/////////////////////////////////////////// // Running convGAN-majority-5 on folding_yeast4 /////////////////////////////////////////// Load 'data_input/folding_yeast4' 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 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 259, 28 GAN fn, tp: 2, 9 GAN f1 score: 0.375 GAN cohens kappa score: 0.337 -> test with 'LR' LR tn, fp: 245, 42 LR fn, tp: 2, 9 LR f1 score: 0.290 LR cohens kappa score: 0.244 LR average precision score: 0.379 -> test with 'GB' GB tn, fp: 287, 0 GB fn, tp: 10, 1 GB f1 score: 0.167 GB cohens kappa score: 0.162 -> test with 'KNN' KNN tn, fp: 261, 26 KNN fn, tp: 3, 8 KNN f1 score: 0.356 KNN cohens kappa score: 0.317 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 241, 46 GAN fn, tp: 1, 10 GAN f1 score: 0.299 GAN cohens kappa score: 0.252 -> test with 'LR' LR tn, fp: 229, 58 LR fn, tp: 1, 10 LR f1 score: 0.253 LR cohens kappa score: 0.202 LR average precision score: 0.620 -> test with 'GB' GB tn, fp: 283, 4 GB fn, tp: 7, 4 GB f1 score: 0.421 GB cohens kappa score: 0.402 -> test with 'KNN' KNN tn, fp: 251, 36 KNN fn, tp: 0, 11 KNN f1 score: 0.379 KNN cohens kappa score: 0.340 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 240, 47 GAN fn, tp: 1, 10 GAN f1 score: 0.294 GAN cohens kappa score: 0.248 -> test with 'LR' LR tn, fp: 243, 44 LR fn, tp: 1, 10 LR f1 score: 0.308 LR cohens kappa score: 0.262 LR average precision score: 0.257 -> test with 'GB' GB tn, fp: 286, 1 GB fn, tp: 10, 1 GB f1 score: 0.154 GB cohens kappa score: 0.144 -> test with 'KNN' KNN tn, fp: 250, 37 KNN fn, tp: 3, 8 KNN f1 score: 0.286 KNN cohens kappa score: 0.241 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 258, 29 GAN fn, tp: 6, 5 GAN f1 score: 0.222 GAN cohens kappa score: 0.176 -> test with 'LR' LR tn, fp: 250, 37 LR fn, tp: 6, 5 LR f1 score: 0.189 LR cohens kappa score: 0.138 LR average precision score: 0.183 -> test with 'GB' GB tn, fp: 281, 6 GB fn, tp: 8, 3 GB f1 score: 0.300 GB cohens kappa score: 0.276 -> test with 'KNN' KNN tn, fp: 262, 25 KNN fn, tp: 5, 6 KNN f1 score: 0.286 KNN cohens kappa score: 0.245 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1104 synthetic samples -> test with GAN.predict GAN tn, fp: 240, 45 GAN fn, tp: 1, 6 GAN f1 score: 0.207 GAN cohens kappa score: 0.172 -> test with 'LR' LR tn, fp: 234, 51 LR fn, tp: 1, 6 LR f1 score: 0.188 LR cohens kappa score: 0.151 LR average precision score: 0.398 -> test with 'GB' GB tn, fp: 284, 1 GB fn, tp: 6, 1 GB f1 score: 0.222 GB cohens kappa score: 0.214 -> test with 'KNN' KNN tn, fp: 257, 28 KNN fn, tp: 1, 6 KNN f1 score: 0.293 KNN cohens kappa score: 0.263 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 261, 26 GAN fn, tp: 2, 9 GAN f1 score: 0.391 GAN cohens kappa score: 0.355 -> test with 'LR' LR tn, fp: 244, 43 LR fn, tp: 2, 9 LR f1 score: 0.286 LR cohens kappa score: 0.239 LR average precision score: 0.317 -> test with 'GB' GB tn, fp: 282, 5 GB fn, tp: 10, 1 GB f1 score: 0.118 GB cohens kappa score: 0.094 -> test with 'KNN' KNN tn, fp: 261, 26 KNN fn, tp: 3, 8 KNN f1 score: 0.356 KNN cohens kappa score: 0.317 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 257, 30 GAN fn, tp: 3, 8 GAN f1 score: 0.327 GAN cohens kappa score: 0.286 -> test with 'LR' LR tn, fp: 242, 45 LR fn, tp: 3, 8 LR f1 score: 0.250 LR cohens kappa score: 0.201 LR average precision score: 0.459 -> test with 'GB' GB tn, fp: 285, 2 GB fn, tp: 7, 4 GB f1 score: 0.471 GB cohens kappa score: 0.456 -> test with 'KNN' KNN tn, fp: 239, 48 KNN fn, tp: 3, 8 KNN f1 score: 0.239 KNN cohens kappa score: 0.189 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 257, 30 GAN fn, tp: 4, 7 GAN f1 score: 0.292 GAN cohens kappa score: 0.249 -> test with 'LR' LR tn, fp: 244, 43 LR fn, tp: 4, 7 LR f1 score: 0.230 LR cohens kappa score: 0.180 LR average precision score: 0.376 -> test with 'GB' GB tn, fp: 284, 3 GB fn, tp: 8, 3 GB f1 score: 0.353 GB cohens kappa score: 0.336 -> test with 'KNN' KNN tn, fp: 252, 35 KNN fn, tp: 2, 9 KNN f1 score: 0.327 KNN cohens kappa score: 0.285 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 248, 39 GAN fn, tp: 2, 9 GAN f1 score: 0.305 GAN cohens kappa score: 0.261 -> test with 'LR' LR tn, fp: 243, 44 LR fn, tp: 2, 9 LR f1 score: 0.281 LR cohens kappa score: 0.234 LR average precision score: 0.313 -> test with 'GB' GB tn, fp: 285, 2 GB fn, tp: 9, 2 GB f1 score: 0.267 GB cohens kappa score: 0.252 -> test with 'KNN' KNN tn, fp: 263, 24 KNN fn, tp: 3, 8 KNN f1 score: 0.372 KNN cohens kappa score: 0.336 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1104 synthetic samples -> test with GAN.predict GAN tn, fp: 239, 46 GAN fn, tp: 1, 6 GAN f1 score: 0.203 GAN cohens kappa score: 0.168 -> test with 'LR' LR tn, fp: 229, 56 LR fn, tp: 1, 6 LR f1 score: 0.174 LR cohens kappa score: 0.137 LR average precision score: 0.392 -> test with 'GB' GB tn, fp: 284, 1 GB fn, tp: 6, 1 GB f1 score: 0.222 GB cohens kappa score: 0.214 -> test with 'KNN' KNN tn, fp: 261, 24 KNN fn, tp: 1, 6 KNN f1 score: 0.324 KNN cohens kappa score: 0.297 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 258, 29 GAN fn, tp: 3, 8 GAN f1 score: 0.333 GAN cohens kappa score: 0.293 -> test with 'LR' LR tn, fp: 241, 46 LR fn, tp: 2, 9 LR f1 score: 0.273 LR cohens kappa score: 0.225 LR average precision score: 0.456 -> test with 'GB' GB tn, fp: 283, 4 GB fn, tp: 10, 1 GB f1 score: 0.125 GB cohens kappa score: 0.104 -> test with 'KNN' KNN tn, fp: 263, 24 KNN fn, tp: 3, 8 KNN f1 score: 0.372 KNN cohens kappa score: 0.336 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 260, 27 GAN fn, tp: 2, 9 GAN f1 score: 0.383 GAN cohens kappa score: 0.346 -> test with 'LR' LR tn, fp: 240, 47 LR fn, tp: 2, 9 LR f1 score: 0.269 LR cohens kappa score: 0.221 LR average precision score: 0.406 -> test with 'GB' GB tn, fp: 285, 2 GB fn, tp: 10, 1 GB f1 score: 0.143 GB cohens kappa score: 0.129 -> test with 'KNN' KNN tn, fp: 260, 27 KNN fn, tp: 1, 10 KNN f1 score: 0.417 KNN cohens kappa score: 0.381 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 248, 39 GAN fn, tp: 4, 7 GAN f1 score: 0.246 GAN cohens kappa score: 0.198 -> test with 'LR' LR tn, fp: 240, 47 LR fn, tp: 3, 8 LR f1 score: 0.242 LR cohens kappa score: 0.193 LR average precision score: 0.240 -> test with 'GB' GB tn, fp: 283, 4 GB fn, tp: 10, 1 GB f1 score: 0.125 GB cohens kappa score: 0.104 -> test with 'KNN' KNN tn, fp: 256, 31 KNN fn, tp: 3, 8 KNN f1 score: 0.320 KNN cohens kappa score: 0.278 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 254, 33 GAN fn, tp: 3, 8 GAN f1 score: 0.308 GAN cohens kappa score: 0.265 -> test with 'LR' LR tn, fp: 241, 46 LR fn, tp: 3, 8 LR f1 score: 0.246 LR cohens kappa score: 0.197 LR average precision score: 0.513 -> test with 'GB' GB tn, fp: 284, 3 GB fn, tp: 9, 2 GB f1 score: 0.250 GB cohens kappa score: 0.232 -> test with 'KNN' KNN tn, fp: 260, 27 KNN fn, tp: 5, 6 KNN f1 score: 0.273 KNN cohens kappa score: 0.230 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1104 synthetic samples -> test with GAN.predict GAN tn, fp: 254, 31 GAN fn, tp: 3, 4 GAN f1 score: 0.190 GAN cohens kappa score: 0.157 -> test with 'LR' LR tn, fp: 245, 40 LR fn, tp: 2, 5 LR f1 score: 0.192 LR cohens kappa score: 0.157 LR average precision score: 0.392 -> test with 'GB' GB tn, fp: 284, 1 GB fn, tp: 4, 3 GB f1 score: 0.545 GB cohens kappa score: 0.537 -> test with 'KNN' KNN tn, fp: 253, 32 KNN fn, tp: 1, 6 KNN f1 score: 0.267 KNN cohens kappa score: 0.236 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 263, 24 GAN fn, tp: 5, 6 GAN f1 score: 0.293 GAN cohens kappa score: 0.252 -> test with 'LR' LR tn, fp: 252, 35 LR fn, tp: 4, 7 LR f1 score: 0.264 LR cohens kappa score: 0.218 LR average precision score: 0.456 -> test with 'GB' GB tn, fp: 286, 1 GB fn, tp: 9, 2 GB f1 score: 0.286 GB cohens kappa score: 0.274 -> test with 'KNN' KNN tn, fp: 263, 24 KNN fn, tp: 6, 5 KNN f1 score: 0.250 KNN cohens kappa score: 0.208 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 246, 41 GAN fn, tp: 2, 9 GAN f1 score: 0.295 GAN cohens kappa score: 0.250 -> test with 'LR' LR tn, fp: 239, 48 LR fn, tp: 2, 9 LR f1 score: 0.265 LR cohens kappa score: 0.216 LR average precision score: 0.289 -> test with 'GB' GB tn, fp: 283, 4 GB fn, tp: 9, 2 GB f1 score: 0.235 GB cohens kappa score: 0.215 -> test with 'KNN' KNN tn, fp: 255, 32 KNN fn, tp: 3, 8 KNN f1 score: 0.314 KNN cohens kappa score: 0.272 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 245, 42 GAN fn, tp: 2, 9 GAN f1 score: 0.290 GAN cohens kappa score: 0.244 -> test with 'LR' LR tn, fp: 230, 57 LR fn, tp: 2, 9 LR f1 score: 0.234 LR cohens kappa score: 0.182 LR average precision score: 0.237 -> test with 'GB' GB tn, fp: 282, 5 GB fn, tp: 10, 1 GB f1 score: 0.118 GB cohens kappa score: 0.094 -> test with 'KNN' KNN tn, fp: 250, 37 KNN fn, tp: 1, 10 KNN f1 score: 0.345 KNN cohens kappa score: 0.303 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 252, 35 GAN fn, tp: 3, 8 GAN f1 score: 0.296 GAN cohens kappa score: 0.252 -> test with 'LR' LR tn, fp: 245, 42 LR fn, tp: 3, 8 LR f1 score: 0.262 LR cohens kappa score: 0.215 LR average precision score: 0.294 -> test with 'GB' GB tn, fp: 283, 4 GB fn, tp: 9, 2 GB f1 score: 0.235 GB cohens kappa score: 0.215 -> test with 'KNN' KNN tn, fp: 255, 32 KNN fn, tp: 4, 7 KNN f1 score: 0.280 KNN cohens kappa score: 0.236 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1104 synthetic samples -> test with GAN.predict GAN tn, fp: 255, 30 GAN fn, tp: 2, 5 GAN f1 score: 0.238 GAN cohens kappa score: 0.206 -> test with 'LR' LR tn, fp: 249, 36 LR fn, tp: 2, 5 LR f1 score: 0.208 LR cohens kappa score: 0.175 LR average precision score: 0.540 -> test with 'GB' GB tn, fp: 282, 3 GB fn, tp: 4, 3 GB f1 score: 0.462 GB cohens kappa score: 0.449 -> test with 'KNN' KNN tn, fp: 255, 30 KNN fn, tp: 2, 5 KNN f1 score: 0.238 KNN cohens kappa score: 0.206 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 254, 33 GAN fn, tp: 3, 8 GAN f1 score: 0.308 GAN cohens kappa score: 0.265 -> test with 'LR' LR tn, fp: 248, 39 LR fn, tp: 3, 8 LR f1 score: 0.276 LR cohens kappa score: 0.230 LR average precision score: 0.205 -> test with 'GB' GB tn, fp: 284, 3 GB fn, tp: 11, 0 GB f1 score: 0.000 GB cohens kappa score: -0.016 -> test with 'KNN' KNN tn, fp: 260, 27 KNN fn, tp: 1, 10 KNN f1 score: 0.417 KNN cohens kappa score: 0.381 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 252, 35 GAN fn, tp: 2, 9 GAN f1 score: 0.327 GAN cohens kappa score: 0.285 -> test with 'LR' LR tn, fp: 229, 58 LR fn, tp: 2, 9 LR f1 score: 0.231 LR cohens kappa score: 0.179 LR average precision score: 0.519 -> test with 'GB' GB tn, fp: 286, 1 GB fn, tp: 9, 2 GB f1 score: 0.286 GB cohens kappa score: 0.274 -> test with 'KNN' KNN tn, fp: 251, 36 KNN fn, tp: 1, 10 KNN f1 score: 0.351 KNN cohens kappa score: 0.310 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 257, 30 GAN fn, tp: 3, 8 GAN f1 score: 0.327 GAN cohens kappa score: 0.286 -> test with 'LR' LR tn, fp: 237, 50 LR fn, tp: 3, 8 LR f1 score: 0.232 LR cohens kappa score: 0.181 LR average precision score: 0.539 -> test with 'GB' GB tn, fp: 287, 0 GB fn, tp: 9, 2 GB f1 score: 0.308 GB cohens kappa score: 0.300 -> test with 'KNN' KNN tn, fp: 258, 29 KNN fn, tp: 4, 7 KNN f1 score: 0.298 KNN cohens kappa score: 0.256 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1106 synthetic samples -> test with GAN.predict GAN tn, fp: 254, 33 GAN fn, tp: 1, 10 GAN f1 score: 0.370 GAN cohens kappa score: 0.331 -> test with 'LR' LR tn, fp: 244, 43 LR fn, tp: 1, 10 LR f1 score: 0.312 LR cohens kappa score: 0.268 LR average precision score: 0.525 -> test with 'GB' GB tn, fp: 281, 6 GB fn, tp: 9, 2 GB f1 score: 0.211 GB cohens kappa score: 0.185 -> test with 'KNN' KNN tn, fp: 258, 29 KNN fn, tp: 4, 7 KNN f1 score: 0.298 KNN cohens kappa score: 0.256 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 1104 synthetic samples -> test with GAN.predict GAN tn, fp: 245, 40 GAN fn, tp: 2, 5 GAN f1 score: 0.192 GAN cohens kappa score: 0.157 -> test with 'LR' LR tn, fp: 233, 52 LR fn, tp: 2, 5 LR f1 score: 0.156 LR cohens kappa score: 0.119 LR average precision score: 0.133 -> test with 'GB' GB tn, fp: 281, 4 GB fn, tp: 5, 2 GB f1 score: 0.308 GB cohens kappa score: 0.292 -> test with 'KNN' KNN tn, fp: 264, 21 KNN fn, tp: 1, 6 KNN f1 score: 0.353 KNN cohens kappa score: 0.327 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 252, 58 LR fn, tp: 6, 10 LR f1 score: 0.312 LR cohens kappa score: 0.268 LR average precision score: 0.620 average: LR tn, fp: 240.64, 45.96 LR fn, tp: 2.36, 7.84 LR f1 score: 0.244 LR cohens kappa score: 0.199 LR average precision score: 0.378 minimum: LR tn, fp: 229, 35 LR fn, tp: 1, 5 LR f1 score: 0.156 LR cohens kappa score: 0.119 LR average precision score: 0.133 -----[ GB ]----- maximum: GB tn, fp: 287, 6 GB fn, tp: 11, 4 GB f1 score: 0.545 GB cohens kappa score: 0.537 average: GB tn, fp: 283.8, 2.8 GB fn, tp: 8.32, 1.88 GB f1 score: 0.253 GB cohens kappa score: 0.238 minimum: GB tn, fp: 281, 0 GB fn, tp: 4, 0 GB f1 score: 0.000 GB cohens kappa score: -0.016 -----[ KNN ]----- maximum: KNN tn, fp: 264, 48 KNN fn, tp: 6, 11 KNN f1 score: 0.417 KNN cohens kappa score: 0.381 average: KNN tn, fp: 256.72, 29.88 KNN fn, tp: 2.56, 7.64 KNN f1 score: 0.320 KNN cohens kappa score: 0.282 minimum: KNN tn, fp: 239, 21 KNN fn, tp: 0, 5 KNN f1 score: 0.238 KNN cohens kappa score: 0.189 -----[ GAN ]----- maximum: GAN tn, fp: 263, 47 GAN fn, tp: 6, 10 GAN f1 score: 0.391 GAN cohens kappa score: 0.355 average: GAN tn, fp: 251.88, 34.72 GAN fn, tp: 2.52, 7.68 GAN f1 score: 0.292 GAN cohens kappa score: 0.252 minimum: GAN tn, fp: 239, 24 GAN fn, tp: 1, 4 GAN f1 score: 0.190 GAN cohens kappa score: 0.157