/////////////////////////////////////////// // Running convGAN-majority-5 on imblearn_protein_homo /////////////////////////////////////////// Load 'data_input/imblearn_protein_homo' from imblearn 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 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28661, 230 GAN fn, tp: 40, 220 GAN f1 score: 0.620 GAN cohens kappa score: 0.615 -> test with 'LR' LR tn, fp: 27665, 1226 LR fn, tp: 16, 244 LR f1 score: 0.282 LR cohens kappa score: 0.271 LR average precision score: 0.856 -> test with 'GB' GB tn, fp: 28409, 482 GB fn, tp: 19, 241 GB f1 score: 0.490 GB cohens kappa score: 0.484 -> test with 'KNN' KNN tn, fp: 28546, 345 KNN fn, tp: 96, 164 KNN f1 score: 0.427 KNN cohens kappa score: 0.420 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28614, 277 GAN fn, tp: 32, 228 GAN f1 score: 0.596 GAN cohens kappa score: 0.591 -> test with 'LR' LR tn, fp: 27811, 1080 LR fn, tp: 15, 245 LR f1 score: 0.309 LR cohens kappa score: 0.299 LR average precision score: 0.886 -> test with 'GB' GB tn, fp: 28409, 482 GB fn, tp: 15, 245 GB f1 score: 0.496 GB cohens kappa score: 0.490 -> test with 'KNN' KNN tn, fp: 28324, 567 KNN fn, tp: 80, 180 KNN f1 score: 0.357 KNN cohens kappa score: 0.349 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28694, 197 GAN fn, tp: 42, 218 GAN f1 score: 0.646 GAN cohens kappa score: 0.642 -> test with 'LR' LR tn, fp: 27637, 1254 LR fn, tp: 7, 253 LR f1 score: 0.286 LR cohens kappa score: 0.275 LR average precision score: 0.886 -> test with 'GB' GB tn, fp: 28369, 522 GB fn, tp: 10, 250 GB f1 score: 0.484 GB cohens kappa score: 0.478 -> test with 'KNN' KNN tn, fp: 28486, 405 KNN fn, tp: 110, 150 KNN f1 score: 0.368 KNN cohens kappa score: 0.360 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28678, 213 GAN fn, tp: 45, 215 GAN f1 score: 0.625 GAN cohens kappa score: 0.621 -> test with 'LR' LR tn, fp: 27749, 1142 LR fn, tp: 14, 246 LR f1 score: 0.299 LR cohens kappa score: 0.288 LR average precision score: 0.857 -> test with 'GB' GB tn, fp: 28437, 454 GB fn, tp: 19, 241 GB f1 score: 0.505 GB cohens kappa score: 0.498 -> test with 'KNN' KNN tn, fp: 28499, 392 KNN fn, tp: 94, 166 KNN f1 score: 0.406 KNN cohens kappa score: 0.399 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114524 synthetic samples -> test with GAN.predict GAN tn, fp: 28596, 295 GAN fn, tp: 49, 207 GAN f1 score: 0.546 GAN cohens kappa score: 0.541 -> test with 'LR' LR tn, fp: 27808, 1083 LR fn, tp: 20, 236 LR f1 score: 0.300 LR cohens kappa score: 0.289 LR average precision score: 0.818 -> test with 'GB' GB tn, fp: 28506, 385 GB fn, tp: 27, 229 GB f1 score: 0.526 GB cohens kappa score: 0.520 -> test with 'KNN' KNN tn, fp: 28504, 387 KNN fn, tp: 113, 143 KNN f1 score: 0.364 KNN cohens kappa score: 0.356 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28700, 191 GAN fn, tp: 47, 213 GAN f1 score: 0.642 GAN cohens kappa score: 0.638 -> test with 'LR' LR tn, fp: 27768, 1123 LR fn, tp: 11, 249 LR f1 score: 0.305 LR cohens kappa score: 0.295 LR average precision score: 0.866 -> test with 'GB' GB tn, fp: 28459, 432 GB fn, tp: 18, 242 GB f1 score: 0.518 GB cohens kappa score: 0.512 -> test with 'KNN' KNN tn, fp: 28547, 344 KNN fn, tp: 100, 160 KNN f1 score: 0.419 KNN cohens kappa score: 0.412 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28593, 298 GAN fn, tp: 33, 227 GAN f1 score: 0.578 GAN cohens kappa score: 0.573 -> test with 'LR' LR tn, fp: 27734, 1157 LR fn, tp: 13, 247 LR f1 score: 0.297 LR cohens kappa score: 0.286 LR average precision score: 0.891 -> test with 'GB' GB tn, fp: 28376, 515 GB fn, tp: 18, 242 GB f1 score: 0.476 GB cohens kappa score: 0.469 -> test with 'KNN' KNN tn, fp: 28525, 366 KNN fn, tp: 93, 167 KNN f1 score: 0.421 KNN cohens kappa score: 0.414 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28408, 483 GAN fn, tp: 47, 213 GAN f1 score: 0.446 GAN cohens kappa score: 0.438 -> test with 'LR' LR tn, fp: 27736, 1155 LR fn, tp: 17, 243 LR f1 score: 0.293 LR cohens kappa score: 0.282 LR average precision score: 0.833 -> test with 'GB' GB tn, fp: 28429, 462 GB fn, tp: 22, 238 GB f1 score: 0.496 GB cohens kappa score: 0.489 -> test with 'KNN' KNN tn, fp: 28497, 394 KNN fn, tp: 99, 161 KNN f1 score: 0.395 KNN cohens kappa score: 0.388 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28715, 176 GAN fn, tp: 41, 219 GAN f1 score: 0.669 GAN cohens kappa score: 0.665 -> test with 'LR' LR tn, fp: 27708, 1183 LR fn, tp: 14, 246 LR f1 score: 0.291 LR cohens kappa score: 0.280 LR average precision score: 0.863 -> test with 'GB' GB tn, fp: 28422, 469 GB fn, tp: 16, 244 GB f1 score: 0.502 GB cohens kappa score: 0.495 -> test with 'KNN' KNN tn, fp: 28502, 389 KNN fn, tp: 90, 170 KNN f1 score: 0.415 KNN cohens kappa score: 0.408 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114524 synthetic samples -> test with GAN.predict GAN tn, fp: 28634, 257 GAN fn, tp: 46, 210 GAN f1 score: 0.581 GAN cohens kappa score: 0.576 -> test with 'LR' LR tn, fp: 27656, 1235 LR fn, tp: 13, 243 LR f1 score: 0.280 LR cohens kappa score: 0.269 LR average precision score: 0.845 -> test with 'GB' GB tn, fp: 28396, 495 GB fn, tp: 19, 237 GB f1 score: 0.480 GB cohens kappa score: 0.473 -> test with 'KNN' KNN tn, fp: 28518, 373 KNN fn, tp: 97, 159 KNN f1 score: 0.404 KNN cohens kappa score: 0.396 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28708, 183 GAN fn, tp: 42, 218 GAN f1 score: 0.660 GAN cohens kappa score: 0.656 -> test with 'LR' LR tn, fp: 27787, 1104 LR fn, tp: 17, 243 LR f1 score: 0.302 LR cohens kappa score: 0.292 LR average precision score: 0.868 -> test with 'GB' GB tn, fp: 28456, 435 GB fn, tp: 20, 240 GB f1 score: 0.513 GB cohens kappa score: 0.507 -> test with 'KNN' KNN tn, fp: 28501, 390 KNN fn, tp: 91, 169 KNN f1 score: 0.413 KNN cohens kappa score: 0.405 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28752, 139 GAN fn, tp: 41, 219 GAN f1 score: 0.709 GAN cohens kappa score: 0.706 -> test with 'LR' LR tn, fp: 27705, 1186 LR fn, tp: 12, 248 LR f1 score: 0.293 LR cohens kappa score: 0.282 LR average precision score: 0.863 -> test with 'GB' GB tn, fp: 28407, 484 GB fn, tp: 18, 242 GB f1 score: 0.491 GB cohens kappa score: 0.484 -> test with 'KNN' KNN tn, fp: 28539, 352 KNN fn, tp: 108, 152 KNN f1 score: 0.398 KNN cohens kappa score: 0.391 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 27444, 1447 GAN fn, tp: 34, 226 GAN f1 score: 0.234 GAN cohens kappa score: 0.222 -> test with 'LR' LR tn, fp: 27732, 1159 LR fn, tp: 17, 243 LR f1 score: 0.292 LR cohens kappa score: 0.282 LR average precision score: 0.830 -> test with 'GB' GB tn, fp: 28434, 457 GB fn, tp: 22, 238 GB f1 score: 0.498 GB cohens kappa score: 0.492 -> test with 'KNN' KNN tn, fp: 28523, 368 KNN fn, tp: 101, 159 KNN f1 score: 0.404 KNN cohens kappa score: 0.397 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28674, 217 GAN fn, tp: 40, 220 GAN f1 score: 0.631 GAN cohens kappa score: 0.627 -> test with 'LR' LR tn, fp: 27674, 1217 LR fn, tp: 12, 248 LR f1 score: 0.288 LR cohens kappa score: 0.277 LR average precision score: 0.865 -> test with 'GB' GB tn, fp: 28435, 456 GB fn, tp: 10, 250 GB f1 score: 0.518 GB cohens kappa score: 0.511 -> test with 'KNN' KNN tn, fp: 28495, 396 KNN fn, tp: 99, 161 KNN f1 score: 0.394 KNN cohens kappa score: 0.387 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114524 synthetic samples -> test with GAN.predict GAN tn, fp: 28663, 228 GAN fn, tp: 36, 220 GAN f1 score: 0.625 GAN cohens kappa score: 0.621 -> test with 'LR' LR tn, fp: 27697, 1194 LR fn, tp: 12, 244 LR f1 score: 0.288 LR cohens kappa score: 0.277 LR average precision score: 0.882 -> test with 'GB' GB tn, fp: 28412, 479 GB fn, tp: 16, 240 GB f1 score: 0.492 GB cohens kappa score: 0.486 -> test with 'KNN' KNN tn, fp: 28504, 387 KNN fn, tp: 92, 164 KNN f1 score: 0.406 KNN cohens kappa score: 0.399 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28749, 142 GAN fn, tp: 39, 221 GAN f1 score: 0.709 GAN cohens kappa score: 0.706 -> test with 'LR' LR tn, fp: 27806, 1085 LR fn, tp: 13, 247 LR f1 score: 0.310 LR cohens kappa score: 0.300 LR average precision score: 0.873 -> test with 'GB' GB tn, fp: 28419, 472 GB fn, tp: 16, 244 GB f1 score: 0.500 GB cohens kappa score: 0.493 -> test with 'KNN' KNN tn, fp: 28105, 786 KNN fn, tp: 90, 170 KNN f1 score: 0.280 KNN cohens kappa score: 0.269 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28647, 244 GAN fn, tp: 45, 215 GAN f1 score: 0.598 GAN cohens kappa score: 0.593 -> test with 'LR' LR tn, fp: 27761, 1130 LR fn, tp: 15, 245 LR f1 score: 0.300 LR cohens kappa score: 0.289 LR average precision score: 0.839 -> test with 'GB' GB tn, fp: 28379, 512 GB fn, tp: 20, 240 GB f1 score: 0.474 GB cohens kappa score: 0.467 -> test with 'KNN' KNN tn, fp: 28527, 364 KNN fn, tp: 105, 155 KNN f1 score: 0.398 KNN cohens kappa score: 0.391 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28737, 154 GAN fn, tp: 45, 215 GAN f1 score: 0.684 GAN cohens kappa score: 0.680 -> test with 'LR' LR tn, fp: 27737, 1154 LR fn, tp: 18, 242 LR f1 score: 0.292 LR cohens kappa score: 0.281 LR average precision score: 0.855 -> test with 'GB' GB tn, fp: 28447, 444 GB fn, tp: 20, 240 GB f1 score: 0.508 GB cohens kappa score: 0.502 -> test with 'KNN' KNN tn, fp: 28499, 392 KNN fn, tp: 89, 171 KNN f1 score: 0.416 KNN cohens kappa score: 0.408 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28675, 216 GAN fn, tp: 45, 215 GAN f1 score: 0.622 GAN cohens kappa score: 0.618 -> test with 'LR' LR tn, fp: 27710, 1181 LR fn, tp: 11, 249 LR f1 score: 0.295 LR cohens kappa score: 0.284 LR average precision score: 0.879 -> test with 'GB' GB tn, fp: 28466, 425 GB fn, tp: 14, 246 GB f1 score: 0.528 GB cohens kappa score: 0.522 -> test with 'KNN' KNN tn, fp: 28523, 368 KNN fn, tp: 93, 167 KNN f1 score: 0.420 KNN cohens kappa score: 0.413 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114524 synthetic samples -> test with GAN.predict GAN tn, fp: 28660, 231 GAN fn, tp: 31, 225 GAN f1 score: 0.632 GAN cohens kappa score: 0.628 -> test with 'LR' LR tn, fp: 27747, 1144 LR fn, tp: 16, 240 LR f1 score: 0.293 LR cohens kappa score: 0.282 LR average precision score: 0.839 -> test with 'GB' GB tn, fp: 28389, 502 GB fn, tp: 15, 241 GB f1 score: 0.482 GB cohens kappa score: 0.476 -> test with 'KNN' KNN tn, fp: 28492, 399 KNN fn, tp: 89, 167 KNN f1 score: 0.406 KNN cohens kappa score: 0.399 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28616, 275 GAN fn, tp: 36, 224 GAN f1 score: 0.590 GAN cohens kappa score: 0.585 -> test with 'LR' LR tn, fp: 27773, 1118 LR fn, tp: 13, 247 LR f1 score: 0.304 LR cohens kappa score: 0.293 LR average precision score: 0.863 -> test with 'GB' GB tn, fp: 28432, 459 GB fn, tp: 17, 243 GB f1 score: 0.505 GB cohens kappa score: 0.499 -> test with 'KNN' KNN tn, fp: 28548, 343 KNN fn, tp: 100, 160 KNN f1 score: 0.419 KNN cohens kappa score: 0.412 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28606, 285 GAN fn, tp: 40, 220 GAN f1 score: 0.575 GAN cohens kappa score: 0.570 -> test with 'LR' LR tn, fp: 27745, 1146 LR fn, tp: 14, 246 LR f1 score: 0.298 LR cohens kappa score: 0.287 LR average precision score: 0.868 -> test with 'GB' GB tn, fp: 28444, 447 GB fn, tp: 18, 242 GB f1 score: 0.510 GB cohens kappa score: 0.504 -> test with 'KNN' KNN tn, fp: 28518, 373 KNN fn, tp: 100, 160 KNN f1 score: 0.404 KNN cohens kappa score: 0.396 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28741, 150 GAN fn, tp: 46, 214 GAN f1 score: 0.686 GAN cohens kappa score: 0.683 -> test with 'LR' LR tn, fp: 27678, 1213 LR fn, tp: 18, 242 LR f1 score: 0.282 LR cohens kappa score: 0.271 LR average precision score: 0.853 -> test with 'GB' GB tn, fp: 28471, 420 GB fn, tp: 17, 243 GB f1 score: 0.527 GB cohens kappa score: 0.520 -> test with 'KNN' KNN tn, fp: 28515, 376 KNN fn, tp: 106, 154 KNN f1 score: 0.390 KNN cohens kappa score: 0.382 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114528 synthetic samples -> test with GAN.predict GAN tn, fp: 28765, 126 GAN fn, tp: 44, 216 GAN f1 score: 0.718 GAN cohens kappa score: 0.715 -> test with 'LR' LR tn, fp: 27728, 1163 LR fn, tp: 10, 250 LR f1 score: 0.299 LR cohens kappa score: 0.288 LR average precision score: 0.863 -> test with 'GB' GB tn, fp: 28408, 483 GB fn, tp: 18, 242 GB f1 score: 0.491 GB cohens kappa score: 0.485 -> test with 'KNN' KNN tn, fp: 28522, 369 KNN fn, tp: 95, 165 KNN f1 score: 0.416 KNN cohens kappa score: 0.409 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 114524 synthetic samples -> test with GAN.predict GAN tn, fp: 28463, 428 GAN fn, tp: 33, 223 GAN f1 score: 0.492 GAN cohens kappa score: 0.485 -> test with 'LR' LR tn, fp: 27734, 1157 LR fn, tp: 15, 241 LR f1 score: 0.291 LR cohens kappa score: 0.281 LR average precision score: 0.849 -> test with 'GB' GB tn, fp: 28391, 500 GB fn, tp: 15, 241 GB f1 score: 0.483 GB cohens kappa score: 0.477 -> test with 'KNN' KNN tn, fp: 28519, 372 KNN fn, tp: 91, 165 KNN f1 score: 0.416 KNN cohens kappa score: 0.409 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 27811, 1254 LR fn, tp: 20, 253 LR f1 score: 0.310 LR cohens kappa score: 0.300 LR average precision score: 0.891 average: LR tn, fp: 27731.44, 1159.56 LR fn, tp: 14.12, 245.08 LR f1 score: 0.295 LR cohens kappa score: 0.284 LR average precision score: 0.860 minimum: LR tn, fp: 27637, 1080 LR fn, tp: 7, 236 LR f1 score: 0.280 LR cohens kappa score: 0.269 LR average precision score: 0.818 -----[ GB ]----- maximum: GB tn, fp: 28506, 522 GB fn, tp: 27, 250 GB f1 score: 0.528 GB cohens kappa score: 0.522 average: GB tn, fp: 28424.08, 466.92 GB fn, tp: 17.56, 241.64 GB f1 score: 0.500 GB cohens kappa score: 0.493 minimum: GB tn, fp: 28369, 385 GB fn, tp: 10, 229 GB f1 score: 0.474 GB cohens kappa score: 0.467 -----[ KNN ]----- maximum: KNN tn, fp: 28548, 786 KNN fn, tp: 113, 180 KNN f1 score: 0.427 KNN cohens kappa score: 0.420 average: KNN tn, fp: 28491.12, 399.88 KNN fn, tp: 96.84, 162.36 KNN f1 score: 0.398 KNN cohens kappa score: 0.391 minimum: KNN tn, fp: 28105, 343 KNN fn, tp: 80, 143 KNN f1 score: 0.280 KNN cohens kappa score: 0.269 -----[ GAN ]----- maximum: GAN tn, fp: 28765, 1447 GAN fn, tp: 49, 228 GAN f1 score: 0.718 GAN cohens kappa score: 0.715 average: GAN tn, fp: 28607.72, 283.28 GAN fn, tp: 40.76, 218.44 GAN f1 score: 0.605 GAN cohens kappa score: 0.600 minimum: GAN tn, fp: 27444, 126 GAN fn, tp: 31, 207 GAN f1 score: 0.234 GAN cohens kappa score: 0.222