/////////////////////////////////////////// // Running convGAN on kaggle_creditcard /////////////////////////////////////////// Load 'data_input/kaggle_creditcard' 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 227059 synthetic samples -> test with 'LR' LR tn, fp: 54819, 2044 LR fn, tp: 16, 83 LR f1 score: 0.075 LR cohens kappa score: 0.071 LR average precision score: 0.579 -> test with 'GB' GB tn, fp: 56577, 286 GB fn, tp: 20, 79 GB f1 score: 0.341 GB cohens kappa score: 0.339 -> test with 'KNN' KNN tn, fp: 56521, 342 KNN fn, tp: 76, 23 KNN f1 score: 0.099 KNN cohens kappa score: 0.097 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54137, 2726 LR fn, tp: 10, 89 LR f1 score: 0.061 LR cohens kappa score: 0.058 LR average precision score: 0.703 -> test with 'GB' GB tn, fp: 56550, 313 GB fn, tp: 10, 89 GB f1 score: 0.355 GB cohens kappa score: 0.353 -> test with 'KNN' KNN tn, fp: 56389, 474 KNN fn, tp: 76, 23 KNN f1 score: 0.077 KNN cohens kappa score: 0.074 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 55190, 1673 LR fn, tp: 10, 89 LR f1 score: 0.096 LR cohens kappa score: 0.093 LR average precision score: 0.682 -> test with 'GB' GB tn, fp: 56545, 318 GB fn, tp: 13, 86 GB f1 score: 0.342 GB cohens kappa score: 0.340 -> test with 'KNN' KNN tn, fp: 56445, 418 KNN fn, tp: 77, 22 KNN f1 score: 0.082 KNN cohens kappa score: 0.079 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 55233, 1630 LR fn, tp: 7, 92 LR f1 score: 0.101 LR cohens kappa score: 0.098 LR average precision score: 0.758 -> test with 'GB' GB tn, fp: 56495, 368 GB fn, tp: 8, 91 GB f1 score: 0.326 GB cohens kappa score: 0.324 -> test with 'KNN' KNN tn, fp: 56516, 347 KNN fn, tp: 67, 32 KNN f1 score: 0.134 KNN cohens kappa score: 0.131 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227056 synthetic samples -> test with 'LR' LR tn, fp: 54918, 1945 LR fn, tp: 9, 87 LR f1 score: 0.082 LR cohens kappa score: 0.079 LR average precision score: 0.801 -> test with 'GB' GB tn, fp: 56450, 413 GB fn, tp: 9, 87 GB f1 score: 0.292 GB cohens kappa score: 0.290 -> test with 'KNN' KNN tn, fp: 56499, 364 KNN fn, tp: 70, 26 KNN f1 score: 0.107 KNN cohens kappa score: 0.105 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 53350, 3513 LR fn, tp: 12, 87 LR f1 score: 0.047 LR cohens kappa score: 0.044 LR average precision score: 0.699 -> test with 'GB' GB tn, fp: 56557, 306 GB fn, tp: 10, 89 GB f1 score: 0.360 GB cohens kappa score: 0.359 -> test with 'KNN' KNN tn, fp: 56382, 481 KNN fn, tp: 73, 26 KNN f1 score: 0.086 KNN cohens kappa score: 0.083 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54196, 2667 LR fn, tp: 10, 89 LR f1 score: 0.062 LR cohens kappa score: 0.059 LR average precision score: 0.652 -> test with 'GB' GB tn, fp: 56434, 429 GB fn, tp: 10, 89 GB f1 score: 0.288 GB cohens kappa score: 0.286 -> test with 'KNN' KNN tn, fp: 56532, 331 KNN fn, tp: 72, 27 KNN f1 score: 0.118 KNN cohens kappa score: 0.116 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54521, 2342 LR fn, tp: 9, 90 LR f1 score: 0.071 LR cohens kappa score: 0.068 LR average precision score: 0.714 -> test with 'GB' GB tn, fp: 56529, 334 GB fn, tp: 11, 88 GB f1 score: 0.338 GB cohens kappa score: 0.336 -> test with 'KNN' KNN tn, fp: 56627, 236 KNN fn, tp: 69, 30 KNN f1 score: 0.164 KNN cohens kappa score: 0.162 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54666, 2197 LR fn, tp: 10, 89 LR f1 score: 0.075 LR cohens kappa score: 0.072 LR average precision score: 0.716 -> test with 'GB' GB tn, fp: 56528, 335 GB fn, tp: 12, 87 GB f1 score: 0.334 GB cohens kappa score: 0.332 -> test with 'KNN' KNN tn, fp: 56610, 253 KNN fn, tp: 77, 22 KNN f1 score: 0.118 KNN cohens kappa score: 0.115 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227056 synthetic samples -> test with 'LR' LR tn, fp: 54988, 1875 LR fn, tp: 10, 86 LR f1 score: 0.084 LR cohens kappa score: 0.081 LR average precision score: 0.749 -> test with 'GB' GB tn, fp: 56518, 345 GB fn, tp: 15, 81 GB f1 score: 0.310 GB cohens kappa score: 0.308 -> test with 'KNN' KNN tn, fp: 56611, 252 KNN fn, tp: 74, 22 KNN f1 score: 0.119 KNN cohens kappa score: 0.117 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54577, 2286 LR fn, tp: 10, 89 LR f1 score: 0.072 LR cohens kappa score: 0.069 LR average precision score: 0.672 -> test with 'GB' GB tn, fp: 56596, 267 GB fn, tp: 14, 85 GB f1 score: 0.377 GB cohens kappa score: 0.375 -> test with 'KNN' KNN tn, fp: 56344, 519 KNN fn, tp: 75, 24 KNN f1 score: 0.075 KNN cohens kappa score: 0.072 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 55026, 1837 LR fn, tp: 9, 90 LR f1 score: 0.089 LR cohens kappa score: 0.086 LR average precision score: 0.642 -> test with 'GB' GB tn, fp: 56583, 280 GB fn, tp: 11, 88 GB f1 score: 0.377 GB cohens kappa score: 0.375 -> test with 'KNN' KNN tn, fp: 56348, 515 KNN fn, tp: 73, 26 KNN f1 score: 0.081 KNN cohens kappa score: 0.079 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54796, 2067 LR fn, tp: 10, 89 LR f1 score: 0.079 LR cohens kappa score: 0.076 LR average precision score: 0.703 -> test with 'GB' GB tn, fp: 56500, 363 GB fn, tp: 13, 86 GB f1 score: 0.314 GB cohens kappa score: 0.312 -> test with 'KNN' KNN tn, fp: 56714, 149 KNN fn, tp: 79, 20 KNN f1 score: 0.149 KNN cohens kappa score: 0.147 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 55442, 1421 LR fn, tp: 9, 90 LR f1 score: 0.112 LR cohens kappa score: 0.109 LR average precision score: 0.780 -> test with 'GB' GB tn, fp: 56547, 316 GB fn, tp: 11, 88 GB f1 score: 0.350 GB cohens kappa score: 0.348 -> test with 'KNN' KNN tn, fp: 56278, 585 KNN fn, tp: 70, 29 KNN f1 score: 0.081 KNN cohens kappa score: 0.079 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227056 synthetic samples -> test with 'LR' LR tn, fp: 55364, 1499 LR fn, tp: 11, 85 LR f1 score: 0.101 LR cohens kappa score: 0.098 LR average precision score: 0.770 -> test with 'GB' GB tn, fp: 56585, 278 GB fn, tp: 13, 83 GB f1 score: 0.363 GB cohens kappa score: 0.362 -> test with 'KNN' KNN tn, fp: 56435, 428 KNN fn, tp: 69, 27 KNN f1 score: 0.098 KNN cohens kappa score: 0.095 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54526, 2337 LR fn, tp: 6, 93 LR f1 score: 0.074 LR cohens kappa score: 0.070 LR average precision score: 0.693 -> test with 'GB' GB tn, fp: 56538, 325 GB fn, tp: 8, 91 GB f1 score: 0.353 GB cohens kappa score: 0.352 -> test with 'KNN' KNN tn, fp: 56573, 290 KNN fn, tp: 80, 19 KNN f1 score: 0.093 KNN cohens kappa score: 0.091 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54747, 2116 LR fn, tp: 12, 87 LR f1 score: 0.076 LR cohens kappa score: 0.073 LR average precision score: 0.650 -> test with 'GB' GB tn, fp: 56562, 301 GB fn, tp: 13, 86 GB f1 score: 0.354 GB cohens kappa score: 0.352 -> test with 'KNN' KNN tn, fp: 56524, 339 KNN fn, tp: 79, 20 KNN f1 score: 0.087 KNN cohens kappa score: 0.085 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 55021, 1842 LR fn, tp: 10, 89 LR f1 score: 0.088 LR cohens kappa score: 0.085 LR average precision score: 0.733 -> test with 'GB' GB tn, fp: 56539, 324 GB fn, tp: 13, 86 GB f1 score: 0.338 GB cohens kappa score: 0.336 -> test with 'KNN' KNN tn, fp: 56689, 174 KNN fn, tp: 77, 22 KNN f1 score: 0.149 KNN cohens kappa score: 0.147 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 55209, 1654 LR fn, tp: 7, 92 LR f1 score: 0.100 LR cohens kappa score: 0.097 LR average precision score: 0.777 -> test with 'GB' GB tn, fp: 56435, 428 GB fn, tp: 10, 89 GB f1 score: 0.289 GB cohens kappa score: 0.287 -> test with 'KNN' KNN tn, fp: 56295, 568 KNN fn, tp: 62, 37 KNN f1 score: 0.105 KNN cohens kappa score: 0.102 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227056 synthetic samples -> test with 'LR' LR tn, fp: 55824, 1039 LR fn, tp: 12, 84 LR f1 score: 0.138 LR cohens kappa score: 0.135 LR average precision score: 0.729 -> test with 'GB' GB tn, fp: 56665, 198 GB fn, tp: 16, 80 GB f1 score: 0.428 GB cohens kappa score: 0.426 -> test with 'KNN' KNN tn, fp: 56285, 578 KNN fn, tp: 69, 27 KNN f1 score: 0.077 KNN cohens kappa score: 0.074 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54476, 2387 LR fn, tp: 14, 85 LR f1 score: 0.066 LR cohens kappa score: 0.063 LR average precision score: 0.664 -> test with 'GB' GB tn, fp: 56599, 264 GB fn, tp: 18, 81 GB f1 score: 0.365 GB cohens kappa score: 0.363 -> test with 'KNN' KNN tn, fp: 56702, 161 KNN fn, tp: 77, 22 KNN f1 score: 0.156 KNN cohens kappa score: 0.154 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54533, 2330 LR fn, tp: 7, 92 LR f1 score: 0.073 LR cohens kappa score: 0.070 LR average precision score: 0.764 -> test with 'GB' GB tn, fp: 56521, 342 GB fn, tp: 8, 91 GB f1 score: 0.342 GB cohens kappa score: 0.340 -> test with 'KNN' KNN tn, fp: 56554, 309 KNN fn, tp: 74, 25 KNN f1 score: 0.115 KNN cohens kappa score: 0.113 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54993, 1870 LR fn, tp: 11, 88 LR f1 score: 0.086 LR cohens kappa score: 0.083 LR average precision score: 0.687 -> test with 'GB' GB tn, fp: 56481, 382 GB fn, tp: 13, 86 GB f1 score: 0.303 GB cohens kappa score: 0.301 -> test with 'KNN' KNN tn, fp: 56727, 136 KNN fn, tp: 76, 23 KNN f1 score: 0.178 KNN cohens kappa score: 0.177 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227059 synthetic samples -> test with 'LR' LR tn, fp: 54666, 2197 LR fn, tp: 8, 91 LR f1 score: 0.076 LR cohens kappa score: 0.073 LR average precision score: 0.775 -> test with 'GB' GB tn, fp: 56535, 328 GB fn, tp: 10, 89 GB f1 score: 0.345 GB cohens kappa score: 0.343 -> test with 'KNN' KNN tn, fp: 56518, 345 KNN fn, tp: 76, 23 KNN f1 score: 0.099 KNN cohens kappa score: 0.096 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 227056 synthetic samples -> test with 'LR' LR tn, fp: 55361, 1502 LR fn, tp: 8, 88 LR f1 score: 0.104 LR cohens kappa score: 0.102 LR average precision score: 0.688 -> test with 'GB' GB tn, fp: 56557, 306 GB fn, tp: 10, 86 GB f1 score: 0.352 GB cohens kappa score: 0.351 -> test with 'KNN' KNN tn, fp: 56283, 580 KNN fn, tp: 67, 29 KNN f1 score: 0.082 KNN cohens kappa score: 0.080 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 55824, 3513 LR fn, tp: 16, 93 LR f1 score: 0.138 LR cohens kappa score: 0.135 LR average precision score: 0.801 average: LR tn, fp: 54823.16, 2039.84 LR fn, tp: 9.88, 88.52 LR f1 score: 0.083 LR cohens kappa score: 0.080 LR average precision score: 0.711 minimum: LR tn, fp: 53350, 1039 LR fn, tp: 6, 83 LR f1 score: 0.047 LR cohens kappa score: 0.044 LR average precision score: 0.579 -----[ GB ]----- maximum: GB tn, fp: 56665, 429 GB fn, tp: 20, 91 GB f1 score: 0.428 GB cohens kappa score: 0.426 average: GB tn, fp: 56537.04, 325.96 GB fn, tp: 11.96, 86.44 GB f1 score: 0.341 GB cohens kappa score: 0.340 minimum: GB tn, fp: 56434, 198 GB fn, tp: 8, 79 GB f1 score: 0.288 GB cohens kappa score: 0.286 -----[ KNN ]----- maximum: KNN tn, fp: 56727, 585 KNN fn, tp: 80, 37 KNN f1 score: 0.178 KNN cohens kappa score: 0.177 average: KNN tn, fp: 56496.04, 366.96 KNN fn, tp: 73.36, 25.04 KNN f1 score: 0.109 KNN cohens kappa score: 0.107 minimum: KNN tn, fp: 56278, 136 KNN fn, tp: 62, 19 KNN f1 score: 0.075 KNN cohens kappa score: 0.072