/////////////////////////////////////////// // Running SimpleGAN on folding_winequality-red-4 /////////////////////////////////////////// Load 'data_input/folding_winequality-red-4' 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 Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 309, 1 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.006 LR average precision score: 0.149 -> test with 'GB' GB tn, fp: 307, 3 GB fn, tp: 10, 1 GB f1 score: 0.133 GB cohens kappa score: 0.117 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 309, 1 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.006 LR average precision score: 0.128 -> test with 'GB' GB tn, fp: 308, 2 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.011 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.177 -> test with 'GB' GB tn, fp: 306, 4 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.019 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 308, 2 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.011 LR average precision score: 0.081 -> test with 'GB' GB tn, fp: 307, 3 GB fn, tp: 10, 1 GB f1 score: 0.133 GB cohens kappa score: 0.117 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1196 synthetic samples -> test with 'LR' LR tn, fp: 306, 0 LR fn, tp: 9, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.220 -> test with 'GB' GB tn, fp: 306, 0 GB fn, tp: 8, 1 GB f1 score: 0.200 GB cohens kappa score: 0.195 -> test with 'KNN' KNN tn, fp: 306, 0 KNN fn, tp: 9, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ====== 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 1194 synthetic samples -> test with 'LR' LR tn, fp: 309, 1 LR fn, tp: 10, 1 LR f1 score: 0.154 LR cohens kappa score: 0.145 LR average precision score: 0.130 -> test with 'GB' GB tn, fp: 306, 4 GB fn, tp: 10, 1 GB f1 score: 0.125 GB cohens kappa score: 0.106 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.152 -> test with 'GB' GB tn, fp: 307, 3 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.015 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 308, 2 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.011 LR average precision score: 0.145 -> test with 'GB' GB tn, fp: 308, 2 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.011 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 10, 1 LR f1 score: 0.167 LR cohens kappa score: 0.162 LR average precision score: 0.306 -> test with 'GB' GB tn, fp: 310, 0 GB fn, tp: 10, 1 GB f1 score: 0.167 GB cohens kappa score: 0.162 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1196 synthetic samples -> test with 'LR' LR tn, fp: 306, 0 LR fn, tp: 9, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.077 -> test with 'GB' GB tn, fp: 299, 7 GB fn, tp: 8, 1 GB f1 score: 0.118 GB cohens kappa score: 0.093 -> test with 'KNN' KNN tn, fp: 306, 0 KNN fn, tp: 9, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ====== 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 1194 synthetic samples -> test with 'LR' LR tn, fp: 309, 1 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.006 LR average precision score: 0.168 -> test with 'GB' GB tn, fp: 307, 3 GB fn, tp: 10, 1 GB f1 score: 0.133 GB cohens kappa score: 0.117 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.171 -> test with 'GB' GB tn, fp: 310, 0 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: 0.000 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 308, 2 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.011 LR average precision score: 0.080 -> test with 'GB' GB tn, fp: 308, 2 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.011 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 309, 1 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.006 LR average precision score: 0.172 -> test with 'GB' GB tn, fp: 307, 3 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.015 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1196 synthetic samples -> test with 'LR' LR tn, fp: 306, 0 LR fn, tp: 9, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.121 -> test with 'GB' GB tn, fp: 306, 0 GB fn, tp: 9, 0 GB f1 score: 0.000 GB cohens kappa score: 0.000 -> test with 'KNN' KNN tn, fp: 306, 0 KNN fn, tp: 9, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ====== 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 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.315 -> test with 'GB' GB tn, fp: 308, 2 GB fn, tp: 9, 2 GB f1 score: 0.267 GB cohens kappa score: 0.253 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 309, 1 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.006 LR average precision score: 0.160 -> test with 'GB' GB tn, fp: 308, 2 GB fn, tp: 9, 2 GB f1 score: 0.267 GB cohens kappa score: 0.253 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.102 -> test with 'GB' GB tn, fp: 307, 3 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.015 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 307, 3 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.015 LR average precision score: 0.166 -> test with 'GB' GB tn, fp: 304, 6 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.025 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1196 synthetic samples -> test with 'LR' LR tn, fp: 306, 0 LR fn, tp: 9, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.047 -> test with 'GB' GB tn, fp: 305, 1 GB fn, tp: 9, 0 GB f1 score: 0.000 GB cohens kappa score: -0.006 -> test with 'KNN' KNN tn, fp: 306, 0 KNN fn, tp: 9, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ====== 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 1194 synthetic samples -> test with 'LR' LR tn, fp: 309, 1 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: -0.006 LR average precision score: 0.084 -> test with 'GB' GB tn, fp: 304, 6 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.025 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.099 -> test with 'GB' GB tn, fp: 308, 2 GB fn, tp: 10, 1 GB f1 score: 0.143 GB cohens kappa score: 0.130 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.201 -> test with 'GB' GB tn, fp: 308, 2 GB fn, tp: 10, 1 GB f1 score: 0.143 GB cohens kappa score: 0.130 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1194 synthetic samples -> test with 'LR' LR tn, fp: 310, 0 LR fn, tp: 11, 0 LR f1 score: 0.000 LR cohens kappa score: 0.000 LR average precision score: 0.159 -> test with 'GB' GB tn, fp: 310, 0 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: 0.000 -> test with 'KNN' KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples Epoch 1/3 Epoch 2/3 Epoch 3/3 -> create 1196 synthetic samples -> test with 'LR' LR tn, fp: 305, 1 LR fn, tp: 9, 0 LR f1 score: 0.000 LR cohens kappa score: -0.006 LR average precision score: 0.164 -> test with 'GB' GB tn, fp: 301, 5 GB fn, tp: 9, 0 GB f1 score: 0.000 GB cohens kappa score: -0.021 -> test with 'KNN' KNN tn, fp: 306, 0 KNN fn, tp: 9, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 310, 3 LR fn, tp: 11, 1 LR f1 score: 0.167 LR cohens kappa score: 0.162 LR average precision score: 0.315 average: LR tn, fp: 308.52, 0.68 LR fn, tp: 10.52, 0.08 LR f1 score: 0.013 LR cohens kappa score: 0.009 LR average precision score: 0.151 minimum: LR tn, fp: 305, 0 LR fn, tp: 9, 0 LR f1 score: 0.000 LR cohens kappa score: -0.015 LR average precision score: 0.047 -----[ GB ]----- maximum: GB tn, fp: 310, 7 GB fn, tp: 11, 2 GB f1 score: 0.267 GB cohens kappa score: 0.253 average: GB tn, fp: 306.6, 2.6 GB fn, tp: 10.08, 0.52 GB f1 score: 0.073 GB cohens kappa score: 0.060 minimum: GB tn, fp: 299, 0 GB fn, tp: 8, 0 GB f1 score: 0.000 GB cohens kappa score: -0.025 -----[ KNN ]----- maximum: KNN tn, fp: 310, 0 KNN fn, tp: 11, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 average: KNN tn, fp: 309.2, 0.0 KNN fn, tp: 10.6, 0.0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000 minimum: KNN tn, fp: 306, 0 KNN fn, tp: 9, 0 KNN f1 score: 0.000 KNN cohens kappa score: 0.000