290, 0 GB fn, tp: 5, 2 GB f1 score: 0.444 GB cohens kappa score: 0.439 -> test with 'KNN' KNN tn, fp: 271, 19 KNN fn, tp: 1, 6 KNN f1 score: 0.375 KNN cohens kappa score: 0.351 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 277, 13 GAN fn, tp: 2, 5 GAN f1 score: 0.400 GAN cohens kappa score: 0.379 -> test with 'LR' LR tn, fp: 269, 21 LR fn, tp: 2, 5 LR f1 score: 0.303 LR cohens kappa score: 0.276 LR average precision score: 0.552 -> test with 'GB' GB tn, fp: 288, 2 GB fn, tp: 4, 3 GB f1 score: 0.500 GB cohens kappa score: 0.490 -> test with 'KNN' KNN tn, fp: 274, 16 KNN fn, tp: 1, 6 KNN f1 score: 0.414 KNN cohens kappa score: 0.392 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1132 synthetic samples -> test with GAN.predict GAN tn, fp: 250, 39 GAN fn, tp: 0, 7 GAN f1 score: 0.264 GAN cohens kappa score: 0.233 -> test with 'LR' LR tn, fp: 245, 44 LR fn, tp: 0, 7 LR f1 score: 0.241 LR cohens kappa score: 0.208 LR average precision score: 0.554 -> test with 'GB' GB tn, fp: 289, 0 GB fn, tp: 3, 4 GB f1 score: 0.727 GB cohens kappa score: 0.722 -> test with 'KNN' KNN tn, fp: 261, 28 KNN fn, tp: 1, 6 KNN f1 score: 0.293 KNN cohens kappa score: 0.264 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 268, 22 GAN fn, tp: 1, 6 GAN f1 score: 0.343 GAN cohens kappa score: 0.317 -> test with 'LR' LR tn, fp: 264, 26 LR fn, tp: 0, 7 LR f1 score: 0.350 LR cohens kappa score: 0.324 LR average precision score: 0.678 -> test with 'GB' GB tn, fp: 286, 4 GB fn, tp: 3, 4 GB f1 score: 0.533 GB cohens kappa score: 0.521 -> test with 'KNN' KNN tn, fp: 273, 17 KNN fn, tp: 1, 6 KNN f1 score: 0.400 KNN cohens kappa score: 0.378 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 257, 33 GAN fn, tp: 0, 7 GAN f1 score: 0.298 GAN cohens kappa score: 0.269 -> test with 'LR' LR tn, fp: 251, 39 LR fn, tp: 0, 7 LR f1 score: 0.264 LR cohens kappa score: 0.233 LR average precision score: 0.245 -> test with 'GB' GB tn, fp: 288, 2 GB fn, tp: 4, 3 GB f1 score: 0.500 GB cohens kappa score: 0.490 -> test with 'KNN' KNN tn, fp: 262, 28 KNN fn, tp: 0, 7 KNN f1 score: 0.333 KNN cohens kappa score: 0.306 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 265, 25 GAN fn, tp: 1, 6 GAN f1 score: 0.316 GAN cohens kappa score: 0.288 -> test with 'LR' LR tn, fp: 261, 29 LR fn, tp: 1, 6 LR f1 score: 0.286 LR cohens kappa score: 0.257 LR average precision score: 0.511 -> test with 'GB' GB tn, fp: 287, 3 GB fn, tp: 4, 3 GB f1 score: 0.462 GB cohens kappa score: 0.450 -> test with 'KNN' KNN tn, fp: 266, 24 KNN fn, tp: 2, 5 KNN f1 score: 0.278 KNN cohens kappa score: 0.249 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 269, 21 GAN fn, tp: 2, 5 GAN f1 score: 0.303 GAN cohens kappa score: 0.276 -> test with 'LR' LR tn, fp: 260, 30 LR fn, tp: 2, 5 LR f1 score: 0.238 LR cohens kappa score: 0.207 LR average precision score: 0.557 -> test with 'GB' GB tn, fp: 287, 3 GB fn, tp: 5, 2 GB f1 score: 0.333 GB cohens kappa score: 0.320 -> test with 'KNN' KNN tn, fp: 269, 21 KNN fn, tp: 1, 6 KNN f1 score: 0.353 KNN cohens kappa score: 0.328 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1132 synthetic samples -> test with GAN.predict GAN tn, fp: 272, 17 GAN fn, tp: 1, 6 GAN f1 score: 0.400 GAN cohens kappa score: 0.377 -> test with 'LR' LR tn, fp: 269, 20 LR fn, tp: 1, 6 LR f1 score: 0.364 LR cohens kappa score: 0.339 LR average precision score: 0.527 -> test with 'GB' GB tn, fp: 289, 0 GB fn, tp: 6, 1 GB f1 score: 0.250 GB cohens kappa score: 0.246 -> test with 'KNN' KNN tn, fp: 274, 15 KNN fn, tp: 2, 5 KNN f1 score: 0.370 KNN cohens kappa score: 0.348 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 265, 25 GAN fn, tp: 1, 6 GAN f1 score: 0.316 GAN cohens kappa score: 0.288 -> test with 'LR' LR tn, fp: 261, 29 LR fn, tp: 1, 6 LR f1 score: 0.286 LR cohens kappa score: 0.257 LR average precision score: 0.636 -> test with 'GB' GB tn, fp: 289, 1 GB fn, tp: 3, 4 GB f1 score: 0.667 GB cohens kappa score: 0.660 -> test with 'KNN' KNN tn, fp: 267, 23 KNN fn, tp: 1, 6 KNN f1 score: 0.333 KNN cohens kappa score: 0.307 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 262, 28 GAN fn, tp: 0, 7 GAN f1 score: 0.333 GAN cohens kappa score: 0.306 -> test with 'LR' LR tn, fp: 247, 43 LR fn, tp: 0, 7 LR f1 score: 0.246 LR cohens kappa score: 0.213 LR average precision score: 0.813 -> test with 'GB' GB tn, fp: 289, 1 GB fn, tp: 4, 3 GB f1 score: 0.545 GB cohens kappa score: 0.538 -> test with 'KNN' KNN tn, fp: 260, 30 KNN fn, tp: 0, 7 KNN f1 score: 0.318 KNN cohens kappa score: 0.290 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 275, 15 GAN fn, tp: 2, 5 GAN f1 score: 0.370 GAN cohens kappa score: 0.348 -> test with 'LR' LR tn, fp: 267, 23 LR fn, tp: 2, 5 LR f1 score: 0.286 LR cohens kappa score: 0.258 LR average precision score: 0.437 -> test with 'GB' GB tn, fp: 288, 2 GB fn, tp: 4, 3 GB f1 score: 0.500 GB cohens kappa score: 0.490 -> test with 'KNN' KNN tn, fp: 275, 15 KNN fn, tp: 3, 4 KNN f1 score: 0.308 KNN cohens kappa score: 0.283 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 262, 28 GAN fn, tp: 1, 6 GAN f1 score: 0.293 GAN cohens kappa score: 0.264 -> test with 'LR' LR tn, fp: 256, 34 LR fn, tp: 0, 7 LR f1 score: 0.292 LR cohens kappa score: 0.262 LR average precision score: 0.383 -> test with 'GB' GB tn, fp: 285, 5 GB fn, tp: 3, 4 GB f1 score: 0.500 GB cohens kappa score: 0.486 -> test with 'KNN' KNN tn, fp: 262, 28 KNN fn, tp: 1, 6 KNN f1 score: 0.293 KNN cohens kappa score: 0.264 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1132 synthetic samples -> test with GAN.predict GAN tn, fp: 272, 17 GAN fn, tp: 1, 6 GAN f1 score: 0.400 GAN cohens kappa score: 0.377 -> test with 'LR' LR tn, fp: 268, 21 LR fn, tp: 1, 6 LR f1 score: 0.353 LR cohens kappa score: 0.328 LR average precision score: 0.405 -> test with 'GB' GB tn, fp: 288, 1 GB fn, tp: 7, 0 GB f1 score: 0.000 GB cohens kappa score: -0.006 -> test with 'KNN' KNN tn, fp: 270, 19 KNN fn, tp: 1, 6 KNN f1 score: 0.375 KNN cohens kappa score: 0.351 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 277, 13 GAN fn, tp: 1, 6 GAN f1 score: 0.462 GAN cohens kappa score: 0.442 -> test with 'LR' LR tn, fp: 272, 18 LR fn, tp: 1, 6 LR f1 score: 0.387 LR cohens kappa score: 0.364 LR average precision score: 0.704 -> test with 'GB' GB tn, fp: 289, 1 GB fn, tp: 3, 4 GB f1 score: 0.667 GB cohens kappa score: 0.660 -> test with 'KNN' KNN tn, fp: 270, 20 KNN fn, tp: 1, 6 KNN f1 score: 0.364 KNN cohens kappa score: 0.339 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 270, 20 GAN fn, tp: 0, 7 GAN f1 score: 0.412 GAN cohens kappa score: 0.389 -> test with 'LR' LR tn, fp: 259, 31 LR fn, tp: 0, 7 LR f1 score: 0.311 LR cohens kappa score: 0.283 LR average precision score: 0.246 -> test with 'GB' GB tn, fp: 286, 4 GB fn, tp: 4, 3 GB f1 score: 0.429 GB cohens kappa score: 0.415 -> test with 'KNN' KNN tn, fp: 275, 15 KNN fn, tp: 0, 7 KNN f1 score: 0.483 KNN cohens kappa score: 0.464 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 254, 36 GAN fn, tp: 1, 6 GAN f1 score: 0.245 GAN cohens kappa score: 0.213 -> test with 'LR' LR tn, fp: 250, 40 LR fn, tp: 1, 6 LR f1 score: 0.226 LR cohens kappa score: 0.193 LR average precision score: 0.550 -> test with 'GB' GB tn, fp: 287, 3 GB fn, tp: 2, 5 GB f1 score: 0.667 GB cohens kappa score: 0.658 -> test with 'KNN' KNN tn, fp: 252, 38 KNN fn, tp: 0, 7 KNN f1 score: 0.269 KNN cohens kappa score: 0.238 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 272, 18 GAN fn, tp: 1, 6 GAN f1 score: 0.387 GAN cohens kappa score: 0.364 -> test with 'LR' LR tn, fp: 268, 22 LR fn, tp: 1, 6 LR f1 score: 0.343 LR cohens kappa score: 0.317 LR average precision score: 0.652 -> test with 'GB' GB tn, fp: 287, 3 GB fn, tp: 4, 3 GB f1 score: 0.462 GB cohens kappa score: 0.450 -> test with 'KNN' KNN tn, fp: 277, 13 KNN fn, tp: 2, 5 KNN f1 score: 0.400 KNN cohens kappa score: 0.379 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1132 synthetic samples -> test with GAN.predict GAN tn, fp: 274, 15 GAN fn, tp: 2, 5 GAN f1 score: 0.370 GAN cohens kappa score: 0.348 -> test with 'LR' LR tn, fp: 267, 22 LR fn, tp: 2, 5 LR f1 score: 0.294 LR cohens kappa score: 0.267 LR average precision score: 0.679 -> test with 'GB' GB tn, fp: 288, 1 GB fn, tp: 4, 3 GB f1 score: 0.545 GB cohens kappa score: 0.537 -> test with 'KNN' KNN tn, fp: 275, 14 KNN fn, tp: 2, 5 KNN f1 score: 0.385 KNN cohens kappa score: 0.363 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 263, 27 GAN fn, tp: 0, 7 GAN f1 score: 0.341 GAN cohens kappa score: 0.315 -> test with 'LR' LR tn, fp: 249, 41 LR fn, tp: 0, 7 LR f1 score: 0.255 LR cohens kappa score: 0.223 LR average precision score: 0.500 -> test with 'GB' GB tn, fp: 288, 2 GB fn, tp: 3, 4 GB f1 score: 0.615 GB cohens kappa score: 0.607 -> test with 'KNN' KNN tn, fp: 259, 31 KNN fn, tp: 2, 5 KNN f1 score: 0.233 KNN cohens kappa score: 0.201 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 275, 15 GAN fn, tp: 3, 4 GAN f1 score: 0.308 GAN cohens kappa score: 0.283 -> test with 'LR' LR tn, fp: 266, 24 LR fn, tp: 3, 4 LR f1 score: 0.229 LR cohens kappa score: 0.198 LR average precision score: 0.217 -> test with 'GB' GB tn, fp: 289, 1 GB fn, tp: 4, 3 GB f1 score: 0.545 GB cohens kappa score: 0.538 -> test with 'KNN' KNN tn, fp: 274, 16 KNN fn, tp: 3, 4 KNN f1 score: 0.296 KNN cohens kappa score: 0.271 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 261, 29 GAN fn, tp: 0, 7 GAN f1 score: 0.326 GAN cohens kappa score: 0.298 -> test with 'LR' LR tn, fp: 262, 28 LR fn, tp: 0, 7 LR f1 score: 0.333 LR cohens kappa score: 0.306 LR average precision score: 0.713 -> test with 'GB' GB tn, fp: 288, 2 GB fn, tp: 1, 6 GB f1 score: 0.800 GB cohens kappa score: 0.795 -> test with 'KNN' KNN tn, fp: 267, 23 KNN fn, tp: 0, 7 KNN f1 score: 0.378 KNN cohens kappa score: 0.354 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1131 synthetic samples -> test with GAN.predict GAN tn, fp: 269, 21 GAN fn, tp: 1, 6 GAN f1 score: 0.353 GAN cohens kappa score: 0.328 -> test with 'LR' LR tn, fp: 256, 34 LR fn, tp: 0, 7 LR f1 score: 0.292 LR cohens kappa score: 0.262 LR average precision score: 0.285 -> test with 'GB' GB tn, fp: 289, 1 GB fn, tp: 5, 2 GB f1 score: 0.400 GB cohens kappa score: 0.391 -> test with 'KNN' KNN tn, fp: 271, 19 KNN fn, tp: 2, 5 KNN f1 score: 0.323 KNN cohens kappa score: 0.297 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1132 synthetic samples -> test with GAN.predict GAN tn, fp: 272, 17 GAN fn, tp: 2, 5 GAN f1 score: 0.345 GAN cohens kappa score: 0.320 -> test with 'LR' LR tn, fp: 272, 17 LR fn, tp: 2, 5 LR f1 score: 0.345 LR cohens kappa score: 0.320 LR average precision score: 0.418 -> test with 'GB' GB tn, fp: 286, 3 GB fn, tp: 5, 2 GB f1 score: 0.333 GB cohens kappa score: 0.320 -> test with 'KNN' KNN tn, fp: 280, 9 KNN fn, tp: 2, 5 KNN f1 score: 0.476 KNN cohens kappa score: 0.459 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 272, 44 LR fn, tp: 3, 7 LR f1 score: 0.387 LR cohens kappa score: 0.364 LR average precision score: 0.813 average: LR tn, fp: 261.04, 28.76 LR fn, tp: 0.92, 6.08 LR f1 score: 0.296 LR cohens kappa score: 0.268 LR average precision score: 0.507 minimum: LR tn, fp: 245, 17 LR fn, tp: 0, 4 LR f1 score: 0.226 LR cohens kappa score: 0.193 LR average precision score: 0.217 -----[ GB ]----- maximum: GB tn, fp: 290, 5 GB fn, tp: 7, 6 GB f1 score: 0.800 GB cohens kappa score: 0.795 average: GB tn, fp: 287.72, 2.08 GB fn, tp: 3.92, 3.08 GB f1 score: 0.493 GB cohens kappa score: 0.483 minimum: GB tn, fp: 285, 0 GB fn, tp: 1, 0 GB f1 score: 0.000 GB cohens kappa score: -0.006 -----[ KNN ]----- maximum: KNN tn, fp: 280, 38 KNN fn, tp: 3, 7 KNN f1 score: 0.483 KNN cohens kappa score: 0.464 average: KNN tn, fp: 268.6, 21.2 KNN fn, tp: 1.32, 5.68 KNN f1 score: 0.344 KNN cohens kappa score: 0.318 minimum: KNN tn, fp: 252, 9 KNN fn, tp: 0, 4 KNN f1 score: 0.229 KNN cohens kappa score: 0.198 -----[ GAN ]----- maximum: GAN tn, fp: 277, 39 GAN fn, tp: 3, 7 GAN f1 score: 0.462 GAN cohens kappa score: 0.442 average: GAN tn, fp: 267.48, 22.32 GAN fn, tp: 1.08, 5.92 GAN f1 score: 0.344 GAN cohens kappa score: 0.318 minimum: GAN tn, fp: 250, 13 GAN fn, tp: 0, 4 GAN f1 score: 0.245 GAN cohens kappa score: 0.213