/////////////////////////////////////////// // Running convGAN-majority-full on folding_kr-vs-k-zero-one_vs_draw /////////////////////////////////////////// Load 'data_input/folding_kr-vs-k-zero-one_vs_draw' 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 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 556, 4 GAN fn, tp: 2, 19 GAN f1 score: 0.864 GAN cohens kappa score: 0.858 -> test with 'LR' LR tn, fp: 549, 11 LR fn, tp: 3, 18 LR f1 score: 0.720 LR cohens kappa score: 0.708 LR average precision score: 0.844 -> test with 'GB' GB tn, fp: 558, 2 GB fn, tp: 1, 20 GB f1 score: 0.930 GB cohens kappa score: 0.928 -> test with 'KNN' KNN tn, fp: 554, 6 KNN fn, tp: 3, 18 KNN f1 score: 0.800 KNN cohens kappa score: 0.792 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 559, 1 GAN fn, tp: 1, 20 GAN f1 score: 0.952 GAN cohens kappa score: 0.951 -> test with 'LR' LR tn, fp: 548, 12 LR fn, tp: 0, 21 LR f1 score: 0.778 LR cohens kappa score: 0.768 LR average precision score: 0.928 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 3, 18 GB f1 score: 0.923 GB cohens kappa score: 0.920 -> test with 'KNN' KNN tn, fp: 558, 2 KNN fn, tp: 0, 21 KNN f1 score: 0.955 KNN cohens kappa score: 0.953 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 556, 4 GAN fn, tp: 1, 20 GAN f1 score: 0.889 GAN cohens kappa score: 0.884 -> test with 'LR' LR tn, fp: 537, 23 LR fn, tp: 0, 21 LR f1 score: 0.646 LR cohens kappa score: 0.628 LR average precision score: 0.820 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 1, 20 GB f1 score: 0.976 GB cohens kappa score: 0.975 -> test with 'KNN' KNN tn, fp: 553, 7 KNN fn, tp: 0, 21 KNN f1 score: 0.857 KNN cohens kappa score: 0.851 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 559, 1 GAN fn, tp: 1, 20 GAN f1 score: 0.952 GAN cohens kappa score: 0.951 -> test with 'LR' LR tn, fp: 543, 17 LR fn, tp: 2, 19 LR f1 score: 0.667 LR cohens kappa score: 0.651 LR average precision score: 0.882 -> test with 'GB' GB tn, fp: 556, 4 GB fn, tp: 1, 20 GB f1 score: 0.889 GB cohens kappa score: 0.884 -> test with 'KNN' KNN tn, fp: 551, 9 KNN fn, tp: 0, 21 KNN f1 score: 0.824 KNN cohens kappa score: 0.816 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2156 synthetic samples -> test with GAN.predict GAN tn, fp: 556, 0 GAN fn, tp: 0, 21 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 537, 19 LR fn, tp: 0, 21 LR f1 score: 0.689 LR cohens kappa score: 0.673 LR average precision score: 0.982 -> test with 'GB' GB tn, fp: 556, 0 GB fn, tp: 1, 20 GB f1 score: 0.976 GB cohens kappa score: 0.975 -> test with 'KNN' KNN tn, fp: 549, 7 KNN fn, tp: 0, 21 KNN f1 score: 0.857 KNN cohens kappa score: 0.851 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 560, 0 GAN fn, tp: 1, 20 GAN f1 score: 0.976 GAN cohens kappa score: 0.975 -> test with 'LR' LR tn, fp: 546, 14 LR fn, tp: 1, 20 LR f1 score: 0.727 LR cohens kappa score: 0.715 LR average precision score: 0.927 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 1, 20 GB f1 score: 0.976 GB cohens kappa score: 0.975 -> test with 'KNN' KNN tn, fp: 554, 6 KNN fn, tp: 0, 21 KNN f1 score: 0.875 KNN cohens kappa score: 0.870 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 560, 0 GAN fn, tp: 2, 19 GAN f1 score: 0.950 GAN cohens kappa score: 0.948 -> test with 'LR' LR tn, fp: 553, 7 LR fn, tp: 2, 19 LR f1 score: 0.809 LR cohens kappa score: 0.801 LR average precision score: 0.919 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 1, 20 GB f1 score: 0.976 GB cohens kappa score: 0.975 -> test with 'KNN' KNN tn, fp: 556, 4 KNN fn, tp: 2, 19 KNN f1 score: 0.864 KNN cohens kappa score: 0.858 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 558, 2 GAN fn, tp: 1, 20 GAN f1 score: 0.930 GAN cohens kappa score: 0.928 -> test with 'LR' LR tn, fp: 546, 14 LR fn, tp: 0, 21 LR f1 score: 0.750 LR cohens kappa score: 0.738 LR average precision score: 0.905 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 1, 20 GB f1 score: 0.976 GB cohens kappa score: 0.975 -> test with 'KNN' KNN tn, fp: 558, 2 KNN fn, tp: 1, 20 KNN f1 score: 0.930 KNN cohens kappa score: 0.928 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 554, 6 GAN fn, tp: 0, 21 GAN f1 score: 0.875 GAN cohens kappa score: 0.870 -> test with 'LR' LR tn, fp: 539, 21 LR fn, tp: 0, 21 LR f1 score: 0.667 LR cohens kappa score: 0.650 LR average precision score: 0.883 -> test with 'GB' GB tn, fp: 559, 1 GB fn, tp: 2, 19 GB f1 score: 0.927 GB cohens kappa score: 0.924 -> test with 'KNN' KNN tn, fp: 550, 10 KNN fn, tp: 1, 20 KNN f1 score: 0.784 KNN cohens kappa score: 0.775 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2156 synthetic samples -> test with GAN.predict GAN tn, fp: 556, 0 GAN fn, tp: 1, 20 GAN f1 score: 0.976 GAN cohens kappa score: 0.975 -> test with 'LR' LR tn, fp: 536, 20 LR fn, tp: 0, 21 LR f1 score: 0.677 LR cohens kappa score: 0.661 LR average precision score: 0.896 -> test with 'GB' GB tn, fp: 556, 0 GB fn, tp: 0, 21 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 553, 3 KNN fn, tp: 0, 21 KNN f1 score: 0.933 KNN cohens kappa score: 0.931 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 558, 2 GAN fn, tp: 2, 19 GAN f1 score: 0.905 GAN cohens kappa score: 0.901 -> test with 'LR' LR tn, fp: 547, 13 LR fn, tp: 0, 21 LR f1 score: 0.764 LR cohens kappa score: 0.753 LR average precision score: 0.946 -> test with 'GB' GB tn, fp: 559, 1 GB fn, tp: 1, 20 GB f1 score: 0.952 GB cohens kappa score: 0.951 -> test with 'KNN' KNN tn, fp: 552, 8 KNN fn, tp: 0, 21 KNN f1 score: 0.840 KNN cohens kappa score: 0.833 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 558, 2 GAN fn, tp: 2, 19 GAN f1 score: 0.905 GAN cohens kappa score: 0.901 -> test with 'LR' LR tn, fp: 554, 6 LR fn, tp: 1, 20 LR f1 score: 0.851 LR cohens kappa score: 0.845 LR average precision score: 0.890 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 1, 20 GB f1 score: 0.976 GB cohens kappa score: 0.975 -> test with 'KNN' KNN tn, fp: 555, 5 KNN fn, tp: 0, 21 KNN f1 score: 0.894 KNN cohens kappa score: 0.889 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 559, 1 GAN fn, tp: 1, 20 GAN f1 score: 0.952 GAN cohens kappa score: 0.951 -> test with 'LR' LR tn, fp: 538, 22 LR fn, tp: 1, 20 LR f1 score: 0.635 LR cohens kappa score: 0.616 LR average precision score: 0.865 -> test with 'GB' GB tn, fp: 559, 1 GB fn, tp: 2, 19 GB f1 score: 0.927 GB cohens kappa score: 0.924 -> test with 'KNN' KNN tn, fp: 553, 7 KNN fn, tp: 1, 20 KNN f1 score: 0.833 KNN cohens kappa score: 0.826 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 557, 3 GAN fn, tp: 1, 20 GAN f1 score: 0.909 GAN cohens kappa score: 0.906 -> test with 'LR' LR tn, fp: 541, 19 LR fn, tp: 1, 20 LR f1 score: 0.667 LR cohens kappa score: 0.650 LR average precision score: 0.938 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 1, 20 GB f1 score: 0.976 GB cohens kappa score: 0.975 -> test with 'KNN' KNN tn, fp: 556, 4 KNN fn, tp: 0, 21 KNN f1 score: 0.913 KNN cohens kappa score: 0.909 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2156 synthetic samples -> test with GAN.predict GAN tn, fp: 552, 4 GAN fn, tp: 0, 21 GAN f1 score: 0.913 GAN cohens kappa score: 0.909 -> test with 'LR' LR tn, fp: 533, 23 LR fn, tp: 0, 21 LR f1 score: 0.646 LR cohens kappa score: 0.628 LR average precision score: 0.905 -> test with 'GB' GB tn, fp: 555, 1 GB fn, tp: 0, 21 GB f1 score: 0.977 GB cohens kappa score: 0.976 -> test with 'KNN' KNN tn, fp: 545, 11 KNN fn, tp: 0, 21 KNN f1 score: 0.792 KNN cohens kappa score: 0.783 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 557, 3 GAN fn, tp: 1, 20 GAN f1 score: 0.909 GAN cohens kappa score: 0.906 -> test with 'LR' LR tn, fp: 549, 11 LR fn, tp: 1, 20 LR f1 score: 0.769 LR cohens kappa score: 0.759 LR average precision score: 0.936 -> test with 'GB' GB tn, fp: 558, 2 GB fn, tp: 0, 21 GB f1 score: 0.955 GB cohens kappa score: 0.953 -> test with 'KNN' KNN tn, fp: 553, 7 KNN fn, tp: 0, 21 KNN f1 score: 0.857 KNN cohens kappa score: 0.851 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 554, 6 GAN fn, tp: 0, 21 GAN f1 score: 0.875 GAN cohens kappa score: 0.870 -> test with 'LR' LR tn, fp: 542, 18 LR fn, tp: 0, 21 LR f1 score: 0.700 LR cohens kappa score: 0.685 LR average precision score: 0.959 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 0, 21 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 553, 7 KNN fn, tp: 0, 21 KNN f1 score: 0.857 KNN cohens kappa score: 0.851 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 560, 0 GAN fn, tp: 5, 16 GAN f1 score: 0.865 GAN cohens kappa score: 0.861 -> test with 'LR' LR tn, fp: 544, 16 LR fn, tp: 4, 17 LR f1 score: 0.630 LR cohens kappa score: 0.613 LR average precision score: 0.808 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 4, 17 GB f1 score: 0.895 GB cohens kappa score: 0.891 -> test with 'KNN' KNN tn, fp: 552, 8 KNN fn, tp: 1, 20 KNN f1 score: 0.816 KNN cohens kappa score: 0.808 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 557, 3 GAN fn, tp: 1, 20 GAN f1 score: 0.909 GAN cohens kappa score: 0.906 -> test with 'LR' LR tn, fp: 542, 18 LR fn, tp: 0, 21 LR f1 score: 0.700 LR cohens kappa score: 0.685 LR average precision score: 0.910 -> test with 'GB' GB tn, fp: 558, 2 GB fn, tp: 1, 20 GB f1 score: 0.930 GB cohens kappa score: 0.928 -> test with 'KNN' KNN tn, fp: 554, 6 KNN fn, tp: 0, 21 KNN f1 score: 0.875 KNN cohens kappa score: 0.870 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2156 synthetic samples -> test with GAN.predict GAN tn, fp: 556, 0 GAN fn, tp: 2, 19 GAN f1 score: 0.950 GAN cohens kappa score: 0.948 -> test with 'LR' LR tn, fp: 553, 3 LR fn, tp: 1, 20 LR f1 score: 0.909 LR cohens kappa score: 0.905 LR average precision score: 0.945 -> test with 'GB' GB tn, fp: 556, 0 GB fn, tp: 2, 19 GB f1 score: 0.950 GB cohens kappa score: 0.948 -> test with 'KNN' KNN tn, fp: 549, 7 KNN fn, tp: 1, 20 KNN f1 score: 0.833 KNN cohens kappa score: 0.826 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 558, 2 GAN fn, tp: 0, 21 GAN f1 score: 0.955 GAN cohens kappa score: 0.953 -> test with 'LR' LR tn, fp: 543, 17 LR fn, tp: 0, 21 LR f1 score: 0.712 LR cohens kappa score: 0.698 LR average precision score: 0.963 -> test with 'GB' GB tn, fp: 558, 2 GB fn, tp: 1, 20 GB f1 score: 0.930 GB cohens kappa score: 0.928 -> test with 'KNN' KNN tn, fp: 555, 5 KNN fn, tp: 1, 20 KNN f1 score: 0.870 KNN cohens kappa score: 0.864 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 558, 2 GAN fn, tp: 0, 21 GAN f1 score: 0.955 GAN cohens kappa score: 0.953 -> test with 'LR' LR tn, fp: 547, 13 LR fn, tp: 1, 20 LR f1 score: 0.741 LR cohens kappa score: 0.729 LR average precision score: 0.911 -> test with 'GB' GB tn, fp: 560, 0 GB fn, tp: 3, 18 GB f1 score: 0.923 GB cohens kappa score: 0.920 -> test with 'KNN' KNN tn, fp: 554, 6 KNN fn, tp: 1, 20 KNN f1 score: 0.851 KNN cohens kappa score: 0.845 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 558, 2 GAN fn, tp: 0, 21 GAN f1 score: 0.955 GAN cohens kappa score: 0.953 -> test with 'LR' LR tn, fp: 541, 19 LR fn, tp: 0, 21 LR f1 score: 0.689 LR cohens kappa score: 0.673 LR average precision score: 0.838 -> test with 'GB' GB tn, fp: 559, 1 GB fn, tp: 0, 21 GB f1 score: 0.977 GB cohens kappa score: 0.976 -> test with 'KNN' KNN tn, fp: 552, 8 KNN fn, tp: 0, 21 KNN f1 score: 0.840 KNN cohens kappa score: 0.833 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2152 synthetic samples -> test with GAN.predict GAN tn, fp: 556, 4 GAN fn, tp: 1, 20 GAN f1 score: 0.889 GAN cohens kappa score: 0.884 -> test with 'LR' LR tn, fp: 541, 19 LR fn, tp: 1, 20 LR f1 score: 0.667 LR cohens kappa score: 0.650 LR average precision score: 0.886 -> test with 'GB' GB tn, fp: 557, 3 GB fn, tp: 1, 20 GB f1 score: 0.909 GB cohens kappa score: 0.906 -> test with 'KNN' KNN tn, fp: 551, 9 KNN fn, tp: 1, 20 KNN f1 score: 0.800 KNN cohens kappa score: 0.791 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2156 synthetic samples -> test with GAN.predict GAN tn, fp: 552, 4 GAN fn, tp: 0, 21 GAN f1 score: 0.913 GAN cohens kappa score: 0.909 -> test with 'LR' LR tn, fp: 542, 14 LR fn, tp: 1, 20 LR f1 score: 0.727 LR cohens kappa score: 0.714 LR average precision score: 0.915 -> test with 'GB' GB tn, fp: 555, 1 GB fn, tp: 0, 21 GB f1 score: 0.977 GB cohens kappa score: 0.976 -> test with 'KNN' KNN tn, fp: 549, 7 KNN fn, tp: 0, 21 KNN f1 score: 0.857 KNN cohens kappa score: 0.851 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 554, 23 LR fn, tp: 4, 21 LR f1 score: 0.909 LR cohens kappa score: 0.905 LR average precision score: 0.982 average: LR tn, fp: 543.64, 15.56 LR fn, tp: 0.8, 20.2 LR f1 score: 0.717 LR cohens kappa score: 0.704 LR average precision score: 0.904 minimum: LR tn, fp: 533, 3 LR fn, tp: 0, 17 LR f1 score: 0.630 LR cohens kappa score: 0.613 LR average precision score: 0.808 -----[ GB ]----- maximum: GB tn, fp: 560, 4 GB fn, tp: 4, 21 GB f1 score: 1.000 GB cohens kappa score: 1.000 average: GB tn, fp: 558.36, 0.84 GB fn, tp: 1.16, 19.84 GB f1 score: 0.952 GB cohens kappa score: 0.950 minimum: GB tn, fp: 555, 0 GB fn, tp: 0, 17 GB f1 score: 0.889 GB cohens kappa score: 0.884 -----[ KNN ]----- maximum: KNN tn, fp: 558, 11 KNN fn, tp: 3, 21 KNN f1 score: 0.955 KNN cohens kappa score: 0.953 average: KNN tn, fp: 552.76, 6.44 KNN fn, tp: 0.52, 20.48 KNN f1 score: 0.856 KNN cohens kappa score: 0.850 minimum: KNN tn, fp: 545, 2 KNN fn, tp: 0, 18 KNN f1 score: 0.784 KNN cohens kappa score: 0.775 -----[ GAN ]----- maximum: GAN tn, fp: 560, 6 GAN fn, tp: 5, 21 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 average: GAN tn, fp: 556.96, 2.24 GAN fn, tp: 1.04, 19.96 GAN f1 score: 0.925 GAN cohens kappa score: 0.922 minimum: GAN tn, fp: 552, 0 GAN fn, tp: 0, 16 GAN f1 score: 0.864 GAN cohens kappa score: 0.858