/////////////////////////////////////////// // Running convGAN-majority-full on imblearn_mammography /////////////////////////////////////////// Load 'data_input/imblearn_mammography' from imblearn non empty cut in data_input/imblearn_mammography! (7 points) 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 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2122, 63 GAN fn, tp: 12, 40 GAN f1 score: 0.516 GAN cohens kappa score: 0.501 -> test with 'LR' LR tn, fp: 1916, 269 LR fn, tp: 5, 47 LR f1 score: 0.255 LR cohens kappa score: 0.224 LR average precision score: 0.611 -> test with 'GB' GB tn, fp: 2118, 67 GB fn, tp: 10, 42 GB f1 score: 0.522 GB cohens kappa score: 0.506 -> test with 'KNN' KNN tn, fp: 2119, 66 KNN fn, tp: 10, 42 KNN f1 score: 0.525 KNN cohens kappa score: 0.510 ------ Step 1/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2137, 48 GAN fn, tp: 13, 39 GAN f1 score: 0.561 GAN cohens kappa score: 0.548 -> test with 'LR' LR tn, fp: 1943, 242 LR fn, tp: 7, 45 LR f1 score: 0.265 LR cohens kappa score: 0.235 LR average precision score: 0.591 -> test with 'GB' GB tn, fp: 2119, 66 GB fn, tp: 10, 42 GB f1 score: 0.525 GB cohens kappa score: 0.510 -> test with 'KNN' KNN tn, fp: 2122, 63 KNN fn, tp: 10, 42 KNN f1 score: 0.535 KNN cohens kappa score: 0.520 ------ Step 1/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2136, 49 GAN fn, tp: 11, 41 GAN f1 score: 0.577 GAN cohens kappa score: 0.565 -> test with 'LR' LR tn, fp: 1928, 257 LR fn, tp: 7, 45 LR f1 score: 0.254 LR cohens kappa score: 0.223 LR average precision score: 0.664 -> test with 'GB' GB tn, fp: 2123, 62 GB fn, tp: 8, 44 GB f1 score: 0.557 GB cohens kappa score: 0.543 -> test with 'KNN' KNN tn, fp: 2128, 57 KNN fn, tp: 9, 43 KNN f1 score: 0.566 KNN cohens kappa score: 0.552 ------ Step 1/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2128, 57 GAN fn, tp: 8, 44 GAN f1 score: 0.575 GAN cohens kappa score: 0.562 -> test with 'LR' LR tn, fp: 1940, 245 LR fn, tp: 6, 46 LR f1 score: 0.268 LR cohens kappa score: 0.238 LR average precision score: 0.504 -> test with 'GB' GB tn, fp: 2121, 64 GB fn, tp: 10, 42 GB f1 score: 0.532 GB cohens kappa score: 0.517 -> test with 'KNN' KNN tn, fp: 2112, 73 KNN fn, tp: 9, 43 KNN f1 score: 0.512 KNN cohens kappa score: 0.496 ------ Step 1/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8532 synthetic samples -> test with GAN.predict GAN tn, fp: 2129, 54 GAN fn, tp: 9, 43 GAN f1 score: 0.577 GAN cohens kappa score: 0.564 -> test with 'LR' LR tn, fp: 1989, 194 LR fn, tp: 7, 45 LR f1 score: 0.309 LR cohens kappa score: 0.282 LR average precision score: 0.575 -> test with 'GB' GB tn, fp: 2128, 55 GB fn, tp: 10, 42 GB f1 score: 0.564 GB cohens kappa score: 0.550 -> test with 'KNN' KNN tn, fp: 1504, 679 KNN fn, tp: 11, 41 KNN f1 score: 0.106 KNN cohens kappa score: 0.066 ====== Step 2/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 2/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2107, 78 GAN fn, tp: 11, 41 GAN f1 score: 0.480 GAN cohens kappa score: 0.462 -> test with 'LR' LR tn, fp: 1905, 280 LR fn, tp: 6, 46 LR f1 score: 0.243 LR cohens kappa score: 0.212 LR average precision score: 0.587 -> test with 'GB' GB tn, fp: 2105, 80 GB fn, tp: 11, 41 GB f1 score: 0.474 GB cohens kappa score: 0.456 -> test with 'KNN' KNN tn, fp: 2106, 79 KNN fn, tp: 9, 43 KNN f1 score: 0.494 KNN cohens kappa score: 0.477 ------ Step 2/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2123, 62 GAN fn, tp: 10, 42 GAN f1 score: 0.538 GAN cohens kappa score: 0.524 -> test with 'LR' LR tn, fp: 1918, 267 LR fn, tp: 5, 47 LR f1 score: 0.257 LR cohens kappa score: 0.226 LR average precision score: 0.568 -> test with 'GB' GB tn, fp: 2096, 89 GB fn, tp: 8, 44 GB f1 score: 0.476 GB cohens kappa score: 0.458 -> test with 'KNN' KNN tn, fp: 2100, 85 KNN fn, tp: 9, 43 KNN f1 score: 0.478 KNN cohens kappa score: 0.460 ------ Step 2/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2148, 37 GAN fn, tp: 12, 40 GAN f1 score: 0.620 GAN cohens kappa score: 0.609 -> test with 'LR' LR tn, fp: 1965, 220 LR fn, tp: 7, 45 LR f1 score: 0.284 LR cohens kappa score: 0.255 LR average precision score: 0.582 -> test with 'GB' GB tn, fp: 2124, 61 GB fn, tp: 10, 42 GB f1 score: 0.542 GB cohens kappa score: 0.527 -> test with 'KNN' KNN tn, fp: 2120, 65 KNN fn, tp: 11, 41 KNN f1 score: 0.519 KNN cohens kappa score: 0.504 ------ Step 2/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2135, 50 GAN fn, tp: 9, 43 GAN f1 score: 0.593 GAN cohens kappa score: 0.581 -> test with 'LR' LR tn, fp: 1934, 251 LR fn, tp: 4, 48 LR f1 score: 0.274 LR cohens kappa score: 0.244 LR average precision score: 0.635 -> test with 'GB' GB tn, fp: 2123, 62 GB fn, tp: 6, 46 GB f1 score: 0.575 GB cohens kappa score: 0.561 -> test with 'KNN' KNN tn, fp: 2111, 74 KNN fn, tp: 5, 47 KNN f1 score: 0.543 KNN cohens kappa score: 0.528 ------ Step 2/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8532 synthetic samples -> test with GAN.predict GAN tn, fp: 2142, 41 GAN fn, tp: 16, 36 GAN f1 score: 0.558 GAN cohens kappa score: 0.546 -> test with 'LR' LR tn, fp: 1964, 219 LR fn, tp: 9, 43 LR f1 score: 0.274 LR cohens kappa score: 0.245 LR average precision score: 0.599 -> test with 'GB' GB tn, fp: 2130, 53 GB fn, tp: 12, 40 GB f1 score: 0.552 GB cohens kappa score: 0.538 -> test with 'KNN' KNN tn, fp: 2136, 47 KNN fn, tp: 13, 39 KNN f1 score: 0.565 KNN cohens kappa score: 0.552 ====== Step 3/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 3/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2133, 52 GAN fn, tp: 11, 41 GAN f1 score: 0.566 GAN cohens kappa score: 0.552 -> test with 'LR' LR tn, fp: 1988, 197 LR fn, tp: 5, 47 LR f1 score: 0.318 LR cohens kappa score: 0.290 LR average precision score: 0.688 -> test with 'GB' GB tn, fp: 2121, 64 GB fn, tp: 8, 44 GB f1 score: 0.550 GB cohens kappa score: 0.535 -> test with 'KNN' KNN tn, fp: 2127, 58 KNN fn, tp: 7, 45 KNN f1 score: 0.581 KNN cohens kappa score: 0.567 ------ Step 3/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2139, 46 GAN fn, tp: 12, 40 GAN f1 score: 0.580 GAN cohens kappa score: 0.567 -> test with 'LR' LR tn, fp: 1938, 247 LR fn, tp: 6, 46 LR f1 score: 0.267 LR cohens kappa score: 0.237 LR average precision score: 0.505 -> test with 'GB' GB tn, fp: 2125, 60 GB fn, tp: 12, 40 GB f1 score: 0.526 GB cohens kappa score: 0.511 -> test with 'KNN' KNN tn, fp: 1448, 737 KNN fn, tp: 10, 42 KNN f1 score: 0.101 KNN cohens kappa score: 0.060 ------ Step 3/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2148, 37 GAN fn, tp: 6, 46 GAN f1 score: 0.681 GAN cohens kappa score: 0.672 -> test with 'LR' LR tn, fp: 1948, 237 LR fn, tp: 3, 49 LR f1 score: 0.290 LR cohens kappa score: 0.261 LR average precision score: 0.608 -> test with 'GB' GB tn, fp: 2109, 76 GB fn, tp: 3, 49 GB f1 score: 0.554 GB cohens kappa score: 0.539 -> test with 'KNN' KNN tn, fp: 2121, 64 KNN fn, tp: 6, 46 KNN f1 score: 0.568 KNN cohens kappa score: 0.554 ------ Step 3/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2129, 56 GAN fn, tp: 13, 39 GAN f1 score: 0.531 GAN cohens kappa score: 0.516 -> test with 'LR' LR tn, fp: 1950, 235 LR fn, tp: 10, 42 LR f1 score: 0.255 LR cohens kappa score: 0.225 LR average precision score: 0.566 -> test with 'GB' GB tn, fp: 2114, 71 GB fn, tp: 13, 39 GB f1 score: 0.481 GB cohens kappa score: 0.465 -> test with 'KNN' KNN tn, fp: 2111, 74 KNN fn, tp: 11, 41 KNN f1 score: 0.491 KNN cohens kappa score: 0.474 ------ Step 3/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8532 synthetic samples -> test with GAN.predict GAN tn, fp: 2130, 53 GAN fn, tp: 17, 35 GAN f1 score: 0.500 GAN cohens kappa score: 0.485 -> test with 'LR' LR tn, fp: 1934, 249 LR fn, tp: 7, 45 LR f1 score: 0.260 LR cohens kappa score: 0.230 LR average precision score: 0.649 -> test with 'GB' GB tn, fp: 2121, 62 GB fn, tp: 10, 42 GB f1 score: 0.538 GB cohens kappa score: 0.524 -> test with 'KNN' KNN tn, fp: 2115, 68 KNN fn, tp: 11, 41 KNN f1 score: 0.509 KNN cohens kappa score: 0.493 ====== Step 4/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 4/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2144, 41 GAN fn, tp: 14, 38 GAN f1 score: 0.580 GAN cohens kappa score: 0.568 -> test with 'LR' LR tn, fp: 1957, 228 LR fn, tp: 8, 44 LR f1 score: 0.272 LR cohens kappa score: 0.242 LR average precision score: 0.631 -> test with 'GB' GB tn, fp: 2129, 56 GB fn, tp: 15, 37 GB f1 score: 0.510 GB cohens kappa score: 0.495 -> test with 'KNN' KNN tn, fp: 1493, 692 KNN fn, tp: 12, 40 KNN f1 score: 0.102 KNN cohens kappa score: 0.061 ------ Step 4/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2138, 47 GAN fn, tp: 11, 41 GAN f1 score: 0.586 GAN cohens kappa score: 0.573 -> test with 'LR' LR tn, fp: 1935, 250 LR fn, tp: 5, 47 LR f1 score: 0.269 LR cohens kappa score: 0.239 LR average precision score: 0.561 -> test with 'GB' GB tn, fp: 2108, 77 GB fn, tp: 7, 45 GB f1 score: 0.517 GB cohens kappa score: 0.501 -> test with 'KNN' KNN tn, fp: 1500, 685 KNN fn, tp: 7, 45 KNN f1 score: 0.115 KNN cohens kappa score: 0.075 ------ Step 4/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2138, 47 GAN fn, tp: 10, 42 GAN f1 score: 0.596 GAN cohens kappa score: 0.584 -> test with 'LR' LR tn, fp: 1955, 230 LR fn, tp: 7, 45 LR f1 score: 0.275 LR cohens kappa score: 0.246 LR average precision score: 0.687 -> test with 'GB' GB tn, fp: 2119, 66 GB fn, tp: 7, 45 GB f1 score: 0.552 GB cohens kappa score: 0.538 -> test with 'KNN' KNN tn, fp: 2114, 71 KNN fn, tp: 9, 43 KNN f1 score: 0.518 KNN cohens kappa score: 0.502 ------ Step 4/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2137, 48 GAN fn, tp: 10, 42 GAN f1 score: 0.592 GAN cohens kappa score: 0.579 -> test with 'LR' LR tn, fp: 1931, 254 LR fn, tp: 9, 43 LR f1 score: 0.246 LR cohens kappa score: 0.215 LR average precision score: 0.501 -> test with 'GB' GB tn, fp: 2117, 68 GB fn, tp: 10, 42 GB f1 score: 0.519 GB cohens kappa score: 0.503 -> test with 'KNN' KNN tn, fp: 2117, 68 KNN fn, tp: 12, 40 KNN f1 score: 0.500 KNN cohens kappa score: 0.484 ------ Step 4/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8532 synthetic samples -> test with GAN.predict GAN tn, fp: 2134, 49 GAN fn, tp: 9, 43 GAN f1 score: 0.597 GAN cohens kappa score: 0.585 -> test with 'LR' LR tn, fp: 1909, 274 LR fn, tp: 2, 50 LR f1 score: 0.266 LR cohens kappa score: 0.235 LR average precision score: 0.598 -> test with 'GB' GB tn, fp: 2119, 64 GB fn, tp: 6, 46 GB f1 score: 0.568 GB cohens kappa score: 0.554 -> test with 'KNN' KNN tn, fp: 2111, 72 KNN fn, tp: 10, 42 KNN f1 score: 0.506 KNN cohens kappa score: 0.490 ====== Step 5/5 ======= -> Shuffling data -> Spliting data to slices ------ Step 5/5: Slice 1/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2141, 44 GAN fn, tp: 13, 39 GAN f1 score: 0.578 GAN cohens kappa score: 0.565 -> test with 'LR' LR tn, fp: 1918, 267 LR fn, tp: 4, 48 LR f1 score: 0.262 LR cohens kappa score: 0.231 LR average precision score: 0.644 -> test with 'GB' GB tn, fp: 2127, 58 GB fn, tp: 5, 47 GB f1 score: 0.599 GB cohens kappa score: 0.586 -> test with 'KNN' KNN tn, fp: 2129, 56 KNN fn, tp: 9, 43 KNN f1 score: 0.570 KNN cohens kappa score: 0.556 ------ Step 5/5: Slice 2/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2116, 69 GAN fn, tp: 6, 46 GAN f1 score: 0.551 GAN cohens kappa score: 0.536 -> test with 'LR' LR tn, fp: 1926, 259 LR fn, tp: 5, 47 LR f1 score: 0.263 LR cohens kappa score: 0.232 LR average precision score: 0.550 -> test with 'GB' GB tn, fp: 2112, 73 GB fn, tp: 6, 46 GB f1 score: 0.538 GB cohens kappa score: 0.523 -> test with 'KNN' KNN tn, fp: 2116, 69 KNN fn, tp: 8, 44 KNN f1 score: 0.533 KNN cohens kappa score: 0.518 ------ Step 5/5: Slice 3/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2131, 54 GAN fn, tp: 16, 36 GAN f1 score: 0.507 GAN cohens kappa score: 0.492 -> test with 'LR' LR tn, fp: 1958, 227 LR fn, tp: 9, 43 LR f1 score: 0.267 LR cohens kappa score: 0.237 LR average precision score: 0.577 -> test with 'GB' GB tn, fp: 2126, 59 GB fn, tp: 13, 39 GB f1 score: 0.520 GB cohens kappa score: 0.505 -> test with 'KNN' KNN tn, fp: 2117, 68 KNN fn, tp: 13, 39 KNN f1 score: 0.491 KNN cohens kappa score: 0.474 ------ Step 5/5: Slice 4/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8530 synthetic samples -> test with GAN.predict GAN tn, fp: 2135, 50 GAN fn, tp: 10, 42 GAN f1 score: 0.583 GAN cohens kappa score: 0.571 -> test with 'LR' LR tn, fp: 1911, 274 LR fn, tp: 4, 48 LR f1 score: 0.257 LR cohens kappa score: 0.226 LR average precision score: 0.610 -> test with 'GB' GB tn, fp: 2109, 76 GB fn, tp: 8, 44 GB f1 score: 0.512 GB cohens kappa score: 0.495 -> test with 'KNN' KNN tn, fp: 2115, 70 KNN fn, tp: 9, 43 KNN f1 score: 0.521 KNN cohens kappa score: 0.505 ------ Step 5/5: Slice 5/5 ------- -> Reset the GAN -> Train generator for synthetic samples -> create 8532 synthetic samples -> test with GAN.predict GAN tn, fp: 2134, 49 GAN fn, tp: 13, 39 GAN f1 score: 0.557 GAN cohens kappa score: 0.544 -> test with 'LR' LR tn, fp: 1948, 235 LR fn, tp: 8, 44 LR f1 score: 0.266 LR cohens kappa score: 0.236 LR average precision score: 0.624 -> test with 'GB' GB tn, fp: 2119, 64 GB fn, tp: 15, 37 GB f1 score: 0.484 GB cohens kappa score: 0.467 -> test with 'KNN' KNN tn, fp: 2124, 59 KNN fn, tp: 12, 40 KNN f1 score: 0.530 KNN cohens kappa score: 0.515 ### Exercise is done. -----[ LR ]----- maximum: LR tn, fp: 1989, 280 LR fn, tp: 10, 50 LR f1 score: 0.318 LR cohens kappa score: 0.290 LR average precision score: 0.688 average: LR tn, fp: 1940.32, 244.28 LR fn, tp: 6.2, 45.8 LR f1 score: 0.269 LR cohens kappa score: 0.239 LR average precision score: 0.597 minimum: LR tn, fp: 1905, 194 LR fn, tp: 2, 42 LR f1 score: 0.243 LR cohens kappa score: 0.212 LR average precision score: 0.501 -----[ GB ]----- maximum: GB tn, fp: 2130, 89 GB fn, tp: 15, 49 GB f1 score: 0.599 GB cohens kappa score: 0.586 average: GB tn, fp: 2118.48, 66.12 GB fn, tp: 9.32, 42.68 GB f1 score: 0.531 GB cohens kappa score: 0.516 minimum: GB tn, fp: 2096, 53 GB fn, tp: 3, 37 GB f1 score: 0.474 GB cohens kappa score: 0.456 -----[ KNN ]----- maximum: KNN tn, fp: 2136, 737 KNN fn, tp: 13, 47 KNN f1 score: 0.581 KNN cohens kappa score: 0.567 average: KNN tn, fp: 2016.64, 167.96 KNN fn, tp: 9.68, 42.32 KNN f1 score: 0.459 KNN cohens kappa score: 0.440 minimum: KNN tn, fp: 1448, 47 KNN fn, tp: 5, 39 KNN f1 score: 0.101 KNN cohens kappa score: 0.060 -----[ GAN ]----- maximum: GAN tn, fp: 2148, 78 GAN fn, tp: 17, 46 GAN f1 score: 0.681 GAN cohens kappa score: 0.672 average: GAN tn, fp: 2133.36, 51.24 GAN fn, tp: 11.28, 40.72 GAN f1 score: 0.567 GAN cohens kappa score: 0.554 minimum: GAN tn, fp: 2107, 37 GAN fn, tp: 6, 35 GAN f1 score: 0.480 GAN cohens kappa score: 0.462