/////////////////////////////////////////// // Running convGAN-majority-5 on folding_shuttle-2_vs_5 /////////////////////////////////////////// Load 'data_input/folding_shuttle-2_vs_5' 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 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 653, 1 LR fn, tp: 0, 10 LR f1 score: 0.952 LR cohens kappa score: 0.952 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 653, 1 KNN fn, tp: 0, 10 KNN f1 score: 0.952 KNN cohens kappa score: 0.952 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 653, 1 KNN fn, tp: 0, 10 KNN f1 score: 0.952 KNN cohens kappa score: 0.952 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 652, 2 KNN fn, tp: 0, 10 KNN f1 score: 0.909 KNN cohens kappa score: 0.908 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2576 synthetic samples -> test with GAN.predict GAN tn, fp: 651, 0 GAN fn, tp: 0, 9 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 651, 0 LR fn, tp: 0, 9 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 651, 0 GB fn, tp: 0, 9 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 651, 0 KNN fn, tp: 0, 9 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 653, 1 KNN fn, tp: 0, 10 KNN f1 score: 0.952 KNN cohens kappa score: 0.952 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 652, 2 KNN fn, tp: 0, 10 KNN f1 score: 0.909 KNN cohens kappa score: 0.908 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2576 synthetic samples -> test with GAN.predict GAN tn, fp: 651, 0 GAN fn, tp: 0, 9 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 650, 1 LR fn, tp: 0, 9 LR f1 score: 0.947 LR cohens kappa score: 0.947 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 651, 0 GB fn, tp: 0, 9 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 650, 1 KNN fn, tp: 0, 9 KNN f1 score: 0.947 KNN cohens kappa score: 0.947 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 653, 1 KNN fn, tp: 0, 10 KNN f1 score: 0.952 KNN cohens kappa score: 0.952 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 653, 1 KNN fn, tp: 0, 10 KNN f1 score: 0.952 KNN cohens kappa score: 0.952 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2576 synthetic samples -> test with GAN.predict GAN tn, fp: 651, 0 GAN fn, tp: 0, 9 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 651, 0 LR fn, tp: 0, 9 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 650, 1 GB fn, tp: 0, 9 GB f1 score: 0.947 GB cohens kappa score: 0.947 -> test with 'KNN' KNN tn, fp: 649, 2 KNN fn, tp: 0, 9 KNN f1 score: 0.900 KNN cohens kappa score: 0.898 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 651, 3 KNN fn, tp: 0, 10 KNN f1 score: 0.870 KNN cohens kappa score: 0.867 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 653, 1 KNN fn, tp: 0, 10 KNN f1 score: 0.952 KNN cohens kappa score: 0.952 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2576 synthetic samples -> test with GAN.predict GAN tn, fp: 651, 0 GAN fn, tp: 0, 9 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 651, 0 LR fn, tp: 0, 9 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 651, 0 GB fn, tp: 0, 9 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 650, 1 KNN fn, tp: 0, 9 KNN f1 score: 0.947 KNN cohens kappa score: 0.947 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 653, 1 KNN fn, tp: 0, 10 KNN f1 score: 0.952 KNN cohens kappa score: 0.952 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 653, 1 LR fn, tp: 0, 10 LR f1 score: 0.952 LR cohens kappa score: 0.952 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 651, 3 KNN fn, tp: 0, 10 KNN f1 score: 0.870 KNN cohens kappa score: 0.867 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2574 synthetic samples -> test with GAN.predict GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 654, 0 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 654, 0 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 654, 0 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 2576 synthetic samples -> test with GAN.predict GAN tn, fp: 651, 0 GAN fn, tp: 0, 9 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 -> test with 'LR' LR tn, fp: 651, 0 LR fn, tp: 0, 9 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 -> test with 'GB' GB tn, fp: 651, 0 GB fn, tp: 0, 9 GB f1 score: 1.000 GB cohens kappa score: 1.000 -> test with 'KNN' KNN tn, fp: 650, 1 KNN fn, tp: 0, 9 KNN f1 score: 0.947 KNN cohens kappa score: 0.947 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 654, 1 LR fn, tp: 0, 10 LR f1 score: 1.000 LR cohens kappa score: 1.000 LR average precision score: 1.000 average: LR tn, fp: 653.28, 0.12 LR fn, tp: 0.0, 9.8 LR f1 score: 0.994 LR cohens kappa score: 0.994 LR average precision score: 1.000 minimum: LR tn, fp: 650, 0 LR fn, tp: 0, 9 LR f1 score: 0.947 LR cohens kappa score: 0.947 LR average precision score: 1.000 -----[ GB ]----- maximum: GB tn, fp: 654, 1 GB fn, tp: 0, 10 GB f1 score: 1.000 GB cohens kappa score: 1.000 average: GB tn, fp: 653.36, 0.04 GB fn, tp: 0.0, 9.8 GB f1 score: 0.998 GB cohens kappa score: 0.998 minimum: GB tn, fp: 650, 0 GB fn, tp: 0, 9 GB f1 score: 0.947 GB cohens kappa score: 0.947 -----[ KNN ]----- maximum: KNN tn, fp: 654, 3 KNN fn, tp: 0, 10 KNN f1 score: 1.000 KNN cohens kappa score: 1.000 average: KNN tn, fp: 652.52, 0.88 KNN fn, tp: 0.0, 9.8 KNN f1 score: 0.959 KNN cohens kappa score: 0.958 minimum: KNN tn, fp: 649, 0 KNN fn, tp: 0, 9 KNN f1 score: 0.870 KNN cohens kappa score: 0.867 -----[ GAN ]----- maximum: GAN tn, fp: 654, 0 GAN fn, tp: 0, 10 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 average: GAN tn, fp: 653.4, 0.0 GAN fn, tp: 0.0, 9.8 GAN f1 score: 1.000 GAN cohens kappa score: 1.000 minimum: GAN tn, fp: 651, 0 GAN fn, tp: 0, 9 GAN f1 score: 1.000 GAN cohens kappa score: 1.000