/////////////////////////////////////////// // Running SimpleGAN 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 Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28867, 24 LR fn, tp: 64, 196 LR f1 score: 0.817 LR cohens kappa score: 0.815 LR average precision score: 0.868 -> test with 'GB' GB tn, fp: 28874, 17 GB fn, tp: 59, 201 GB f1 score: 0.841 GB cohens kappa score: 0.840 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 165, 95 KNN f1 score: 0.535 KNN cohens kappa score: 0.533 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28883, 8 LR fn, tp: 59, 201 LR f1 score: 0.857 LR cohens kappa score: 0.856 LR average precision score: 0.889 -> test with 'GB' GB tn, fp: 28886, 5 GB fn, tp: 56, 204 GB f1 score: 0.870 GB cohens kappa score: 0.869 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 153, 107 KNN f1 score: 0.583 KNN cohens kappa score: 0.581 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28867, 24 LR fn, tp: 58, 202 LR f1 score: 0.831 LR cohens kappa score: 0.830 LR average precision score: 0.890 -> test with 'GB' GB tn, fp: 28885, 6 GB fn, tp: 71, 189 GB f1 score: 0.831 GB cohens kappa score: 0.829 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 178, 82 KNN f1 score: 0.478 KNN cohens kappa score: 0.476 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28877, 14 LR fn, tp: 69, 191 LR f1 score: 0.822 LR cohens kappa score: 0.820 LR average precision score: 0.857 -> test with 'GB' GB tn, fp: 28883, 8 GB fn, tp: 72, 188 GB f1 score: 0.825 GB cohens kappa score: 0.823 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 166, 94 KNN f1 score: 0.530 KNN cohens kappa score: 0.527 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114524 synthetic samples -> test with 'LR' LR tn, fp: 28879, 12 LR fn, tp: 82, 174 LR f1 score: 0.787 LR cohens kappa score: 0.786 LR average precision score: 0.831 -> test with 'GB' GB tn, fp: 28880, 11 GB fn, tp: 85, 171 GB f1 score: 0.781 GB cohens kappa score: 0.779 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 177, 79 KNN f1 score: 0.472 KNN cohens kappa score: 0.469 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28880, 11 LR fn, tp: 74, 186 LR f1 score: 0.814 LR cohens kappa score: 0.813 LR average precision score: 0.871 -> test with 'GB' GB tn, fp: 28886, 5 GB fn, tp: 79, 181 GB f1 score: 0.812 GB cohens kappa score: 0.810 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 167, 93 KNN f1 score: 0.527 KNN cohens kappa score: 0.525 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28869, 22 LR fn, tp: 60, 200 LR f1 score: 0.830 LR cohens kappa score: 0.828 LR average precision score: 0.894 -> test with 'GB' GB tn, fp: 28880, 11 GB fn, tp: 58, 202 GB f1 score: 0.854 GB cohens kappa score: 0.853 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 164, 96 KNN f1 score: 0.539 KNN cohens kappa score: 0.537 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28871, 20 LR fn, tp: 63, 197 LR f1 score: 0.826 LR cohens kappa score: 0.825 LR average precision score: 0.835 -> test with 'GB' GB tn, fp: 28880, 11 GB fn, tp: 71, 189 GB f1 score: 0.822 GB cohens kappa score: 0.820 -> test with 'KNN' KNN tn, fp: 28889, 2 KNN fn, tp: 162, 98 KNN f1 score: 0.544 KNN cohens kappa score: 0.542 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28868, 23 LR fn, tp: 68, 192 LR f1 score: 0.808 LR cohens kappa score: 0.807 LR average precision score: 0.856 -> test with 'GB' GB tn, fp: 28877, 14 GB fn, tp: 67, 193 GB f1 score: 0.827 GB cohens kappa score: 0.825 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 163, 97 KNN f1 score: 0.543 KNN cohens kappa score: 0.541 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114524 synthetic samples -> test with 'LR' LR tn, fp: 28875, 16 LR fn, tp: 66, 190 LR f1 score: 0.823 LR cohens kappa score: 0.821 LR average precision score: 0.868 -> test with 'GB' GB tn, fp: 28876, 15 GB fn, tp: 69, 187 GB f1 score: 0.817 GB cohens kappa score: 0.815 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 169, 87 KNN f1 score: 0.506 KNN cohens kappa score: 0.504 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28874, 17 LR fn, tp: 68, 192 LR f1 score: 0.819 LR cohens kappa score: 0.817 LR average precision score: 0.869 -> test with 'GB' GB tn, fp: 28876, 15 GB fn, tp: 71, 189 GB f1 score: 0.815 GB cohens kappa score: 0.813 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 175, 85 KNN f1 score: 0.493 KNN cohens kappa score: 0.491 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28877, 14 LR fn, tp: 66, 194 LR f1 score: 0.829 LR cohens kappa score: 0.828 LR average precision score: 0.861 -> test with 'GB' GB tn, fp: 28879, 12 GB fn, tp: 68, 192 GB f1 score: 0.828 GB cohens kappa score: 0.826 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 174, 86 KNN f1 score: 0.496 KNN cohens kappa score: 0.493 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28866, 25 LR fn, tp: 71, 189 LR f1 score: 0.797 LR cohens kappa score: 0.796 LR average precision score: 0.834 -> test with 'GB' GB tn, fp: 28880, 11 GB fn, tp: 74, 186 GB f1 score: 0.814 GB cohens kappa score: 0.813 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 167, 93 KNN f1 score: 0.525 KNN cohens kappa score: 0.523 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28880, 11 LR fn, tp: 70, 190 LR f1 score: 0.824 LR cohens kappa score: 0.823 LR average precision score: 0.873 -> test with 'GB' GB tn, fp: 28880, 11 GB fn, tp: 72, 188 GB f1 score: 0.819 GB cohens kappa score: 0.818 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 171, 89 KNN f1 score: 0.509 KNN cohens kappa score: 0.506 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114524 synthetic samples -> test with 'LR' LR tn, fp: 28873, 18 LR fn, tp: 65, 191 LR f1 score: 0.822 LR cohens kappa score: 0.820 LR average precision score: 0.874 -> test with 'GB' GB tn, fp: 28879, 12 GB fn, tp: 64, 192 GB f1 score: 0.835 GB cohens kappa score: 0.833 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 152, 104 KNN f1 score: 0.578 KNN cohens kappa score: 0.576 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28872, 19 LR fn, tp: 68, 192 LR f1 score: 0.815 LR cohens kappa score: 0.814 LR average precision score: 0.871 -> test with 'GB' GB tn, fp: 28882, 9 GB fn, tp: 64, 196 GB f1 score: 0.843 GB cohens kappa score: 0.842 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 167, 93 KNN f1 score: 0.527 KNN cohens kappa score: 0.525 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28871, 20 LR fn, tp: 74, 186 LR f1 score: 0.798 LR cohens kappa score: 0.797 LR average precision score: 0.841 -> test with 'GB' GB tn, fp: 28868, 23 GB fn, tp: 75, 185 GB f1 score: 0.791 GB cohens kappa score: 0.789 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 166, 94 KNN f1 score: 0.531 KNN cohens kappa score: 0.529 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28882, 9 LR fn, tp: 60, 200 LR f1 score: 0.853 LR cohens kappa score: 0.852 LR average precision score: 0.854 -> test with 'GB' GB tn, fp: 28875, 16 GB fn, tp: 61, 199 GB f1 score: 0.838 GB cohens kappa score: 0.837 -> test with 'KNN' KNN tn, fp: 28889, 2 KNN fn, tp: 175, 85 KNN f1 score: 0.490 KNN cohens kappa score: 0.488 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28875, 16 LR fn, tp: 68, 192 LR f1 score: 0.821 LR cohens kappa score: 0.819 LR average precision score: 0.871 -> test with 'GB' GB tn, fp: 28877, 14 GB fn, tp: 68, 192 GB f1 score: 0.824 GB cohens kappa score: 0.823 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 162, 98 KNN f1 score: 0.547 KNN cohens kappa score: 0.545 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114524 synthetic samples -> test with 'LR' LR tn, fp: 28875, 16 LR fn, tp: 64, 192 LR f1 score: 0.828 LR cohens kappa score: 0.826 LR average precision score: 0.861 -> test with 'GB' GB tn, fp: 28879, 12 GB fn, tp: 67, 189 GB f1 score: 0.827 GB cohens kappa score: 0.826 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 162, 94 KNN f1 score: 0.536 KNN cohens kappa score: 0.533 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28869, 22 LR fn, tp: 65, 195 LR f1 score: 0.818 LR cohens kappa score: 0.816 LR average precision score: 0.865 -> test with 'GB' GB tn, fp: 28878, 13 GB fn, tp: 65, 195 GB f1 score: 0.833 GB cohens kappa score: 0.832 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 172, 88 KNN f1 score: 0.504 KNN cohens kappa score: 0.502 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28880, 11 LR fn, tp: 70, 190 LR f1 score: 0.824 LR cohens kappa score: 0.823 LR average precision score: 0.862 -> test with 'GB' GB tn, fp: 28879, 12 GB fn, tp: 73, 187 GB f1 score: 0.815 GB cohens kappa score: 0.813 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 166, 94 KNN f1 score: 0.531 KNN cohens kappa score: 0.529 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28874, 17 LR fn, tp: 72, 188 LR f1 score: 0.809 LR cohens kappa score: 0.807 LR average precision score: 0.867 -> test with 'GB' GB tn, fp: 28882, 9 GB fn, tp: 74, 186 GB f1 score: 0.818 GB cohens kappa score: 0.816 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 168, 92 KNN f1 score: 0.523 KNN cohens kappa score: 0.520 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114528 synthetic samples -> test with 'LR' LR tn, fp: 28874, 17 LR fn, tp: 65, 195 LR f1 score: 0.826 LR cohens kappa score: 0.825 LR average precision score: 0.876 -> test with 'GB' GB tn, fp: 28878, 13 GB fn, tp: 71, 189 GB f1 score: 0.818 GB cohens kappa score: 0.817 -> test with 'KNN' KNN tn, fp: 28891, 0 KNN fn, tp: 173, 87 KNN f1 score: 0.501 KNN cohens kappa score: 0.499 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 114524 synthetic samples -> test with 'LR' LR tn, fp: 28871, 20 LR fn, tp: 64, 192 LR f1 score: 0.821 LR cohens kappa score: 0.819 LR average precision score: 0.850 -> test with 'GB' GB tn, fp: 28884, 7 GB fn, tp: 63, 193 GB f1 score: 0.846 GB cohens kappa score: 0.845 -> test with 'KNN' KNN tn, fp: 28890, 1 KNN fn, tp: 156, 100 KNN f1 score: 0.560 KNN cohens kappa score: 0.558 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 28883, 25 LR fn, tp: 82, 202 LR f1 score: 0.857 LR cohens kappa score: 0.856 LR average precision score: 0.894 average: LR tn, fp: 28873.96, 17.04 LR fn, tp: 66.92, 192.28 LR f1 score: 0.821 LR cohens kappa score: 0.819 LR average precision score: 0.863 minimum: LR tn, fp: 28866, 8 LR fn, tp: 58, 174 LR f1 score: 0.787 LR cohens kappa score: 0.786 LR average precision score: 0.831 -----[ GB ]----- maximum: GB tn, fp: 28886, 23 GB fn, tp: 85, 204 GB f1 score: 0.870 GB cohens kappa score: 0.869 average: GB tn, fp: 28879.32, 11.68 GB fn, tp: 68.68, 190.52 GB f1 score: 0.826 GB cohens kappa score: 0.824 minimum: GB tn, fp: 28868, 5 GB fn, tp: 56, 171 GB f1 score: 0.781 GB cohens kappa score: 0.779 -----[ KNN ]----- maximum: KNN tn, fp: 28891, 2 KNN fn, tp: 178, 107 KNN f1 score: 0.583 KNN cohens kappa score: 0.581 average: KNN tn, fp: 28890.48, 0.52 KNN fn, tp: 166.8, 92.4 KNN f1 score: 0.524 KNN cohens kappa score: 0.522 minimum: KNN tn, fp: 28889, 0 KNN fn, tp: 152, 79 KNN f1 score: 0.472 KNN cohens kappa score: 0.469