Archive 4 checkpoints (1-4)

paper-releases/pbisc-diffusion/updates/20261003T103040Z-crosscohort-k1000-saturation/diffusion/dataset/server_1/model8/plans/paired_crosscohort_20261002/runs/hoL_giroux2022/K1000/lora64sat/milestones.json

@@ -0,0 +1,42 @@
+[
+ {
+  "milestone_step": 20000,
+  "selected_step": 20000,
+  "selected_inner_val_loss": 0.10309372149934672,
+  "block_improvement": 0.00940420036439342,
+  "total_improvement": 0.00940420036439342,
+  "reason": "block_end",
+  "checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_giroux2022/K1000/lora64sat/checkpoint_ft_s20000.pt",
+  "checkpoint_sha256": "7f4adbbd2ab4dc4af24aef53a491e6b35ea4f66ed7f8ad4f62656ec46411efbf"
+ },
+ {
+  "milestone_step": 40000,
+  "selected_step": 40000,
+  "selected_inner_val_loss": 0.10284607811369981,
+  "block_improvement": 0.000247643385646909,
+  "total_improvement": 0.009651843750040329,
+  "reason": "block_end",
+  "checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_giroux2022/K1000/lora64sat/checkpoint_ft_s40000.pt",
+  "checkpoint_sha256": "220941d33a105c198a89d311b6b61b1747199d1fb33340ae5411b1a6e36f49d4"
+ },
+ {
+  "milestone_step": 60000,
+  "selected_step": 59500,
+  "selected_inner_val_loss": 0.10267554886987187,
+  "block_improvement": 0.00017052924382793266,
+  "total_improvement": 0.009822372993868261,
+  "reason": "block_end",
+  "checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_giroux2022/K1000/lora64sat/checkpoint_ft_s60000.pt",
+  "checkpoint_sha256": "2766d52ee1068e260f8ae8f2a717f17dcaa652661b6384e5c619f2d85e7076a3"
+ },
+ {
+  "milestone_step": 69250,
+  "selected_step": 67250,
+  "selected_inner_val_loss": 0.10262443648058582,
+  "block_improvement": 5.111238928605377e-05,
+  "total_improvement": 0.009873485383154315,
+  "reason": "early_stop",
+  "checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_giroux2022/K1000/lora64sat/checkpoint_ft_s69250.pt",
+  "checkpoint_sha256": "a17c27b305a62b582d7b68eff0dadfb6c39588ba42b7791b9b6d06bd8d6dd37a"
+ }
+]

paper-releases/pbisc-diffusion/updates/20261003T103040Z-crosscohort-k1000-saturation/diffusion/dataset/server_1/model8/plans/paired_crosscohort_20261002/runs/hoL_giroux2022/K1000/lora64sat/train.log

@@ -0,0 +1,284 @@
+2026-10-03T05:08:29 split=hoL_giroux2022 K=1000 mode=lora trainable=5,184,896 balance=cohort immicro: train 95 rows / 347,061 cells, inner 12; ye2026_sepsis: train 73 rows / 529,816 cells, inner 9; imperial_eskd: train 146 rows / 462,172 cells, inner 11; liu2021: train 40 rows / 147,532 cells, inner 4
+2026-10-03T05:08:31 step 0 inner-val loss 0.11250 (pretrained) {'immicro': 0.10636159498244524, 'ye2026_sepsis': 0.12489082664251328, 'imperial_eskd': 0.10889218679883263, 'liu2021': 0.10984707903116941}
+2026-10-03T05:08:43 step 250 train 0.10608 inner-val 0.11153 best 0.11153@250 (12s) immicro=0.1055 ye2026_sepsis=0.1239 imperial_eskd=0.1078 liu2021=0.1089
+2026-10-03T05:08:56 step 500 train 0.10373 inner-val 0.10973 best 0.10973@500 (25s) immicro=0.1040 ye2026_sepsis=0.1220 imperial_eskd=0.1057 liu2021=0.1072
+2026-10-03T05:09:08 step 750 train 0.10381 inner-val 0.10783 best 0.10783@750 (37s) immicro=0.1026 ye2026_sepsis=0.1196 imperial_eskd=0.1035 liu2021=0.1056
+2026-10-03T05:09:21 step 1000 train 0.10334 inner-val 0.10629 best 0.10629@1000 (50s) immicro=0.1015 ye2026_sepsis=0.1175 imperial_eskd=0.1018 liu2021=0.1044
+2026-10-03T05:09:34 step 1250 train 0.10349 inner-val 0.10532 best 0.10532@1250 (63s) immicro=0.1007 ye2026_sepsis=0.1160 imperial_eskd=0.1009 liu2021=0.1036
+2026-10-03T05:09:46 step 1500 train 0.10313 inner-val 0.10472 best 0.10472@1500 (75s) immicro=0.1002 ye2026_sepsis=0.1152 imperial_eskd=0.1004 liu2021=0.1031
+2026-10-03T05:09:58 step 1750 train 0.10313 inner-val 0.10435 best 0.10435@1750 (87s) immicro=0.0998 ye2026_sepsis=0.1147 imperial_eskd=0.1001 liu2021=0.1028
+2026-10-03T05:10:09 step 2000 train 0.10297 inner-val 0.10410 best 0.10410@2000 (98s) immicro=0.0995 ye2026_sepsis=0.1144 imperial_eskd=0.0999 liu2021=0.1027
+2026-10-03T05:10:20 step 2250 train 0.10300 inner-val 0.10393 best 0.10393@2250 (109s) immicro=0.0992 ye2026_sepsis=0.1142 imperial_eskd=0.0997 liu2021=0.1025
+2026-10-03T05:10:31 step 2500 train 0.10314 inner-val 0.10381 best 0.10381@2500 (120s) immicro=0.0991 ye2026_sepsis=0.1141 imperial_eskd=0.0996 liu2021=0.1025
+2026-10-03T05:10:43 step 2750 train 0.10307 inner-val 0.10372 best 0.10372@2750 (132s) immicro=0.0990 ye2026_sepsis=0.1140 imperial_eskd=0.0995 liu2021=0.1024
+2026-10-03T05:10:56 step 3000 train 0.10306 inner-val 0.10365 best 0.10365@3000 (145s) immicro=0.0989 ye2026_sepsis=0.1139 imperial_eskd=0.0995 liu2021=0.1024
+2026-10-03T05:11:08 step 3250 train 0.10306 inner-val 0.10360 best 0.10360@3250 (157s) immicro=0.0988 ye2026_sepsis=0.1139 imperial_eskd=0.0994 liu2021=0.1023
+2026-10-03T05:11:19 step 3500 train 0.10305 inner-val 0.10356 best 0.10356@3500 (168s) immicro=0.0988 ye2026_sepsis=0.1138 imperial_eskd=0.0994 liu2021=0.1023
+2026-10-03T05:11:30 step 3750 train 0.10291 inner-val 0.10353 best 0.10353@3750 (179s) immicro=0.0987 ye2026_sepsis=0.1138 imperial_eskd=0.0993 liu2021=0.1023
+2026-10-03T05:11:41 step 4000 train 0.10273 inner-val 0.10351 best 0.10351@4000 (190s) immicro=0.0987 ye2026_sepsis=0.1138 imperial_eskd=0.0993 liu2021=0.1023
+2026-10-03T05:11:52 step 4250 train 0.10312 inner-val 0.10349 best 0.10349@4250 (201s) immicro=0.0986 ye2026_sepsis=0.1138 imperial_eskd=0.0993 liu2021=0.1022
+2026-10-03T05:12:03 step 4500 train 0.10296 inner-val 0.10348 best 0.10348@4500 (212s) immicro=0.0986 ye2026_sepsis=0.1138 imperial_eskd=0.0993 liu2021=0.1022
+2026-10-03T05:12:13 step 4750 train 0.10292 inner-val 0.10346 best 0.10346@4750 (222s) immicro=0.0986 ye2026_sepsis=0.1137 imperial_eskd=0.0993 liu2021=0.1022
+2026-10-03T05:12:24 step 5000 train 0.10314 inner-val 0.10344 best 0.10344@5000 (233s) immicro=0.0986 ye2026_sepsis=0.1137 imperial_eskd=0.0992 liu2021=0.1022
+2026-10-03T05:12:35 step 5250 train 0.10275 inner-val 0.10343 best 0.10343@5250 (244s) immicro=0.0986 ye2026_sepsis=0.1137 imperial_eskd=0.0992 liu2021=0.1022
+2026-10-03T05:12:46 step 5500 train 0.10284 inner-val 0.10342 best 0.10342@5500 (255s) immicro=0.0985 ye2026_sepsis=0.1137 imperial_eskd=0.0992 liu2021=0.1022
+2026-10-03T05:12:57 step 5750 train 0.10242 inner-val 0.10341 best 0.10341@5750 (266s) immicro=0.0985 ye2026_sepsis=0.1137 imperial_eskd=0.0992 liu2021=0.1022
+2026-10-03T05:13:08 step 6000 train 0.10262 inner-val 0.10340 best 0.10340@6000 (277s) immicro=0.0985 ye2026_sepsis=0.1137 imperial_eskd=0.0992 liu2021=0.1022
+2026-10-03T05:13:18 step 6250 train 0.10287 inner-val 0.10339 best 0.10339@6250 (287s) immicro=0.0985 ye2026_sepsis=0.1137 imperial_eskd=0.0992 liu2021=0.1022
+2026-10-03T05:13:29 step 6500 train 0.10292 inner-val 0.10338 best 0.10338@6500 (298s) immicro=0.0985 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1022
+2026-10-03T05:13:40 step 6750 train 0.10253 inner-val 0.10337 best 0.10337@6750 (309s) immicro=0.0985 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1022
+2026-10-03T05:13:51 step 7000 train 0.10273 inner-val 0.10336 best 0.10336@7000 (320s) immicro=0.0985 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:14:02 step 7250 train 0.10299 inner-val 0.10335 best 0.10335@7250 (331s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:14:13 step 7500 train 0.10277 inner-val 0.10334 best 0.10334@7500 (342s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:14:23 step 7750 train 0.10263 inner-val 0.10334 best 0.10334@7750 (352s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:14:34 step 8000 train 0.10248 inner-val 0.10333 best 0.10333@8000 (363s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:14:45 step 8250 train 0.10281 inner-val 0.10333 best 0.10333@8250 (374s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:14:56 step 8500 train 0.10245 inner-val 0.10332 best 0.10332@8500 (385s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:15:07 step 8750 train 0.10276 inner-val 0.10331 best 0.10331@8750 (396s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:15:18 step 9000 train 0.10268 inner-val 0.10331 best 0.10331@9000 (407s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:15:29 step 9250 train 0.10257 inner-val 0.10330 best 0.10330@9250 (418s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:15:40 step 9500 train 0.10266 inner-val 0.10330 best 0.10330@9500 (429s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:15:50 step 9750 train 0.10258 inner-val 0.10329 best 0.10329@9750 (439s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:16:01 step 10000 train 0.10240 inner-val 0.10329 best 0.10329@10000 (450s) immicro=0.0984 ye2026_sepsis=0.1137 imperial_eskd=0.0991 liu2021=0.1021
+2026-10-03T05:16:12 step 10250 train 0.10257 inner-val 0.10328 best 0.10328@10250 (461s) immicro=0.0983 ye2026_sepsis=0.1137 imperial_eskd=0.0990 liu2021=0.1021
+2026-10-03T05:16:23 step 10500 train 0.10258 inner-val 0.10327 best 0.10327@10500 (472s) immicro=0.0983 ye2026_sepsis=0.1137 imperial_eskd=0.0990 liu2021=0.1021
+2026-10-03T05:16:34 step 10750 train 0.10268 inner-val 0.10327 best 0.10327@10750 (483s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1021
+2026-10-03T05:16:44 step 11000 train 0.10282 inner-val 0.10326 best 0.10326@11000 (494s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1021
+2026-10-03T05:16:55 step 11250 train 0.10247 inner-val 0.10325 best 0.10325@11250 (504s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1021
+2026-10-03T05:17:06 step 11500 train 0.10274 inner-val 0.10325 best 0.10325@11500 (515s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:17:17 step 11750 train 0.10249 inner-val 0.10325 best 0.10325@11750 (526s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:17:28 step 12000 train 0.10229 inner-val 0.10324 best 0.10324@12000 (537s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:17:39 step 12250 train 0.10238 inner-val 0.10324 best 0.10324@12250 (548s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:17:49 step 12500 train 0.10266 inner-val 0.10323 best 0.10323@12500 (558s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:18:00 step 12750 train 0.10251 inner-val 0.10323 best 0.10323@12750 (569s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:18:11 step 13000 train 0.10272 inner-val 0.10322 best 0.10322@13000 (580s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:18:22 step 13250 train 0.10268 inner-val 0.10321 best 0.10321@13250 (591s) immicro=0.0983 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:18:33 step 13500 train 0.10261 inner-val 0.10321 best 0.10321@13500 (602s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:18:44 step 13750 train 0.10264 inner-val 0.10320 best 0.10320@13750 (613s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:18:55 step 14000 train 0.10263 inner-val 0.10320 best 0.10320@14000 (624s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:19:05 step 14250 train 0.10230 inner-val 0.10319 best 0.10319@14250 (634s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:19:16 step 14500 train 0.10262 inner-val 0.10319 best 0.10319@14500 (645s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:19:27 step 14750 train 0.10250 inner-val 0.10319 best 0.10319@14750 (656s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:19:38 step 15000 train 0.10255 inner-val 0.10319 best 0.10319@15000 (667s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:19:49 step 15250 train 0.10237 inner-val 0.10318 best 0.10318@15250 (678s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:20:00 step 15500 train 0.10259 inner-val 0.10318 best 0.10318@15500 (689s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:20:11 step 15750 train 0.10231 inner-val 0.10317 best 0.10317@15750 (700s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:20:22 step 16000 train 0.10229 inner-val 0.10317 best 0.10317@16000 (711s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:20:32 step 16250 train 0.10258 inner-val 0.10316 best 0.10316@16250 (722s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1020
+2026-10-03T05:20:43 step 16500 train 0.10252 inner-val 0.10316 best 0.10316@16500 (732s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1019
+2026-10-03T05:20:54 step 16750 train 0.10219 inner-val 0.10315 best 0.10315@16750 (743s) immicro=0.0982 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1019
+2026-10-03T05:21:05 step 17000 train 0.10256 inner-val 0.10315 best 0.10315@17000 (754s) immicro=0.0981 ye2026_sepsis=0.1136 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:21:16 step 17250 train 0.10250 inner-val 0.10314 best 0.10314@17250 (765s) immicro=0.0981 ye2026_sepsis=0.1136 imperial_eskd=0.0990 liu2021=0.1019
+2026-10-03T05:21:27 step 17500 train 0.10229 inner-val 0.10314 best 0.10314@17500 (776s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:21:38 step 17750 train 0.10214 inner-val 0.10313 best 0.10313@17750 (787s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:21:48 step 18000 train 0.10281 inner-val 0.10313 best 0.10313@18000 (797s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:21:59 step 18250 train 0.10208 inner-val 0.10312 best 0.10312@18250 (808s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:22:10 step 18500 train 0.10251 inner-val 0.10312 best 0.10312@18500 (819s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:22:21 step 18750 train 0.10218 inner-val 0.10311 best 0.10311@18750 (830s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:22:32 step 19000 train 0.10232 inner-val 0.10311 best 0.10311@19000 (841s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:22:42 step 19250 train 0.10236 inner-val 0.10310 best 0.10310@19250 (851s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:22:53 step 19500 train 0.10228 inner-val 0.10310 best 0.10310@19500 (862s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:23:04 step 19750 train 0.10249 inner-val 0.10310 best 0.10310@19750 (873s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:23:15 step 20000 train 0.10244 inner-val 0.10309 best 0.10309@20000 (884s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:23:17 block end 20000: best 0.10309@20000; block improvement 0.009404 (100.00 % of total 0.009404) -> continue
+2026-10-03T05:23:28 step 20250 train 0.10217 inner-val 0.10309 best 0.10309@20250 (897s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:23:39 step 20500 train 0.10220 inner-val 0.10309 best 0.10309@20500 (908s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:23:50 step 20750 train 0.10235 inner-val 0.10308 best 0.10308@20750 (919s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:24:00 step 21000 train 0.10223 inner-val 0.10308 best 0.10308@21000 (929s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:24:11 step 21250 train 0.10228 inner-val 0.10308 best 0.10308@21000 (940s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:24:22 step 21500 train 0.10205 inner-val 0.10307 best 0.10307@21500 (951s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:24:33 step 21750 train 0.10244 inner-val 0.10307 best 0.10307@21750 (962s) immicro=0.0981 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:24:44 step 22000 train 0.10245 inner-val 0.10307 best 0.10307@22000 (973s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:24:55 step 22250 train 0.10228 inner-val 0.10307 best 0.10307@22250 (984s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:25:06 step 22500 train 0.10223 inner-val 0.10307 best 0.10307@22250 (995s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:25:17 step 22750 train 0.10210 inner-val 0.10306 best 0.10306@22750 (1006s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1019
+2026-10-03T05:25:28 step 23000 train 0.10264 inner-val 0.10305 best 0.10305@23000 (1017s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:25:39 step 23250 train 0.10259 inner-val 0.10305 best 0.10305@23250 (1028s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:25:49 step 23500 train 0.10224 inner-val 0.10305 best 0.10305@23500 (1038s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:26:00 step 23750 train 0.10220 inner-val 0.10304 best 0.10304@23750 (1049s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:26:11 step 24000 train 0.10227 inner-val 0.10304 best 0.10304@24000 (1060s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:26:22 step 24250 train 0.10202 inner-val 0.10304 best 0.10304@24000 (1071s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:26:32 step 24500 train 0.10208 inner-val 0.10304 best 0.10304@24000 (1082s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:26:43 step 24750 train 0.10216 inner-val 0.10304 best 0.10304@24750 (1092s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:26:54 step 25000 train 0.10191 inner-val 0.10303 best 0.10303@25000 (1103s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:27:05 step 25250 train 0.10201 inner-val 0.10303 best 0.10303@25250 (1114s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:27:16 step 25500 train 0.10228 inner-val 0.10303 best 0.10303@25500 (1125s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:27:26 step 25750 train 0.10199 inner-val 0.10303 best 0.10303@25750 (1136s) immicro=0.0980 ye2026_sepsis=0.1135 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:27:37 step 26000 train 0.10206 inner-val 0.10302 best 0.10302@26000 (1146s) immicro=0.0980 ye2026_sepsis=0.1134 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:27:48 step 26250 train 0.10230 inner-val 0.10302 best 0.10302@26250 (1157s) immicro=0.0980 ye2026_sepsis=0.1134 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:27:59 step 26500 train 0.10241 inner-val 0.10301 best 0.10301@26500 (1168s) immicro=0.0980 ye2026_sepsis=0.1134 imperial_eskd=0.0989 liu2021=0.1018
+2026-10-03T05:28:10 step 26750 train 0.10210 inner-val 0.10301 best 0.10301@26750 (1179s) immicro=0.0980 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:28:21 step 27000 train 0.10244 inner-val 0.10300 best 0.10300@27000 (1190s) immicro=0.0980 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:28:31 step 27250 train 0.10234 inner-val 0.10300 best 0.10300@27250 (1200s) immicro=0.0980 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:28:42 step 27500 train 0.10221 inner-val 0.10300 best 0.10300@27500 (1211s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:28:53 step 27750 train 0.10211 inner-val 0.10299 best 0.10299@27750 (1222s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:29:04 step 28000 train 0.10210 inner-val 0.10298 best 0.10298@28000 (1233s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:29:15 step 28250 train 0.10219 inner-val 0.10298 best 0.10298@28250 (1244s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:29:26 step 28500 train 0.10212 inner-val 0.10298 best 0.10298@28500 (1255s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:29:36 step 28750 train 0.10198 inner-val 0.10297 best 0.10297@28750 (1265s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:29:47 step 29000 train 0.10220 inner-val 0.10297 best 0.10297@29000 (1276s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:29:58 step 29250 train 0.10225 inner-val 0.10296 best 0.10296@29250 (1287s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:30:09 step 29500 train 0.10191 inner-val 0.10296 best 0.10296@29500 (1298s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:30:20 step 29750 train 0.10205 inner-val 0.10296 best 0.10296@29500 (1309s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1018
+2026-10-03T05:30:31 step 30000 train 0.10197 inner-val 0.10295 best 0.10295@30000 (1320s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:30:42 step 30250 train 0.10181 inner-val 0.10295 best 0.10295@30250 (1331s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:30:53 step 30500 train 0.10187 inner-val 0.10295 best 0.10295@30500 (1342s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:31:03 step 30750 train 0.10207 inner-val 0.10294 best 0.10294@30750 (1352s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:31:15 step 31000 train 0.10175 inner-val 0.10294 best 0.10294@31000 (1364s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:31:28 step 31250 train 0.10219 inner-val 0.10293 best 0.10293@31250 (1377s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:31:40 step 31500 train 0.10177 inner-val 0.10293 best 0.10293@31250 (1389s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:31:53 step 31750 train 0.10221 inner-val 0.10293 best 0.10293@31750 (1402s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:32:05 step 32000 train 0.10210 inner-val 0.10293 best 0.10293@31750 (1414s) immicro=0.0979 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:32:17 step 32250 train 0.10215 inner-val 0.10293 best 0.10293@31750 (1426s) immicro=0.0978 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:32:30 step 32500 train 0.10177 inner-val 0.10293 best 0.10293@32500 (1439s) immicro=0.0978 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:32:42 step 32750 train 0.10218 inner-val 0.10293 best 0.10293@32750 (1451s) immicro=0.0978 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:32:54 step 33000 train 0.10207 inner-val 0.10292 best 0.10292@33000 (1463s) immicro=0.0978 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:33:06 step 33250 train 0.10197 inner-val 0.10292 best 0.10292@33000 (1475s) immicro=0.0978 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:33:19 step 33500 train 0.10197 inner-val 0.10292 best 0.10292@33500 (1488s) immicro=0.0978 ye2026_sepsis=0.1134 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:33:31 step 33750 train 0.10204 inner-val 0.10291 best 0.10291@33750 (1500s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:33:43 step 34000 train 0.10218 inner-val 0.10291 best 0.10291@34000 (1512s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:33:55 step 34250 train 0.10204 inner-val 0.10291 best 0.10291@34000 (1524s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:34:07 step 34500 train 0.10208 inner-val 0.10290 best 0.10290@34500 (1536s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:34:20 step 34750 train 0.10215 inner-val 0.10290 best 0.10290@34500 (1549s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:34:32 step 35000 train 0.10225 inner-val 0.10290 best 0.10290@35000 (1561s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:34:44 step 35250 train 0.10175 inner-val 0.10290 best 0.10290@35250 (1573s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:34:56 step 35500 train 0.10227 inner-val 0.10290 best 0.10290@35250 (1586s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:35:09 step 35750 train 0.10204 inner-val 0.10289 best 0.10289@35750 (1598s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:35:21 step 36000 train 0.10204 inner-val 0.10289 best 0.10289@36000 (1610s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:35:34 step 36250 train 0.10224 inner-val 0.10289 best 0.10289@36000 (1623s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1017
+2026-10-03T05:35:46 step 36500 train 0.10179 inner-val 0.10289 best 0.10289@36500 (1635s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:35:59 step 36750 train 0.10204 inner-val 0.10289 best 0.10289@36750 (1648s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:36:11 step 37000 train 0.10185 inner-val 0.10288 best 0.10288@37000 (1660s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:36:23 step 37250 train 0.10199 inner-val 0.10287 best 0.10287@37250 (1673s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:36:36 step 37500 train 0.10209 inner-val 0.10287 best 0.10287@37500 (1685s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:36:48 step 37750 train 0.10203 inner-val 0.10287 best 0.10287@37750 (1697s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:37:01 step 38000 train 0.10216 inner-val 0.10287 best 0.10287@38000 (1710s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:37:13 step 38250 train 0.10206 inner-val 0.10286 best 0.10286@38250 (1722s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0988 liu2021=0.1016
+2026-10-03T05:37:26 step 38500 train 0.10203 inner-val 0.10286 best 0.10286@38250 (1735s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:37:38 step 38750 train 0.10199 inner-val 0.10286 best 0.10286@38750 (1747s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:37:51 step 39000 train 0.10201 inner-val 0.10286 best 0.10286@38750 (1760s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:38:03 step 39250 train 0.10193 inner-val 0.10286 best 0.10286@38750 (1772s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:38:16 step 39500 train 0.10205 inner-val 0.10285 best 0.10285@39500 (1785s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:38:28 step 39750 train 0.10206 inner-val 0.10285 best 0.10285@39750 (1797s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:38:41 step 40000 train 0.10189 inner-val 0.10285 best 0.10285@40000 (1810s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:38:42 block end 40000: best 0.10285@40000; block improvement 0.000248 (2.57 % of total 0.009652) -> continue
+2026-10-03T05:38:55 step 40250 train 0.10199 inner-val 0.10284 best 0.10284@40250 (1824s) immicro=0.0978 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:39:08 step 40500 train 0.10192 inner-val 0.10284 best 0.10284@40500 (1837s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:39:20 step 40750 train 0.10169 inner-val 0.10283 best 0.10283@40750 (1849s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:39:33 step 41000 train 0.10176 inner-val 0.10283 best 0.10283@41000 (1862s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:39:45 step 41250 train 0.10195 inner-val 0.10283 best 0.10283@41000 (1874s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:39:58 step 41500 train 0.10205 inner-val 0.10283 best 0.10283@41000 (1887s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:40:10 step 41750 train 0.10194 inner-val 0.10283 best 0.10283@41000 (1899s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:40:23 step 42000 train 0.10174 inner-val 0.10283 best 0.10283@42000 (1912s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:40:35 step 42250 train 0.10187 inner-val 0.10283 best 0.10283@42000 (1924s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:40:47 step 42500 train 0.10203 inner-val 0.10282 best 0.10282@42500 (1936s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:41:00 step 42750 train 0.10173 inner-val 0.10282 best 0.10282@42500 (1949s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:41:12 step 43000 train 0.10156 inner-val 0.10282 best 0.10282@43000 (1961s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:41:25 step 43250 train 0.10210 inner-val 0.10281 best 0.10281@43250 (1974s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:41:37 step 43500 train 0.10210 inner-val 0.10281 best 0.10281@43500 (1986s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:41:50 step 43750 train 0.10213 inner-val 0.10281 best 0.10281@43750 (1999s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:42:02 step 44000 train 0.10167 inner-val 0.10281 best 0.10281@43750 (2012s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:42:15 step 44250 train 0.10201 inner-val 0.10281 best 0.10281@44250 (2024s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:42:27 step 44500 train 0.10203 inner-val 0.10281 best 0.10281@44250 (2036s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:42:39 step 44750 train 0.10188 inner-val 0.10281 best 0.10281@44250 (2048s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:42:52 step 45000 train 0.10194 inner-val 0.10281 best 0.10281@44250 (2061s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:43:04 step 45250 train 0.10205 inner-val 0.10281 best 0.10281@44250 (2073s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:43:16 step 45500 train 0.10248 inner-val 0.10280 best 0.10280@45500 (2085s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1016
+2026-10-03T05:43:28 step 45750 train 0.10180 inner-val 0.10280 best 0.10280@45750 (2097s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:43:41 step 46000 train 0.10222 inner-val 0.10280 best 0.10280@46000 (2110s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:43:53 step 46250 train 0.10192 inner-val 0.10280 best 0.10280@46250 (2122s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:44:05 step 46500 train 0.10198 inner-val 0.10279 best 0.10279@46500 (2134s) immicro=0.0977 ye2026_sepsis=0.1133 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:44:17 step 46750 train 0.10167 inner-val 0.10279 best 0.10279@46750 (2146s) immicro=0.0977 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:44:30 step 47000 train 0.10176 inner-val 0.10278 best 0.10278@47000 (2159s) immicro=0.0977 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:44:42 step 47250 train 0.10185 inner-val 0.10278 best 0.10278@47250 (2171s) immicro=0.0977 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:44:54 step 47500 train 0.10199 inner-val 0.10277 best 0.10277@47500 (2183s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:45:06 step 47750 train 0.10187 inner-val 0.10277 best 0.10277@47750 (2195s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:45:18 step 48000 train 0.10206 inner-val 0.10277 best 0.10277@47750 (2207s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:45:30 step 48250 train 0.10198 inner-val 0.10276 best 0.10276@48250 (2219s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:45:42 step 48500 train 0.10188 inner-val 0.10276 best 0.10276@48500 (2232s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:45:55 step 48750 train 0.10163 inner-val 0.10276 best 0.10276@48500 (2244s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:46:07 step 49000 train 0.10198 inner-val 0.10276 best 0.10276@48500 (2256s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:46:19 step 49250 train 0.10188 inner-val 0.10276 best 0.10276@48500 (2268s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:46:31 step 49500 train 0.10171 inner-val 0.10276 best 0.10276@49500 (2280s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:46:43 step 49750 train 0.10193 inner-val 0.10275 best 0.10275@49750 (2292s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:46:56 step 50000 train 0.10138 inner-val 0.10275 best 0.10275@50000 (2305s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:47:08 step 50250 train 0.10160 inner-val 0.10275 best 0.10275@50000 (2317s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:47:20 step 50500 train 0.10198 inner-val 0.10275 best 0.10275@50500 (2329s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:47:32 step 50750 train 0.10160 inner-val 0.10275 best 0.10275@50750 (2341s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:47:44 step 51000 train 0.10197 inner-val 0.10275 best 0.10275@51000 (2353s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:47:56 step 51250 train 0.10186 inner-val 0.10274 best 0.10274@51250 (2365s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:48:08 step 51500 train 0.10171 inner-val 0.10274 best 0.10274@51500 (2378s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:48:21 step 51750 train 0.10196 inner-val 0.10274 best 0.10274@51750 (2390s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:48:33 step 52000 train 0.10206 inner-val 0.10274 best 0.10274@51750 (2402s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:48:45 step 52250 train 0.10200 inner-val 0.10274 best 0.10274@52250 (2414s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:48:57 step 52500 train 0.10225 inner-val 0.10273 best 0.10273@52500 (2426s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:49:09 step 52750 train 0.10204 inner-val 0.10273 best 0.10273@52750 (2438s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:49:21 step 53000 train 0.10159 inner-val 0.10273 best 0.10273@53000 (2450s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:49:34 step 53250 train 0.10176 inner-val 0.10273 best 0.10273@53250 (2463s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:49:46 step 53500 train 0.10177 inner-val 0.10273 best 0.10273@53250 (2475s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:49:58 step 53750 train 0.10179 inner-val 0.10273 best 0.10273@53250 (2487s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:50:10 step 54000 train 0.10169 inner-val 0.10273 best 0.10273@54000 (2499s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0987 liu2021=0.1015
+2026-10-03T05:50:22 step 54250 train 0.10212 inner-val 0.10272 best 0.10272@54250 (2511s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:50:34 step 54500 train 0.10179 inner-val 0.10272 best 0.10272@54250 (2523s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:50:47 step 54750 train 0.10153 inner-val 0.10272 best 0.10272@54750 (2536s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:50:59 step 55000 train 0.10178 inner-val 0.10271 best 0.10271@55000 (2548s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:51:11 step 55250 train 0.10157 inner-val 0.10271 best 0.10271@55250 (2560s) immicro=0.0976 ye2026_sepsis=0.1132 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:51:23 step 55500 train 0.10179 inner-val 0.10271 best 0.10271@55250 (2572s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:51:35 step 55750 train 0.10189 inner-val 0.10271 best 0.10271@55750 (2584s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:51:47 step 56000 train 0.10171 inner-val 0.10270 best 0.10270@56000 (2596s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:51:59 step 56250 train 0.10193 inner-val 0.10270 best 0.10270@56000 (2608s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:52:12 step 56500 train 0.10158 inner-val 0.10270 best 0.10270@56500 (2621s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:52:24 step 56750 train 0.10161 inner-val 0.10270 best 0.10270@56750 (2633s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:52:36 step 57000 train 0.10176 inner-val 0.10269 best 0.10269@57000 (2645s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:52:48 step 57250 train 0.10212 inner-val 0.10269 best 0.10269@57250 (2657s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:53:00 step 57500 train 0.10172 inner-val 0.10269 best 0.10269@57250 (2669s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:53:12 step 57750 train 0.10144 inner-val 0.10269 best 0.10269@57750 (2681s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:53:24 step 58000 train 0.10169 inner-val 0.10269 best 0.10269@57750 (2694s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:53:36 step 58250 train 0.10182 inner-val 0.10269 best 0.10269@57750 (2705s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:53:47 step 58500 train 0.10198 inner-val 0.10269 best 0.10269@58500 (2716s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:53:58 step 58750 train 0.10214 inner-val 0.10268 best 0.10268@58750 (2727s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:54:09 step 59000 train 0.10202 inner-val 0.10268 best 0.10268@59000 (2738s) immicro=0.0976 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:54:20 step 59250 train 0.10164 inner-val 0.10268 best 0.10268@59250 (2749s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:54:30 step 59500 train 0.10158 inner-val 0.10268 best 0.10268@59500 (2760s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:54:41 step 59750 train 0.10212 inner-val 0.10268 best 0.10268@59500 (2770s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:54:52 step 60000 train 0.10156 inner-val 0.10268 best 0.10268@59500 (2781s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:54:54 block end 60000: best 0.10268@59500; block improvement 0.000171 (1.74 % of total 0.009822) -> continue
+2026-10-03T05:55:05 step 60250 train 0.10166 inner-val 0.10268 best 0.10268@59500 (2794s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:55:15 step 60500 train 0.10187 inner-val 0.10268 best 0.10268@59500 (2804s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:55:26 step 60750 train 0.10152 inner-val 0.10267 best 0.10267@60750 (2815s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:55:37 step 61000 train 0.10202 inner-val 0.10267 best 0.10267@60750 (2826s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:55:48 step 61250 train 0.10199 inner-val 0.10267 best 0.10267@60750 (2837s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:55:59 step 61500 train 0.10176 inner-val 0.10268 best 0.10267@60750 (2848s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:56:10 step 61750 train 0.10182 inner-val 0.10268 best 0.10267@60750 (2859s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:56:21 step 62000 train 0.10173 inner-val 0.10268 best 0.10267@60750 (2870s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:56:32 step 62250 train 0.10170 inner-val 0.10267 best 0.10267@62250 (2881s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1015
+2026-10-03T05:56:43 step 62500 train 0.10167 inner-val 0.10267 best 0.10267@62500 (2892s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:56:54 step 62750 train 0.10167 inner-val 0.10266 best 0.10266@62750 (2903s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:57:05 step 63000 train 0.10161 inner-val 0.10266 best 0.10266@63000 (2914s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:57:15 step 63250 train 0.10206 inner-val 0.10266 best 0.10266@63000 (2924s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:57:26 step 63500 train 0.10169 inner-val 0.10266 best 0.10266@63000 (2936s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:57:37 step 63750 train 0.10192 inner-val 0.10266 best 0.10266@63000 (2946s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:57:48 step 64000 train 0.10188 inner-val 0.10266 best 0.10266@63000 (2957s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:57:59 step 64250 train 0.10179 inner-val 0.10266 best 0.10266@64250 (2968s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:58:10 step 64500 train 0.10178 inner-val 0.10265 best 0.10266@64250 (2979s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:58:21 step 64750 train 0.10193 inner-val 0.10265 best 0.10265@64750 (2990s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:58:32 step 65000 train 0.10194 inner-val 0.10265 best 0.10265@64750 (3001s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:58:43 step 65250 train 0.10184 inner-val 0.10265 best 0.10265@65250 (3012s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:58:54 step 65500 train 0.10156 inner-val 0.10265 best 0.10265@65500 (3023s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:59:04 step 65750 train 0.10180 inner-val 0.10265 best 0.10265@65500 (3034s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:59:15 step 66000 train 0.10186 inner-val 0.10265 best 0.10265@65500 (3044s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:59:26 step 66250 train 0.10165 inner-val 0.10264 best 0.10264@66250 (3055s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:59:37 step 66500 train 0.10197 inner-val 0.10264 best 0.10264@66500 (3066s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:59:48 step 66750 train 0.10145 inner-val 0.10264 best 0.10264@66750 (3077s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T05:59:58 step 67000 train 0.10207 inner-val 0.10263 best 0.10263@67000 (3087s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:00:10 step 67250 train 0.10169 inner-val 0.10262 best 0.10262@67250 (3100s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:00:23 step 67500 train 0.10188 inner-val 0.10263 best 0.10262@67250 (3112s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:00:35 step 67750 train 0.10163 inner-val 0.10263 best 0.10262@67250 (3124s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:00:47 step 68000 train 0.10187 inner-val 0.10263 best 0.10262@67250 (3136s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:01:00 step 68250 train 0.10189 inner-val 0.10263 best 0.10262@67250 (3149s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:01:12 step 68500 train 0.10169 inner-val 0.10263 best 0.10262@67250 (3161s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:01:24 step 68750 train 0.10201 inner-val 0.10263 best 0.10262@67250 (3173s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:01:36 step 69000 train 0.10171 inner-val 0.10263 best 0.10262@67250 (3186s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:01:49 step 69250 train 0.10183 inner-val 0.10263 best 0.10262@67250 (3198s) immicro=0.0975 ye2026_sepsis=0.1131 imperial_eskd=0.0986 liu2021=0.1014
+2026-10-03T06:01:49 early stop: no improvement in 8 evaluations
+2026-10-03T06:01:54 saved /home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_giroux2022/K1000/lora64sat/checkpoint_ft.pt (selected step 67250, inner-val 0.10262 vs pretrained 0.11250); sha 24e23d695129

paper-releases/pbisc-diffusion/updates/20261003T103040Z-crosscohort-k1000-saturation/diffusion/dataset/server_1/model8/plans/paired_crosscohort_20261002/runs/hoL_immicro/K1000/fullsat/RUN.json

@@ -0,0 +1,536 @@
+{
+ "schema": "pbisc_sdd.ft_crosscohort.run.v1",
+ "source_checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/runs/f01_6k_1x_seed20260627/checkpoint_epoch_016.pt",
+ "source_checkpoint_sha256": "8c7cd0e83e2bc19fe03dfcbc084919a1e2f61be8cc8bc12f4caf543dfda91633",
+ "split": "hoL_immicro",
+ "split_file": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/splits/hoL_immicro.json",
+ "split_sha256": "d5646bee4a5f735e64f490824ca865a2cf8191234a8de40c0963cd590ebe6743",
+ "held_out": "immicro",
+ "recipe_k": 1000,
+ "balance": "cohort",
+ "mode": "full",
+ "trainable": {
+  "mode": "full",
+  "n_trainable": 54185184,
+  "n_total_base": 54185184,
+  "trainable_tensors": 64
+ },
+ "lr": 5e-05,
+ "warmup": 200,
+ "batch_size": 256,
+ "ema_decay": 0.999,
+ "max_steps": 200000,
+ "steps_run": 65000,
+ "eval_every": 250,
+ "patience": 8,
+ "selected_step": 63000,
+ "selected_inner_val_loss": 0.10436773711897007,
+ "pretrained_inner_val_loss": 0.11485457591248383,
+ "cohorts": [
+  {
+   "cohort": "ye2026_sepsis",
+   "condition_npz": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/data/ye2026_sepsis/conditions_raw_K1000.npz",
+   "condition_npz_sha256": "ce8b61d7dcd3e62b9e63d52189cfb5bba881f87a661384277a7949a629231cb4",
+   "train_rows": [
+    "Abd-PS04",
+    "Abd-PS05",
+    "Abd-PS07",
+    "Abd-PS19",
+    "Abd-PS20",
+    "Abd-PS27",
+    "Abd-S103",
+    "Abd-S120",
+    "Abd-S123",
+    "Abd-S156",
+    "Abd-S39",
+    "Abd-S42",
+    "Abd-S47",
+    "Abd-S51",
+    "Abd-S57",
+    "Abd-S66",
+    "Abd-S79",
+    "Abd-S83",
+    "HC05",
+    "HC07",
+    "HC09",
+    "HC12",
+    "HC13",
+    "HC16",
+    "IC10",
+    "IC15",
+    "IC16",
+    "IC22",
+    "IC23",
+    "PHC01",
+    "PHC02",
+    "PHC03",
+    "PHC05",
+    "PIC01",
+    "PIC02",
+    "PIC04",
+    "PIC05",
+    "Res-PS02",
+    "Res-PS03",
+    "Res-PS12",
+    "Res-PS21",
+    "Res-PS22",
+    "Res-PS25",
+    "Res-PS26",
+    "Res-S10",
+    "Res-S125",
+    "Res-S127",
+    "Res-S136",
+    "Res-S146",
+    "Res-S149",
+    "Res-S150",
+    "Res-S164",
+    "Res-S166",
+    "Res-S177",
+    "Res-S184",
+    "Res-S19",
+    "Res-S26",
+    "Res-S40",
+    "Res-S77",
+    "Res-S87",
+    "Skin-S140",
+    "Skin-S142",
+    "Skin-S152",
+    "Skin-S38",
+    "Skin-S43",
+    "Skin-S46",
+    "Skin-S52",
+    "Skin-S74",
+    "Uri-S134",
+    "Uri-S147",
+    "Uri-S157",
+    "Uri-S20",
+    "Uri-S93"
+   ],
+   "inner_val_rows": [
+    "Abd-PS23",
+    "Abd-S102",
+    "Abd-S54",
+    "Abd-S98",
+    "HC01",
+    "IC19",
+    "Res-PS16",
+    "Res-S139",
+    "Res-S174"
+   ]
+  },
+  {
+   "cohort": "imperial_eskd",
+   "condition_npz": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/data/imperial_eskd/conditions_raw_K1000.npz",
+   "condition_npz_sha256": "41a62291e98e367d05108a0a95a859ef9ca1997f1fda4656cba58eeefbe56442",
+   "train_rows": [
+    "C20_1004",
+    "C20_1404",
+    "C21_1004",
+    "C21_1704",
+    "C21_2504",
+    "C23_2204",
+    "C31_1304",
+    "C31_2004",
+    "C31_2704",
+    "C33_1401",
+    "C33_1601",
+    "C33_1901",
+    "C34_1404",
+    "C34_2104",
+    "C34_2504",
+    "C36_1504",
+    "C36_1704",
+    "C36_2104",
+    "C40_1504",
+    "C40_2204",
+    "C40_0105",
+    "C42_1504",
+    "C42_2204",
+    "C42_2704",
+    "C58_1604",
+    "C58_2501",
+    "C58_2901",
+    "C58_0102",
+    "C58_0402",
+    "C58_1003",
+    "C60_2504",
+    "C65_1704",
+    "C65_2104",
+    "C65_2504",
+    "C69_1804",
+    "C82_2504",
+    "C85_2004",
+    "C93_2104",
+    "C93_2704",
+    "C93_0405",
+    "C101_0601",
+    "C101_1101",
+    "C101_1301",
+    "C101_1801",
+    "C101_2001",
+    "C101_2502",
+    "C103_2204",
+    "C104_2204",
+    "C106_2204",
+    "C108_2204",
+    "C108_1501",
+    "C108_1801",
+    "C108_2201",
+    "C113_0605",
+    "C123_2404",
+    "C123_2904",
+    "C123_0505",
+    "C124_2404",
+    "C124_2704",
+    "C124_0105",
+    "C126_2504",
+    "C126_3004",
+    "C126_0605",
+    "C127_2504",
+    "C127_0405",
+    "C127_1105",
+    "C132_2804",
+    "C132_0205",
+    "C132_1105",
+    "C137_3004",
+    "C137_2502",
+    "C138_3004",
+    "C138_2301",
+    "C138_3001",
+    "C138_0202",
+    "C138_0402",
+    "C140_3004",
+    "C140_2801",
+    "C140_3001",
+    "C140_0202",
+    "C140_0402",
+    "C140_0602",
+    "C140_0902",
+    "C140_1103",
+    "C141_2001",
+    "C141_2201",
+    "C141_3001",
+    "C141_0202",
+    "C145_3004",
+    "C145_0701",
+    "C145_1201",
+    "C145_1401",
+    "C145_1901",
+    "C145_2301",
+    "C146_3004",
+    "C146_1201",
+    "C146_1401",
+    "C146_1601",
+    "C146_1901",
+    "C146_2301",
+    "C146_2701",
+    "C146_2502",
+    "C152_0405",
+    "C161_1105",
+    "C164_1105",
+    "C167_1205",
+    "C168_1205",
+    "C168_2001",
+    "C168_2301",
+    "C168_3001",
+    "C168_0202",
+    "C168_0403",
+    "C169_1205",
+    "C169_2001",
+    "C169_2301",
+    "C169_0202",
+    "C169_0403",
+    "C170_1205",
+    "C170_2301",
+    "C170_0202",
+    "C170_0402",
+    "C170_0403",
+    "C187_2605",
+    "C187_1201",
+    "C187_1401",
+    "C187_1901",
+    "C187_2301",
+    "C187_2502",
+    "C190_2605",
+    "C190_2901",
+    "C190_0102",
+    "C190_0302",
+    "C190_0502",
+    "C190_0802",
+    "C190_1002",
+    "C190_0903",
+    "C193_2605",
+    "C195_2605",
+    "C197_2605",
+    "C212_2705",
+    "C214_2705",
+    "C237_2805",
+    "C238_2905",
+    "C240_2905",
+    "C246_2905",
+    "C251_2905"
+   ],
+   "inner_val_rows": [
+    "C63_1704",
+    "C63_2404",
+    "C63_2904",
+    "C128_2504",
+    "C128_0405",
+    "C128_1105",
+    "C142_3004",
+    "C147_3004",
+    "C147_1401",
+    "C147_1901",
+    "C234_2805"
+   ]
+  },
+  {
+   "cohort": "giroux2022",
+   "condition_npz": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/data/giroux2022/conditions_raw_K1000.npz",
+   "condition_npz_sha256": "b584655fcfa5fa4c0887d5dd83fcb1c5a0206da4afe5fefab96cba2272b1371c",
+   "train_rows": [
+    "450905_d0",
+    "450905_d14",
+    "0B943B_d0",
+    "0B943B_d14",
+    "7085CA_d0",
+    "7085CA_d7",
+    "7085CA_d14",
+    "2AD75E_d0",
+    "2AD75E_d7",
+    "2AD75E_d14",
+    "B85D75_d0",
+    "B85D75_d7",
+    "180E1A_d0",
+    "180E1A_d7",
+    "180E1A_d14",
+    "82CCF5_d0",
+    "82CCF5_d7",
+    "82CCF5_d14",
+    "BAAF62_d0",
+    "BAAF62_d7",
+    "BAAF62_d14",
+    "0BF51C_d0",
+    "0BF51C_d7",
+    "0BF51C_d14"
+   ],
+   "inner_val_rows": []
+  },
+  {
+   "cohort": "liu2021",
+   "condition_npz": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/data/liu2021/conditions_raw_K1000.npz",
+   "condition_npz_sha256": "ad3131eabcfb11b166ca37c7585c50abb11efb60e2a736fe62de604794535e0c",
+   "train_rows": [
+    "AA220014_HC",
+    "AA220534_HC",
+    "AA220907_HC",
+    "HGR0000071_T0",
+    "HGR0000074_T0",
+    "HGR0000083_T1",
+    "HGR0000089_T1",
+    "HGR0000091_T0",
+    "HGR0000093_T0",
+    "HGR0000136_T0",
+    "HGR0000137_T0",
+    "HGR0000141_T0",
+    "IM000040_HC",
+    "HGR0000042_T1",
+    "HGR0000069_T0",
+    "HGR0000069_T1",
+    "HGR0000092_T0",
+    "HGR0000094_T0",
+    "HGR0000124_T0",
+    "HGR0000144_T2",
+    "HGR0000391_T0",
+    "HGR0000395_T0",
+    "HGR0000418_T1",
+    "SHD4_HC",
+    "SHD7_HC",
+    "HGR0000051_T0",
+    "HGR0000051_T1",
+    "HGR0000051_T3",
+    "HGR0000101_T1",
+    "HGR0000101_T3",
+    "HGR0000102_T0",
+    "HGR0000102_T1",
+    "HGR0000134_T0",
+    "HGR0000135_T0",
+    "HGR0000135_T1",
+    "HGR0000392_T0",
+    "HGR0000429_T1",
+    "HGR0000429_T2",
+    "SHD3_HC",
+    "SHD5_HC"
+   ],
+   "inner_val_rows": [
+    "HGR0000098_T0",
+    "HGR0000143_T0",
+    "HGR0000115_T1",
+    "SHD1_HC"
+   ]
+  }
+ ],
+ "held_out_test_rows_never_loaded": [
+  "HS1",
+  "HS2",
+  "HS3",
+  "HS4",
+  "HS6",
+  "HS7",
+  "HS8",
+  "HS9",
+  "HS10",
+  "HS11",
+  "HS12",
+  "HS13",
+  "HS14",
+  "HS15",
+  "HS16",
+  "HS17",
+  "HS18",
+  "HS19",
+  "HS20",
+  "HS21",
+  "HS23",
+  "HS24",
+  "HS25",
+  "HS26",
+  "HS27",
+  "HS28",
+  "HS29",
+  "HS30",
+  "HS31",
+  "HS32",
+  "HS34",
+  "HS35",
+  "HS36",
+  "HS37",
+  "HS38",
+  "HS39",
+  "HS40",
+  "HS41",
+  "HS42",
+  "HS43",
+  "HS44",
+  "HS45",
+  "HS46",
+  "HS47",
+  "HS48",
+  "HS49",
+  "HS50",
+  "HS51",
+  "HS52",
+  "HS53",
+  "HS54",
+  "HS55",
+  "HS56",
+  "HS57",
+  "HS58",
+  "HS59",
+  "HS60",
+  "HS61",
+  "HS62",
+  "HS63",
+  "HS64",
+  "HS65",
+  "HS66",
+  "HS67",
+  "HS68",
+  "HS69",
+  "HS70",
+  "HS71",
+  "HS72",
+  "HS73",
+  "HS74",
+  "HS75",
+  "HS76",
+  "HS77",
+  "HS78",
+  "HS79",
+  "HS80",
+  "HS81",
+  "HS82",
+  "HS83",
+  "HS84",
+  "HS85",
+  "HS86",
+  "HS87",
+  "HS88",
+  "HS89",
+  "HS90",
+  "HS91",
+  "HS92",
+  "HS93",
+  "HS94",
+  "HS95",
+  "HS96",
+  "HS97",
+  "HS98",
+  "HS100",
+  "HS101",
+  "HS102",
+  "HS103",
+  "HS104",
+  "HS105",
+  "HS106",
+  "HS107",
+  "HS108",
+  "HS109",
+  "HS110",
+  "HS111"
+ ],
+ "max_abs_weight_change": 0.9364820718765259,
+ "n_tensors_changed": 64,
+ "seed": 20261002,
+ "wall_seconds": 3523.7,
+ "smoke": false,
+ "block": 20000,
+ "saturation_frac": 0.01,
+ "stop_reason": "early_stop",
+ "milestones": [
+  {
+   "milestone_step": 20000,
+   "selected_step": 20000,
+   "selected_inner_val_loss": 0.10489554748077083,
+   "block_improvement": 0.009959028431712993,
+   "total_improvement": 0.009959028431712993,
+   "reason": "block_end",
+   "checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_immicro/K1000/fullsat/checkpoint_ft_s20000.pt",
+   "checkpoint_sha256": "079c482ae6d6c25d1dab913b0101807b6f169e1c0c1975b197257918276747e8"
+  },
+  {
+   "milestone_step": 40000,
+   "selected_step": 40000,
+   "selected_inner_val_loss": 0.10458750774776594,
+   "block_improvement": 0.00030803973300488896,
+   "total_improvement": 0.010267068164717882,
+   "reason": "block_end",
+   "checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_immicro/K1000/fullsat/checkpoint_ft_s40000.pt",
+   "checkpoint_sha256": "cb2a07994c9c8c6ebaf9a9b2f424d93a57fc6080b3dd081b288dc3a3c7956e03"
+  },
+  {
+   "milestone_step": 60000,
+   "selected_step": 60000,
+   "selected_inner_val_loss": 0.10440332228662791,
+   "block_improvement": 0.0001841854611380339,
+   "total_improvement": 0.010451253625855916,
+   "reason": "block_end",
+   "checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_immicro/K1000/fullsat/checkpoint_ft_s60000.pt",
+   "checkpoint_sha256": "ed28603294a1a617e007b44d2f58dbc050869a6879bc2a73c7118f59c3fb26d3"
+  },
+  {
+   "milestone_step": 65000,
+   "selected_step": 63000,
+   "selected_inner_val_loss": 0.10436773711897007,
+   "block_improvement": 3.55851676578367e-05,
+   "total_improvement": 0.010486838793513753,
+   "reason": "early_stop",
+   "checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_crosscohort_20261002/runs/hoL_immicro/K1000/fullsat/checkpoint_ft_s65000.pt",
+   "checkpoint_sha256": "5c49c8ecff8e594daae931ae784fa3b289404ca3d1bcf4dfbb3cd9cc34f58be5"
+  }
+ ],
+ "written_utc": "2026-10-03T04:09:41.092010+00:00",
+ "last_milestone_max_abs_diff": 0.0,
+ "checkpoint_sha256": "72a9e1b9fdf5e096a86843f2e79816dfa7f596e873db3d092ce28cbf07ebd269",
+ "reload_forward_max_abs_diff": 0.0
+}

paper-releases/pbisc-diffusion/updates/20261003T103040Z-crosscohort-k1000-saturation/diffusion/dataset/server_1/model8/plans/paired_crosscohort_20261002/runs/hoL_immicro/K1000/fullsat/history.json

@@ -0,0 +1,2872 @@
+[
+ {
+  "step": 0,
+  "inner_val_loss": 0.11485457591248383,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.12609728963838684,
+   "imperial_eskd": 0.1086193590678952,
+   "liu2021": 0.10984707903116941
+  },
+  "train_loss": null
+ },
+ {
+  "step": 250,
+  "inner_val_loss": 0.11293623407656696,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.12422744515869352,
+   "imperial_eskd": 0.10648235340010036,
+   "liu2021": 0.10809890367090702
+  },
+  "train_loss": 0.1066121252477169,
+  "sec": 11.5
+ },
+ {
+  "step": 500,
+  "inner_val_loss": 0.11050809919834137,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.12161432579159737,
+   "imperial_eskd": 0.10389786586165428,
+   "liu2021": 0.10601210594177246
+  },
+  "train_loss": 0.10367642712593078,
+  "sec": 24.5
+ },
+ {
+  "step": 750,
+  "inner_val_loss": 0.1088023175632813,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1195632732576794,
+   "imperial_eskd": 0.10218783061612736,
+   "liu2021": 0.10465584881603718
+  },
+  "train_loss": 0.10342625418305397,
+  "sec": 38.3
+ },
+ {
+  "step": 1000,
+  "inner_val_loss": 0.10773561290945068,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11820382252335548,
+   "imperial_eskd": 0.10118104754523798,
+   "liu2021": 0.10382196865975857
+  },
+  "train_loss": 0.10293849876523017,
+  "sec": 52.4
+ },
+ {
+  "step": 1250,
+  "inner_val_loss": 0.10708637913431884,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11738109630015162,
+   "imperial_eskd": 0.10055816241286018,
+   "liu2021": 0.10331987868994474
+  },
+  "train_loss": 0.10313243559002876,
+  "sec": 66.6
+ },
+ {
+  "step": 1500,
+  "inner_val_loss": 0.1066784006214202,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11685892070333163,
+   "imperial_eskd": 0.10018011250279167,
+   "liu2021": 0.10299616865813732
+  },
+  "train_loss": 0.10295427325367927,
+  "sec": 79.4
+ },
+ {
+  "step": 1750,
+  "inner_val_loss": 0.10640772020989897,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11651934559146564,
+   "imperial_eskd": 0.09991799769076434,
+   "liu2021": 0.10278581734746695
+  },
+  "train_loss": 0.1027361755669117,
+  "sec": 92.2
+ },
+ {
+  "step": 2000,
+  "inner_val_loss": 0.10621871395030306,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11627781473928028,
+   "imperial_eskd": 0.09973372248086063,
+   "liu2021": 0.1026446046307683
+  },
+  "train_loss": 0.10253991255164147,
+  "sec": 105.2
+ },
+ {
+  "step": 2250,
+  "inner_val_loss": 0.10607889734258756,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11609945901566082,
+   "imperial_eskd": 0.09959834373810074,
+   "liu2021": 0.10253888927400112
+  },
+  "train_loss": 0.1025280137360096,
+  "sec": 118.3
+ },
+ {
+  "step": 2500,
+  "inner_val_loss": 0.1059715310017569,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11595803995927174,
+   "imperial_eskd": 0.09949436749924313,
+   "liu2021": 0.10246218554675579
+  },
+  "train_loss": 0.10301806622743606,
+  "sec": 132.4
+ },
+ {
+  "step": 2750,
+  "inner_val_loss": 0.10588359468074059,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11584188499384457,
+   "imperial_eskd": 0.09941044856201518,
+   "liu2021": 0.10239845048636198
+  },
+  "train_loss": 0.10245795056223869,
+  "sec": 145.4
+ },
+ {
+  "step": 3000,
+  "inner_val_loss": 0.10581201738568168,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11574503738019201,
+   "imperial_eskd": 0.0993428761986169,
+   "liu2021": 0.1023481385782361
+  },
+  "train_loss": 0.10245085799694062,
+  "sec": 158.8
+ },
+ {
+  "step": 3250,
+  "inner_val_loss": 0.10575367021766376,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11566006433632639,
+   "imperial_eskd": 0.09928859567100351,
+   "liu2021": 0.10231235064566135
+  },
+  "train_loss": 0.10277113959193229,
+  "sec": 171.8
+ },
+ {
+  "step": 3500,
+  "inner_val_loss": 0.10570121927481609,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11558838768137826,
+   "imperial_eskd": 0.09924177893183449,
+   "liu2021": 0.10227349121123552
+  },
+  "train_loss": 0.10266501396894455,
+  "sec": 186.0
+ },
+ {
+  "step": 3750,
+  "inner_val_loss": 0.10565347139181112,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11551803805761868,
+   "imperial_eskd": 0.09919790957461704,
+   "liu2021": 0.10224446654319763
+  },
+  "train_loss": 0.10230978727340698,
+  "sec": 200.1
+ },
+ {
+  "step": 4000,
+  "inner_val_loss": 0.1056128127174295,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11546093432439698,
+   "imperial_eskd": 0.09916240281679413,
+   "liu2021": 0.10221510101109743
+  },
+  "train_loss": 0.10239347261190414,
+  "sec": 214.5
+ },
+ {
+  "step": 4250,
+  "inner_val_loss": 0.10558171312854647,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11541619234614903,
+   "imperial_eskd": 0.09912993555719202,
+   "liu2021": 0.10219901148229837
+  },
+  "train_loss": 0.10236127415299416,
+  "sec": 228.5
+ },
+ {
+  "step": 4500,
+  "inner_val_loss": 0.1055485403905262,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1153673707611031,
+   "imperial_eskd": 0.09909717806360939,
+   "liu2021": 0.10218107234686613
+  },
+  "train_loss": 0.10237844324111939,
+  "sec": 242.8
+ },
+ {
+  "step": 4750,
+  "inner_val_loss": 0.10552075139939886,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11532616035805808,
+   "imperial_eskd": 0.09907174245877699,
+   "liu2021": 0.10216435138136148
+  },
+  "train_loss": 0.10236009222269059,
+  "sec": 256.8
+ },
+ {
+  "step": 5000,
+  "inner_val_loss": 0.10549191364811526,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1152858038743337,
+   "imperial_eskd": 0.09904267638921738,
+   "liu2021": 0.10214726068079472
+  },
+  "train_loss": 0.10234181824326515,
+  "sec": 271.0
+ },
+ {
+  "step": 5250,
+  "inner_val_loss": 0.10546774895993448,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1152518925567468,
+   "imperial_eskd": 0.09901775893839923,
+   "liu2021": 0.10213359538465738
+  },
+  "train_loss": 0.10225727549195289,
+  "sec": 285.3
+ },
+ {
+  "step": 5500,
+  "inner_val_loss": 0.10544470867599903,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11521996474928325,
+   "imperial_eskd": 0.0989938046444546,
+   "liu2021": 0.10212035663425922
+  },
+  "train_loss": 0.102288833796978,
+  "sec": 299.4
+ },
+ {
+  "step": 5750,
+  "inner_val_loss": 0.10542314607800558,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11519015828768413,
+   "imperial_eskd": 0.0989752018993551,
+   "liu2021": 0.10210407804697752
+  },
+  "train_loss": 0.10188942503929138,
+  "sec": 312.8
+ },
+ {
+  "step": 6000,
+  "inner_val_loss": 0.10540264491807132,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11516530853178766,
+   "imperial_eskd": 0.09895246679132635,
+   "liu2021": 0.10209015943109989
+  },
+  "train_loss": 0.10211304318904876,
+  "sec": 325.4
+ },
+ {
+  "step": 6250,
+  "inner_val_loss": 0.10538300104840637,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11514126426643795,
+   "imperial_eskd": 0.09893160516565497,
+   "liu2021": 0.10207613371312618
+  },
+  "train_loss": 0.1022240168750286,
+  "sec": 339.0
+ },
+ {
+  "step": 6500,
+  "inner_val_loss": 0.10536176590907453,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11511583129564922,
+   "imperial_eskd": 0.09891082481904463,
+   "liu2021": 0.10205864161252975
+  },
+  "train_loss": 0.10223280590772629,
+  "sec": 353.2
+ },
+ {
+  "step": 6750,
+  "inner_val_loss": 0.10534358381447366,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11509696973694696,
+   "imperial_eskd": 0.09888940609314224,
+   "liu2021": 0.1020443756133318
+  },
+  "train_loss": 0.10216372615098954,
+  "sec": 366.7
+ },
+ {
+  "step": 7000,
+  "inner_val_loss": 0.10532697844585585,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11507755352391137,
+   "imperial_eskd": 0.09887337278236043,
+   "liu2021": 0.10203000903129578
+  },
+  "train_loss": 0.10196406161785125,
+  "sec": 380.4
+ },
+ {
+  "step": 7250,
+  "inner_val_loss": 0.10531168120602767,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11505960300564766,
+   "imperial_eskd": 0.098856121301651,
+   "liu2021": 0.10201931931078434
+  },
+  "train_loss": 0.10221881693601609,
+  "sec": 393.9
+ },
+ {
+  "step": 7500,
+  "inner_val_loss": 0.10529890342894628,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11504076090123919,
+   "imperial_eskd": 0.09884262592955069,
+   "liu2021": 0.10201332345604897
+  },
+  "train_loss": 0.10207948490977287,
+  "sec": 407.3
+ },
+ {
+  "step": 7750,
+  "inner_val_loss": 0.10528458861213942,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11502264026138517,
+   "imperial_eskd": 0.09882640940221873,
+   "liu2021": 0.10200471617281437
+  },
+  "train_loss": 0.10203361567854881,
+  "sec": 420.8
+ },
+ {
+  "step": 8000,
+  "inner_val_loss": 0.10527246416568355,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11500651844673687,
+   "imperial_eskd": 0.09881122613495047,
+   "liu2021": 0.10199964791536331
+  },
+  "train_loss": 0.1016498242020607,
+  "sec": 433.6
+ },
+ {
+  "step": 8250,
+  "inner_val_loss": 0.10525846725285855,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11499019546641244,
+   "imperial_eskd": 0.09879745509136807,
+   "liu2021": 0.10198775120079517
+  },
+  "train_loss": 0.1020395722091198,
+  "sec": 446.2
+ },
+ {
+  "step": 8500,
+  "inner_val_loss": 0.10524590140026728,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11497773395644294,
+   "imperial_eskd": 0.09878250550140034,
+   "liu2021": 0.10197746474295855
+  },
+  "train_loss": 0.1017099098265171,
+  "sec": 460.1
+ },
+ {
+  "step": 8750,
+  "inner_val_loss": 0.10523233533545275,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11496259023745854,
+   "imperial_eskd": 0.09877030246637085,
+   "liu2021": 0.10196411330252886
+  },
+  "train_loss": 0.10231356340646744,
+  "sec": 474.4
+ },
+ {
+  "step": 9000,
+  "inner_val_loss": 0.10522304532691988,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11495427290598552,
+   "imperial_eskd": 0.09875950792973692,
+   "liu2021": 0.10195535514503717
+  },
+  "train_loss": 0.10209669968485832,
+  "sec": 488.4
+ },
+ {
+  "step": 9250,
+  "inner_val_loss": 0.10520768389168511,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11493437654442257,
+   "imperial_eskd": 0.09874517944726077,
+   "liu2021": 0.10194349568337202
+  },
+  "train_loss": 0.10193610075116158,
+  "sec": 502.6
+ },
+ {
+  "step": 9500,
+  "inner_val_loss": 0.10519814445176208,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11492305373152097,
+   "imperial_eskd": 0.09873668307607825,
+   "liu2021": 0.10193469654768705
+  },
+  "train_loss": 0.1019431373178959,
+  "sec": 516.7
+ },
+ {
+  "step": 9750,
+  "inner_val_loss": 0.10519139231605963,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11491698771715164,
+   "imperial_eskd": 0.09872647408734668,
+   "liu2021": 0.10193071514368057
+  },
+  "train_loss": 0.1019745808839798,
+  "sec": 530.7
+ },
+ {
+  "step": 10000,
+  "inner_val_loss": 0.10518194826300999,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11490728954474132,
+   "imperial_eskd": 0.09871546483852646,
+   "liu2021": 0.1019230904057622
+  },
+  "train_loss": 0.10190019774436951,
+  "sec": 544.7
+ },
+ {
+  "step": 10250,
+  "inner_val_loss": 0.10517366572599922,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11489470592803425,
+   "imperial_eskd": 0.09870794686404141,
+   "liu2021": 0.10191834438592196
+  },
+  "train_loss": 0.10196795886754989,
+  "sec": 558.7
+ },
+ {
+  "step": 10500,
+  "inner_val_loss": 0.10516500632319387,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11488111524118318,
+   "imperial_eskd": 0.09870063784447583,
+   "liu2021": 0.10191326588392258
+  },
+  "train_loss": 0.10213413700461388,
+  "sec": 572.5
+ },
+ {
+  "step": 10750,
+  "inner_val_loss": 0.10515121460864037,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11486043781042099,
+   "imperial_eskd": 0.09868749129501256,
+   "liu2021": 0.1019057147204876
+  },
+  "train_loss": 0.1019933986067772,
+  "sec": 586.6
+ },
+ {
+  "step": 11000,
+  "inner_val_loss": 0.10514146308743894,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11484590296943982,
+   "imperial_eskd": 0.09868138351223686,
+   "liu2021": 0.10189710278064013
+  },
+  "train_loss": 0.101762630879879,
+  "sec": 600.2
+ },
+ {
+  "step": 11250,
+  "inner_val_loss": 0.10513329303349911,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11483446053332752,
+   "imperial_eskd": 0.09867498685013164,
+   "liu2021": 0.10189043171703815
+  },
+  "train_loss": 0.10176833549141884,
+  "sec": 613.3
+ },
+ {
+  "step": 11500,
+  "inner_val_loss": 0.10512483483399908,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11482630380325848,
+   "imperial_eskd": 0.09866677766496484,
+   "liu2021": 0.1018814230337739
+  },
+  "train_loss": 0.10178321716189384,
+  "sec": 626.6
+ },
+ {
+  "step": 11750,
+  "inner_val_loss": 0.1051165730777112,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11481896042823792,
+   "imperial_eskd": 0.09865729205987671,
+   "liu2021": 0.10187346674501896
+  },
+  "train_loss": 0.10173946261405944,
+  "sec": 640.6
+ },
+ {
+  "step": 12000,
+  "inner_val_loss": 0.10510800749703188,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11481043820579846,
+   "imperial_eskd": 0.0986456915058873,
+   "liu2021": 0.10186789277940989
+  },
+  "train_loss": 0.10153279691934586,
+  "sec": 654.4
+ },
+ {
+  "step": 12250,
+  "inner_val_loss": 0.10509719447192017,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1148014979230033,
+   "imperial_eskd": 0.0986348996785554,
+   "liu2021": 0.10185518581420183
+  },
+  "train_loss": 0.10172624367475509,
+  "sec": 667.9
+ },
+ {
+  "step": 12500,
+  "inner_val_loss": 0.10508928819550108,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11479585369427998,
+   "imperial_eskd": 0.09862608428705823,
+   "liu2021": 0.101845926605165
+  },
+  "train_loss": 0.10192088502645492,
+  "sec": 681.6
+ },
+ {
+  "step": 12750,
+  "inner_val_loss": 0.10508296633396386,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11478723544213507,
+   "imperial_eskd": 0.09861639446832916,
+   "liu2021": 0.10184526909142733
+  },
+  "train_loss": 0.10169373825192451,
+  "sec": 695.1
+ },
+ {
+  "step": 13000,
+  "inner_val_loss": 0.10507559779128343,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11477896571159363,
+   "imperial_eskd": 0.09860791367563335,
+   "liu2021": 0.10183991398662329
+  },
+  "train_loss": 0.10194067013263702,
+  "sec": 708.7
+ },
+ {
+  "step": 13250,
+  "inner_val_loss": 0.10506726968531756,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1147659524447388,
+   "imperial_eskd": 0.09859639711000702,
+   "liu2021": 0.10183945950120687
+  },
+  "train_loss": 0.10183907651901244,
+  "sec": 722.7
+ },
+ {
+  "step": 13500,
+  "inner_val_loss": 0.10506032100605844,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11475799108544986,
+   "imperial_eskd": 0.09858653084798293,
+   "liu2021": 0.10183644108474255
+  },
+  "train_loss": 0.10162449651956558,
+  "sec": 736.8
+ },
+ {
+  "step": 13750,
+  "inner_val_loss": 0.10505211985472478,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11474952101707458,
+   "imperial_eskd": 0.09857827458869327,
+   "liu2021": 0.10182856395840645
+  },
+  "train_loss": 0.10184656926989555,
+  "sec": 750.7
+ },
+ {
+  "step": 14000,
+  "inner_val_loss": 0.10504524016545878,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11473936339219411,
+   "imperial_eskd": 0.0985693447291851,
+   "liu2021": 0.10182701237499714
+  },
+  "train_loss": 0.10175079995393753,
+  "sec": 763.8
+ },
+ {
+  "step": 14250,
+  "inner_val_loss": 0.1050353180711167,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1147271233300368,
+   "imperial_eskd": 0.0985614996064793,
+   "liu2021": 0.10181733127683401
+  },
+  "train_loss": 0.1013484668135643,
+  "sec": 776.3
+ },
+ {
+  "step": 14500,
+  "inner_val_loss": 0.10502553137288952,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11471731381283866,
+   "imperial_eskd": 0.09855202687057582,
+   "liu2021": 0.1018072534352541
+  },
+  "train_loss": 0.10167306759953498,
+  "sec": 789.8
+ },
+ {
+  "step": 14750,
+  "inner_val_loss": 0.10502036582829073,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11470866079131763,
+   "imperial_eskd": 0.09854502027684992,
+   "liu2021": 0.10180741641670465
+  },
+  "train_loss": 0.10195169767737389,
+  "sec": 803.9
+ },
+ {
+  "step": 15000,
+  "inner_val_loss": 0.10501315489862904,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11470130831003189,
+   "imperial_eskd": 0.09853608296676115,
+   "liu2021": 0.10180207341909409
+  },
+  "train_loss": 0.10190697094798087,
+  "sec": 817.9
+ },
+ {
+  "step": 15250,
+  "inner_val_loss": 0.10500957473159238,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11469904250568813,
+   "imperial_eskd": 0.09853208606893366,
+   "liu2021": 0.10179759562015533
+  },
+  "train_loss": 0.10119567328691483,
+  "sec": 832.0
+ },
+ {
+  "step": 15500,
+  "inner_val_loss": 0.10500261762966512,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11469116475847033,
+   "imperial_eskd": 0.0985217849639329,
+   "liu2021": 0.10179490316659212
+  },
+  "train_loss": 0.10176549535989761,
+  "sec": 846.1
+ },
+ {
+  "step": 15750,
+  "inner_val_loss": 0.10499462534858522,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11468494476543532,
+   "imperial_eskd": 0.09851273691112344,
+   "liu2021": 0.10178619436919689
+  },
+  "train_loss": 0.10150144153833389,
+  "sec": 859.9
+ },
+ {
+  "step": 16000,
+  "inner_val_loss": 0.1049899471146039,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1146812385155095,
+   "imperial_eskd": 0.09850372441790321,
+   "liu2021": 0.10178487841039896
+  },
+  "train_loss": 0.1014662393629551,
+  "sec": 873.9
+ },
+ {
+  "step": 16250,
+  "inner_val_loss": 0.10498498239701846,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11467570356196827,
+   "imperial_eskd": 0.09849964861165393,
+   "liu2021": 0.10177959501743317
+  },
+  "train_loss": 0.10177263674139976,
+  "sec": 887.9
+ },
+ {
+  "step": 16500,
+  "inner_val_loss": 0.10498222964580613,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11467620482047398,
+   "imperial_eskd": 0.09849722954359921,
+   "liu2021": 0.10177325457334518
+  },
+  "train_loss": 0.10169405376911163,
+  "sec": 901.9
+ },
+ {
+  "step": 16750,
+  "inner_val_loss": 0.10497610399857012,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11467332889636357,
+   "imperial_eskd": 0.09848950600082224,
+   "liu2021": 0.10176547709852457
+  },
+  "train_loss": 0.10155794039368629,
+  "sec": 915.9
+ },
+ {
+  "step": 17000,
+  "inner_val_loss": 0.10496982919861271,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11466270809372266,
+   "imperial_eskd": 0.09848393059589645,
+   "liu2021": 0.101762848906219
+  },
+  "train_loss": 0.10183827817440033,
+  "sec": 929.7
+ },
+ {
+  "step": 17250,
+  "inner_val_loss": 0.10496420126960233,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11465620042549239,
+   "imperial_eskd": 0.09848138317465782,
+   "liu2021": 0.10175502020865679
+  },
+  "train_loss": 0.10183451759815217,
+  "sec": 943.8
+ },
+ {
+  "step": 17500,
+  "inner_val_loss": 0.10495885972071577,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11465521570709017,
+   "imperial_eskd": 0.09847352653741837,
+   "liu2021": 0.10174783691763878
+  },
+  "train_loss": 0.10189871591329575,
+  "sec": 957.8
+ },
+ {
+  "step": 17750,
+  "inner_val_loss": 0.10495216662835594,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11464607963959376,
+   "imperial_eskd": 0.09846620058471506,
+   "liu2021": 0.10174421966075897
+  },
+  "train_loss": 0.10131221443414688,
+  "sec": 971.8
+ },
+ {
+  "step": 18000,
+  "inner_val_loss": 0.10494480635469107,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1146327435142464,
+   "imperial_eskd": 0.09845974973656914,
+   "liu2021": 0.1017419258132577
+  },
+  "train_loss": 0.10195348662137985,
+  "sec": 984.5
+ },
+ {
+  "step": 18250,
+  "inner_val_loss": 0.10493924107515451,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11462541669607162,
+   "imperial_eskd": 0.09845484827052463,
+   "liu2021": 0.10173745825886726
+  },
+  "train_loss": 0.10137604713439942,
+  "sec": 997.4
+ },
+ {
+  "step": 18500,
+  "inner_val_loss": 0.10493393938812845,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11461851083570057,
+   "imperial_eskd": 0.09845104305581613,
+   "liu2021": 0.10173226427286863
+  },
+  "train_loss": 0.10167447161674499,
+  "sec": 1010.8
+ },
+ {
+  "step": 18750,
+  "inner_val_loss": 0.10492612831984405,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11461402682794465,
+   "imperial_eskd": 0.09844493120908737,
+   "liu2021": 0.10171942692250013
+  },
+  "train_loss": 0.10137038171291352,
+  "sec": 1023.4
+ },
+ {
+  "step": 19000,
+  "inner_val_loss": 0.10492101529305832,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11460842523309919,
+   "imperial_eskd": 0.09843863309784369,
+   "liu2021": 0.10171598754823208
+  },
+  "train_loss": 0.10152953198552131,
+  "sec": 1036.3
+ },
+ {
+  "step": 19250,
+  "inner_val_loss": 0.10491311648603441,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11460336339142588,
+   "imperial_eskd": 0.09843426638028839,
+   "liu2021": 0.10170171968638897
+  },
+  "train_loss": 0.10141369438171387,
+  "sec": 1049.2
+ },
+ {
+  "step": 19500,
+  "inner_val_loss": 0.10490621786614761,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11459456466966206,
+   "imperial_eskd": 0.09842866218902847,
+   "liu2021": 0.10169542673975229
+  },
+  "train_loss": 0.10156156572699547,
+  "sec": 1063.2
+ },
+ {
+  "step": 19750,
+  "inner_val_loss": 0.10490030524091991,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11458494183089998,
+   "imperial_eskd": 0.09842559864575212,
+   "liu2021": 0.10169037524610758
+  },
+  "train_loss": 0.10142052692174912,
+  "sec": 1077.4
+ },
+ {
+  "step": 20000,
+  "inner_val_loss": 0.10489554748077083,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11457915645506647,
+   "imperial_eskd": 0.09841940551996231,
+   "liu2021": 0.10168808046728373
+  },
+  "train_loss": 0.10142033138871193,
+  "sec": 1091.4
+ },
+ {
+  "step": 20250,
+  "inner_val_loss": 0.10489161465605501,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11457257635063595,
+   "imperial_eskd": 0.09841519390994852,
+   "liu2021": 0.10168707370758057
+  },
+  "train_loss": 0.10150583201646805,
+  "sec": 1107.0
+ },
+ {
+  "step": 20500,
+  "inner_val_loss": 0.10488490941260098,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11456103954050276,
+   "imperial_eskd": 0.09840734235265038,
+   "liu2021": 0.10168634634464979
+  },
+  "train_loss": 0.10128593027591705,
+  "sec": 1121.4
+ },
+ {
+  "step": 20750,
+  "inner_val_loss": 0.1048789765052372,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11455841528044806,
+   "imperial_eskd": 0.09840123897249048,
+   "liu2021": 0.10167727526277304
+  },
+  "train_loss": 0.10140911814570427,
+  "sec": 1135.7
+ },
+ {
+  "step": 21000,
+  "inner_val_loss": 0.10487590796354945,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11455210505260362,
+   "imperial_eskd": 0.09839683669534596,
+   "liu2021": 0.10167878214269876
+  },
+  "train_loss": 0.1016389653980732,
+  "sec": 1149.5
+ },
+ {
+  "step": 21250,
+  "inner_val_loss": 0.10487071303054941,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11454674642946985,
+   "imperial_eskd": 0.09839246015657079,
+   "liu2021": 0.1016729325056076
+  },
+  "train_loss": 0.10116986578702927,
+  "sec": 1163.3
+ },
+ {
+  "step": 21500,
+  "inner_val_loss": 0.10486207612059534,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1145364675256941,
+   "imperial_eskd": 0.098383751782504,
+   "liu2021": 0.10166600905358791
+  },
+  "train_loss": 0.10145043817162513,
+  "sec": 1177.1
+ },
+ {
+  "step": 21750,
+  "inner_val_loss": 0.10485584766726301,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11453113580743472,
+   "imperial_eskd": 0.09837980636141518,
+   "liu2021": 0.10165660083293915
+  },
+  "train_loss": 0.10145993492007256,
+  "sec": 1190.9
+ },
+ {
+  "step": 22000,
+  "inner_val_loss": 0.10485372720568469,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11453618854284286,
+   "imperial_eskd": 0.09837404164400967,
+   "liu2021": 0.10165095143020153
+  },
+  "train_loss": 0.1013836508989334,
+  "sec": 1205.0
+ },
+ {
+  "step": 22250,
+  "inner_val_loss": 0.10484638666226105,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11451999843120575,
+   "imperial_eskd": 0.09836756471883167,
+   "liu2021": 0.10165159683674574
+  },
+  "train_loss": 0.10149006056785584,
+  "sec": 1218.7
+ },
+ {
+  "step": 22500,
+  "inner_val_loss": 0.10484208071899133,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1145163161887063,
+   "imperial_eskd": 0.09836185046217659,
+   "liu2021": 0.10164807550609112
+  },
+  "train_loss": 0.10135751581192017,
+  "sec": 1232.6
+ },
+ {
+  "step": 22750,
+  "inner_val_loss": 0.10483940437072355,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11451525572273466,
+   "imperial_eskd": 0.09835535592653534,
+   "liu2021": 0.10164760146290064
+  },
+  "train_loss": 0.10130930358171462,
+  "sec": 1246.6
+ },
+ {
+  "step": 23000,
+  "inner_val_loss": 0.10483304478905418,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11451093852519989,
+   "imperial_eskd": 0.0983478037471121,
+   "liu2021": 0.10164039209485054
+  },
+  "train_loss": 0.10172589853405953,
+  "sec": 1260.4
+ },
+ {
+  "step": 23250,
+  "inner_val_loss": 0.10483177419538652,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11451392827762498,
+   "imperial_eskd": 0.09834461388262836,
+   "liu2021": 0.10163678042590618
+  },
+  "train_loss": 0.1016272228360176,
+  "sec": 1274.4
+ },
+ {
+  "step": 23500,
+  "inner_val_loss": 0.10482673460197468,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1145066987309191,
+   "imperial_eskd": 0.09834146601232616,
+   "liu2021": 0.10163203906267881
+  },
+  "train_loss": 0.10143477946519852,
+  "sec": 1288.3
+ },
+ {
+  "step": 23750,
+  "inner_val_loss": 0.1048204299150292,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11449719924065802,
+   "imperial_eskd": 0.09834089062430641,
+   "liu2021": 0.10162319988012314
+  },
+  "train_loss": 0.10128503468632698,
+  "sec": 1302.3
+ },
+ {
+  "step": 24000,
+  "inner_val_loss": 0.1048161681139409,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11449284230669339,
+   "imperial_eskd": 0.09833525053479454,
+   "liu2021": 0.10162041150033474
+  },
+  "train_loss": 0.10178426477313042,
+  "sec": 1316.0
+ },
+ {
+  "step": 24250,
+  "inner_val_loss": 0.10480981964542709,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11448755363623302,
+   "imperial_eskd": 0.09832808476957408,
+   "liu2021": 0.10161382053047419
+  },
+  "train_loss": 0.10120327860116958,
+  "sec": 1330.0
+ },
+ {
+  "step": 24500,
+  "inner_val_loss": 0.1047987552133925,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11447467654943466,
+   "imperial_eskd": 0.09831875257871368,
+   "liu2021": 0.10160283651202917
+  },
+  "train_loss": 0.10122303530573845,
+  "sec": 1343.8
+ },
+ {
+  "step": 24750,
+  "inner_val_loss": 0.10479422731745223,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11447463225987223,
+   "imperial_eskd": 0.0983151373538104,
+   "liu2021": 0.10159291233867407
+  },
+  "train_loss": 0.10142298182845115,
+  "sec": 1357.7
+ },
+ {
+  "step": 25000,
+  "inner_val_loss": 0.10479024319473641,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1144694553481208,
+   "imperial_eskd": 0.098310841078108,
+   "liu2021": 0.10159043315798044
+  },
+  "train_loss": 0.10126602241396904,
+  "sec": 1371.6
+ },
+ {
+  "step": 25250,
+  "inner_val_loss": 0.10478619472609725,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11446519485778278,
+   "imperial_eskd": 0.09830766679211096,
+   "liu2021": 0.10158572252839804
+  },
+  "train_loss": 0.10134903532266616,
+  "sec": 1385.3
+ },
+ {
+  "step": 25500,
+  "inner_val_loss": 0.1047837770001505,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11446282350354725,
+   "imperial_eskd": 0.09830202500928532,
+   "liu2021": 0.10158648248761892
+  },
+  "train_loss": 0.10110112512111664,
+  "sec": 1399.1
+ },
+ {
+  "step": 25750,
+  "inner_val_loss": 0.10478071059705533,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11446490718258752,
+   "imperial_eskd": 0.09829706393859604,
+   "liu2021": 0.10158016066998243
+  },
+  "train_loss": 0.10135788387060166,
+  "sec": 1413.5
+ },
+ {
+  "step": 26000,
+  "inner_val_loss": 0.10477391609715091,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11445754145582517,
+   "imperial_eskd": 0.09828934818506241,
+   "liu2021": 0.10157485865056515
+  },
+  "train_loss": 0.10147871053218842,
+  "sec": 1427.7
+ },
+ {
+  "step": 26250,
+  "inner_val_loss": 0.104768317888933,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11444770668943723,
+   "imperial_eskd": 0.09828680652109059,
+   "liu2021": 0.10157044045627117
+  },
+  "train_loss": 0.10129926809668541,
+  "sec": 1442.2
+ },
+ {
+  "step": 26500,
+  "inner_val_loss": 0.10476366204432265,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11444401037361887,
+   "imperial_eskd": 0.09828174791552803,
+   "liu2021": 0.10156522784382105
+  },
+  "train_loss": 0.10137804800271988,
+  "sec": 1456.3
+ },
+ {
+  "step": 26750,
+  "inner_val_loss": 0.10475878555779204,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11443626508116722,
+   "imperial_eskd": 0.0982765606181188,
+   "liu2021": 0.1015635309740901
+  },
+  "train_loss": 0.1008530935049057,
+  "sec": 1470.1
+ },
+ {
+  "step": 27000,
+  "inner_val_loss": 0.10475592464145889,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11443270080619389,
+   "imperial_eskd": 0.09827316078272733,
+   "liu2021": 0.10156191233545542
+  },
+  "train_loss": 0.10135508570075034,
+  "sec": 1483.9
+ },
+ {
+  "step": 27250,
+  "inner_val_loss": 0.10475319670042908,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11443017298976581,
+   "imperial_eskd": 0.09826996346766298,
+   "liu2021": 0.10155945364385843
+  },
+  "train_loss": 0.10137590053677559,
+  "sec": 1497.6
+ },
+ {
+  "step": 27500,
+  "inner_val_loss": 0.10474823382040266,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11442345297998852,
+   "imperial_eskd": 0.09826642816716974,
+   "liu2021": 0.10155482031404972
+  },
+  "train_loss": 0.1014763517677784,
+  "sec": 1511.3
+ },
+ {
+  "step": 27750,
+  "inner_val_loss": 0.10474542165896306,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11441936509476767,
+   "imperial_eskd": 0.09826177595691248,
+   "liu2021": 0.10155512392520905
+  },
+  "train_loss": 0.10150577774643899,
+  "sec": 1525.3
+ },
+ {
+  "step": 28000,
+  "inner_val_loss": 0.1047444206502552,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11442530776063602,
+   "imperial_eskd": 0.0982589183205908,
+   "liu2021": 0.10154903586953878
+  },
+  "train_loss": 0.10121049109101296,
+  "sec": 1539.7
+ },
+ {
+  "step": 28250,
+  "inner_val_loss": 0.10474508706016172,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11442782150374518,
+   "imperial_eskd": 0.09826007925651291,
+   "liu2021": 0.10154736042022705
+  },
+  "train_loss": 0.10121887445449829,
+  "sec": 1553.9
+ },
+ {
+  "step": 28500,
+  "inner_val_loss": 0.10473902338820977,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11441384421454535,
+   "imperial_eskd": 0.0982600145719268,
+   "liu2021": 0.10154321137815714
+  },
+  "train_loss": 0.10106964826583863,
+  "sec": 1565.9
+ },
+ {
+  "step": 28750,
+  "inner_val_loss": 0.10473195935978934,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11440798847211732,
+   "imperial_eskd": 0.09825469553470612,
+   "liu2021": 0.10153319407254457
+  },
+  "train_loss": 0.10119320034980774,
+  "sec": 1578.2
+ },
+ {
+  "step": 29000,
+  "inner_val_loss": 0.10472628687154062,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11439722238315476,
+   "imperial_eskd": 0.09824751656163823,
+   "liu2021": 0.10153412166982889
+  },
+  "train_loss": 0.10114939606189728,
+  "sec": 1590.8
+ },
+ {
+  "step": 29250,
+  "inner_val_loss": 0.10472784135239795,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11439732048246595,
+   "imperial_eskd": 0.09824241663921963,
+   "liu2021": 0.10154378693550825
+  },
+  "train_loss": 0.10162037146091461,
+  "sec": 1604.2
+ },
+ {
+  "step": 29500,
+  "inner_val_loss": 0.10472598650903513,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1143999302552806,
+   "imperial_eskd": 0.09823766215281053,
+   "liu2021": 0.10154036711901426
+  },
+  "train_loss": 0.10093125811219215,
+  "sec": 1616.1
+ },
+ {
+  "step": 29750,
+  "inner_val_loss": 0.10472206181849696,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11439583781692717,
+   "imperial_eskd": 0.09823472933335738,
+   "liu2021": 0.1015356183052063
+  },
+  "train_loss": 0.10113678640127181,
+  "sec": 1628.4
+ },
+ {
+  "step": 30000,
+  "inner_val_loss": 0.10472034292000813,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1143946030901538,
+   "imperial_eskd": 0.09823335639455101,
+   "liu2021": 0.10153306927531958
+  },
+  "train_loss": 0.10078215217590332,
+  "sec": 1640.8
+ },
+ {
+  "step": 30250,
+  "inner_val_loss": 0.10471574808486583,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11439270319210158,
+   "imperial_eskd": 0.09822774238207123,
+   "liu2021": 0.10152679868042469
+  },
+  "train_loss": 0.1010243653357029,
+  "sec": 1653.9
+ },
+ {
+  "step": 30500,
+  "inner_val_loss": 0.1047128217752634,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1143859355813927,
+   "imperial_eskd": 0.09822356667030942,
+   "liu2021": 0.1015289630740881
+  },
+  "train_loss": 0.10108919262886047,
+  "sec": 1666.3
+ },
+ {
+  "step": 30750,
+  "inner_val_loss": 0.10471110719786048,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11438792281680638,
+   "imperial_eskd": 0.09821948205882852,
+   "liu2021": 0.10152591671794653
+  },
+  "train_loss": 0.10102685749530792,
+  "sec": 1678.9
+ },
+ {
+  "step": 31000,
+  "inner_val_loss": 0.1047051805620292,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11437807066573037,
+   "imperial_eskd": 0.09821979091926054,
+   "liu2021": 0.10151768010109663
+  },
+  "train_loss": 0.10099230155348778,
+  "sec": 1691.6
+ },
+ {
+  "step": 31250,
+  "inner_val_loss": 0.10470543527091392,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11437827472885449,
+   "imperial_eskd": 0.09821495583111589,
+   "liu2021": 0.10152307525277138
+  },
+  "train_loss": 0.1010059505701065,
+  "sec": 1704.7
+ },
+ {
+  "step": 31500,
+  "inner_val_loss": 0.10469865058282209,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1143732944296466,
+   "imperial_eskd": 0.09820861775766719,
+   "liu2021": 0.10151403956115246
+  },
+  "train_loss": 0.1010197232067585,
+  "sec": 1716.5
+ },
+ {
+  "step": 31750,
+  "inner_val_loss": 0.10469251587322163,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11436479373110665,
+   "imperial_eskd": 0.0982008535753597,
+   "liu2021": 0.10151190031319857
+  },
+  "train_loss": 0.10099384135007858,
+  "sec": 1729.4
+ },
+ {
+  "step": 32000,
+  "inner_val_loss": 0.10468945550020496,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11436277710729176,
+   "imperial_eskd": 0.09819646721536462,
+   "liu2021": 0.10150912217795849
+  },
+  "train_loss": 0.10104307064414024,
+  "sec": 1742.6
+ },
+ {
+  "step": 32250,
+  "inner_val_loss": 0.10468780003826728,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11436515301465988,
+   "imperial_eskd": 0.09819884699853984,
+   "liu2021": 0.10149940010160208
+  },
+  "train_loss": 0.10101970991492272,
+  "sec": 1755.2
+ },
+ {
+  "step": 32500,
+  "inner_val_loss": 0.10468649600792412,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11436071371038754,
+   "imperial_eskd": 0.09819999167864973,
+   "liu2021": 0.10149878263473511
+  },
+  "train_loss": 0.10114361238479615,
+  "sec": 1767.7
+ },
+ {
+  "step": 32750,
+  "inner_val_loss": 0.10468370913330353,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1143605150282383,
+   "imperial_eskd": 0.09819195453416217,
+   "liu2021": 0.10149865783751011
+  },
+  "train_loss": 0.10106919485330582,
+  "sec": 1780.8
+ },
+ {
+  "step": 33000,
+  "inner_val_loss": 0.10467844414414983,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11435454669925901,
+   "imperial_eskd": 0.09818497401746837,
+   "liu2021": 0.10149581171572208
+  },
+  "train_loss": 0.10080297976732254,
+  "sec": 1794.7
+ },
+ {
+  "step": 33250,
+  "inner_val_loss": 0.10467466865964109,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11434902788864242,
+   "imperial_eskd": 0.09818583557551558,
+   "liu2021": 0.10148914251476526
+  },
+  "train_loss": 0.10110863274335861,
+  "sec": 1808.5
+ },
+ {
+  "step": 33500,
+  "inner_val_loss": 0.10467363049210321,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1143535686035951,
+   "imperial_eskd": 0.09818310798569159,
+   "liu2021": 0.10148421488702297
+  },
+  "train_loss": 0.10058023223280907,
+  "sec": 1822.5
+ },
+ {
+  "step": 33750,
+  "inner_val_loss": 0.10466880023931012,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11434577777981758,
+   "imperial_eskd": 0.0981777659194036,
+   "liu2021": 0.10148285701870918
+  },
+  "train_loss": 0.10091716206073761,
+  "sec": 1836.3
+ },
+ {
+  "step": 34000,
+  "inner_val_loss": 0.10466396098373214,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11434573555986087,
+   "imperial_eskd": 0.09817276657982306,
+   "liu2021": 0.10147338081151247
+  },
+  "train_loss": 0.10119396695494652,
+  "sec": 1850.2
+ },
+ {
+  "step": 34250,
+  "inner_val_loss": 0.10466132580180361,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11434026931722958,
+   "imperial_eskd": 0.09816999665715477,
+   "liu2021": 0.10147371143102646
+  },
+  "train_loss": 0.10131568855047227,
+  "sec": 1863.9
+ },
+ {
+  "step": 34500,
+  "inner_val_loss": 0.1046578897577193,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1143334445854028,
+   "imperial_eskd": 0.09816382825374603,
+   "liu2021": 0.10147639643400908
+  },
+  "train_loss": 0.10130748730897904,
+  "sec": 1877.8
+ },
+ {
+  "step": 34750,
+  "inner_val_loss": 0.1046538569835902,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11433210430873765,
+   "imperial_eskd": 0.098161615092646,
+   "liu2021": 0.10146785154938698
+  },
+  "train_loss": 0.10112025767564774,
+  "sec": 1891.6
+ },
+ {
+  "step": 35000,
+  "inner_val_loss": 0.10465070819137273,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11432617985539967,
+   "imperial_eskd": 0.09816128760576248,
+   "liu2021": 0.10146465711295605
+  },
+  "train_loss": 0.1011915420293808,
+  "sec": 1905.5
+ },
+ {
+  "step": 35250,
+  "inner_val_loss": 0.10464763467008818,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11432113043136066,
+   "imperial_eskd": 0.09816107086159966,
+   "liu2021": 0.10146070271730423
+  },
+  "train_loss": 0.1011680948138237,
+  "sec": 1919.5
+ },
+ {
+  "step": 35500,
+  "inner_val_loss": 0.10464511093329198,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11432054473294152,
+   "imperial_eskd": 0.09815964813936841,
+   "liu2021": 0.10145513992756605
+  },
+  "train_loss": 0.10139237797260284,
+  "sec": 1933.7
+ },
+ {
+  "step": 35750,
+  "inner_val_loss": 0.1046426770832352,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11431298529108365,
+   "imperial_eskd": 0.09815875677899881,
+   "liu2021": 0.10145628917962313
+  },
+  "train_loss": 0.10098293110728264,
+  "sec": 1947.9
+ },
+ {
+  "step": 36000,
+  "inner_val_loss": 0.10463791953004671,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11430980265140533,
+   "imperial_eskd": 0.09815613620660522,
+   "liu2021": 0.10144781973212957
+  },
+  "train_loss": 0.10123196741938591,
+  "sec": 1962.3
+ },
+ {
+  "step": 36250,
+  "inner_val_loss": 0.10463105731327023,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11430408888392979,
+   "imperial_eskd": 0.09814789247783748,
+   "liu2021": 0.10144119057804346
+  },
+  "train_loss": 0.10112023383378982,
+  "sec": 1976.6
+ },
+ {
+  "step": 36500,
+  "inner_val_loss": 0.10462561436311063,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11429739329550001,
+   "imperial_eskd": 0.09814575077457861,
+   "liu2021": 0.10143369901925325
+  },
+  "train_loss": 0.10068220686912537,
+  "sec": 1991.1
+ },
+ {
+  "step": 36750,
+  "inner_val_loss": 0.10462106783444734,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11429691024952465,
+   "imperial_eskd": 0.09813820265910843,
+   "liu2021": 0.10142809059470892
+  },
+  "train_loss": 0.10129327425360679,
+  "sec": 2005.5
+ },
+ {
+  "step": 37000,
+  "inner_val_loss": 0.10461723314980766,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11429597975479232,
+   "imperial_eskd": 0.09813530878587202,
+   "liu2021": 0.10142041090875864
+  },
+  "train_loss": 0.10066738468408584,
+  "sec": 2019.4
+ },
+ {
+  "step": 37250,
+  "inner_val_loss": 0.10461752418027281,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11429327436619335,
+   "imperial_eskd": 0.09813372960144823,
+   "liu2021": 0.10142556857317686
+  },
+  "train_loss": 0.1010494943857193,
+  "sec": 2033.6
+ },
+ {
+  "step": 37500,
+  "inner_val_loss": 0.10461402922141232,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428940958446926,
+   "imperial_eskd": 0.0981301912529902,
+   "liu2021": 0.10142248682677746
+  },
+  "train_loss": 0.10110481590032577,
+  "sec": 2046.1
+ },
+ {
+  "step": 37750,
+  "inner_val_loss": 0.1046139640915474,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428713260425462,
+   "imperial_eskd": 0.09812718968499791,
+   "liu2021": 0.10142756998538971
+  },
+  "train_loss": 0.10111588296294212,
+  "sec": 2060.1
+ },
+ {
+  "step": 38000,
+  "inner_val_loss": 0.10460849841335407,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428456587923898,
+   "imperial_eskd": 0.09812285141511397,
+   "liu2021": 0.10141807794570923
+  },
+  "train_loss": 0.10113478055596352,
+  "sec": 2072.5
+ },
+ {
+  "step": 38250,
+  "inner_val_loss": 0.10460401378471644,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428388787640466,
+   "imperial_eskd": 0.0981196364895864,
+   "liu2021": 0.10140851698815823
+  },
+  "train_loss": 0.10120171630382538,
+  "sec": 2086.4
+ },
+ {
+  "step": 38500,
+  "inner_val_loss": 0.10460010711542685,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428313867913352,
+   "imperial_eskd": 0.09811471741307866,
+   "liu2021": 0.10140246525406837
+  },
+  "train_loss": 0.1010114193558693,
+  "sec": 2100.3
+ },
+ {
+  "step": 38750,
+  "inner_val_loss": 0.10459748699144783,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1142796782983674,
+   "imperial_eskd": 0.09811176562851126,
+   "liu2021": 0.10140101704746485
+  },
+  "train_loss": 0.1009778998196125,
+  "sec": 2113.9
+ },
+ {
+  "step": 39000,
+  "inner_val_loss": 0.10459775014181531,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428770505719715,
+   "imperial_eskd": 0.09811333025043661,
+   "liu2021": 0.10139221511781216
+  },
+  "train_loss": 0.1008552266061306,
+  "sec": 2128.0
+ },
+ {
+  "step": 39250,
+  "inner_val_loss": 0.10459477336519092,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428501746720737,
+   "imperial_eskd": 0.09811094877394763,
+   "liu2021": 0.1013883538544178
+  },
+  "train_loss": 0.10095585817098618,
+  "sec": 2140.5
+ },
+ {
+  "step": 39500,
+  "inner_val_loss": 0.10459168642282686,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.114283363852236,
+   "imperial_eskd": 0.09810621562329205,
+   "liu2021": 0.10138547979295254
+  },
+  "train_loss": 0.10111479422450066,
+  "sec": 2154.5
+ },
+ {
+  "step": 39750,
+  "inner_val_loss": 0.10459061092085609,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428860823313396,
+   "imperial_eskd": 0.09810198132287372,
+   "liu2021": 0.10138124320656061
+  },
+  "train_loss": 0.10095193982124329,
+  "sec": 2168.5
+ },
+ {
+  "step": 40000,
+  "inner_val_loss": 0.10458750774776594,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428565863105986,
+   "imperial_eskd": 0.09810030595822768,
+   "liu2021": 0.1013765586540103
+  },
+  "train_loss": 0.1008045875430107,
+  "sec": 2182.6
+ },
+ {
+  "step": 40250,
+  "inner_val_loss": 0.10458788128651253,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1142920781340864,
+   "imperial_eskd": 0.09809902107173746,
+   "liu2021": 0.1013725446537137
+  },
+  "train_loss": 0.10121687072515488,
+  "sec": 2198.1
+ },
+ {
+  "step": 40500,
+  "inner_val_loss": 0.10458204017870554,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428588214847776,
+   "imperial_eskd": 0.09809551963751967,
+   "liu2021": 0.10136471875011921
+  },
+  "train_loss": 0.10120873755216599,
+  "sec": 2210.6
+ },
+ {
+  "step": 40750,
+  "inner_val_loss": 0.10457937434152621,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428508079714245,
+   "imperial_eskd": 0.0980918559838425,
+   "liu2021": 0.10136118624359369
+  },
+  "train_loss": 0.10078801262378692,
+  "sec": 2224.3
+ },
+ {
+  "step": 41000,
+  "inner_val_loss": 0.10457700867243487,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11427975694338481,
+   "imperial_eskd": 0.09808800877495245,
+   "liu2021": 0.10136326029896736
+  },
+  "train_loss": 0.10095579394698143,
+  "sec": 2238.2
+ },
+ {
+  "step": 41250,
+  "inner_val_loss": 0.10457460029420729,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1142765912744734,
+   "imperial_eskd": 0.09808610345829617,
+   "liu2021": 0.10136110614985228
+  },
+  "train_loss": 0.10098979333043098,
+  "sec": 2252.2
+ },
+ {
+  "step": 41500,
+  "inner_val_loss": 0.10457410359824144,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11427582262290849,
+   "imperial_eskd": 0.09808771684765816,
+   "liu2021": 0.10135877132415771
+  },
+  "train_loss": 0.10098774883151054,
+  "sec": 2266.2
+ },
+ {
+  "step": 41750,
+  "inner_val_loss": 0.10457379017526956,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11428307452135616,
+   "imperial_eskd": 0.09808411889455536,
+   "liu2021": 0.10135417710989714
+  },
+  "train_loss": 0.10081213882565498,
+  "sec": 2278.6
+ },
+ {
+  "step": 42000,
+  "inner_val_loss": 0.10456758948914731,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11427025165822771,
+   "imperial_eskd": 0.09807965193282474,
+   "liu2021": 0.1013528648763895
+  },
+  "train_loss": 0.10084121888875962,
+  "sec": 2291.1
+ },
+ {
+  "step": 42250,
+  "inner_val_loss": 0.10456586836740023,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1142669485674964,
+   "imperial_eskd": 0.09807883473959836,
+   "liu2021": 0.10135182179510593
+  },
+  "train_loss": 0.10054939004778862,
+  "sec": 2305.0
+ },
+ {
+  "step": 42500,
+  "inner_val_loss": 0.10456485093354878,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11426732564965884,
+   "imperial_eskd": 0.0980773229490627,
+   "liu2021": 0.1013499042019248
+  },
+  "train_loss": 0.10112120363116264,
+  "sec": 2319.0
+ },
+ {
+  "step": 42750,
+  "inner_val_loss": 0.10456373298584613,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11426171453462707,
+   "imperial_eskd": 0.09807594526897777,
+   "liu2021": 0.10135353915393353
+  },
+  "train_loss": 0.10070882076025009,
+  "sec": 2333.0
+ },
+ {
+  "step": 43000,
+  "inner_val_loss": 0.104562952518714,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11425535794761446,
+   "imperial_eskd": 0.09807269241322171,
+   "liu2021": 0.10136080719530582
+  },
+  "train_loss": 0.10071584907174111,
+  "sec": 2347.0
+ },
+ {
+  "step": 43250,
+  "inner_val_loss": 0.10455797762243134,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.114246666431427,
+   "imperial_eskd": 0.09807205742055719,
+   "liu2021": 0.10135520901530981
+  },
+  "train_loss": 0.10119371819496155,
+  "sec": 2359.5
+ },
+ {
+  "step": 43500,
+  "inner_val_loss": 0.10455920630658315,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11424751372800933,
+   "imperial_eskd": 0.09807503494349393,
+   "liu2021": 0.10135507024824619
+  },
+  "train_loss": 0.10085230925679207,
+  "sec": 2373.2
+ },
+ {
+  "step": 43750,
+  "inner_val_loss": 0.1045561915715084,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11424430459737778,
+   "imperial_eskd": 0.09806829182939096,
+   "liu2021": 0.10135597828775644
+  },
+  "train_loss": 0.10134132665395737,
+  "sec": 2385.7
+ },
+ {
+  "step": 44000,
+  "inner_val_loss": 0.10454758738327508,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11423313369353612,
+   "imperial_eskd": 0.09806136245077307,
+   "liu2021": 0.10134826600551605
+  },
+  "train_loss": 0.10071848747134209,
+  "sec": 2399.5
+ },
+ {
+  "step": 44250,
+  "inner_val_loss": 0.10454427988052067,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1142247940103213,
+   "imperial_eskd": 0.0980606075714935,
+   "liu2021": 0.10134743805974722
+  },
+  "train_loss": 0.10072702571749687,
+  "sec": 2413.4
+ },
+ {
+  "step": 44500,
+  "inner_val_loss": 0.10454359218816867,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1142258308827877,
+   "imperial_eskd": 0.09805955094369975,
+   "liu2021": 0.10134539473801851
+  },
+  "train_loss": 0.10089870867133141,
+  "sec": 2427.1
+ },
+ {
+  "step": 44750,
+  "inner_val_loss": 0.1045441410726921,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11422853213217524,
+   "imperial_eskd": 0.09806267984888771,
+   "liu2021": 0.10134121123701334
+  },
+  "train_loss": 0.10088296815752983,
+  "sec": 2439.6
+ },
+ {
+  "step": 45000,
+  "inner_val_loss": 0.1045377387975653,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11421724781394005,
+   "imperial_eskd": 0.09805592149496078,
+   "liu2021": 0.10134004708379507
+  },
+  "train_loss": 0.10076918339729309,
+  "sec": 2452.2
+ },
+ {
+  "step": 45250,
+  "inner_val_loss": 0.10453468054746236,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11421317027674781,
+   "imperial_eskd": 0.09805360334840688,
+   "liu2021": 0.10133726801723242
+  },
+  "train_loss": 0.10096180698275566,
+  "sec": 2466.1
+ },
+ {
+  "step": 45500,
+  "inner_val_loss": 0.10453060865778514,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11421154315272967,
+   "imperial_eskd": 0.09805007515983148,
+   "liu2021": 0.10133020766079426
+  },
+  "train_loss": 0.10129521849751473,
+  "sec": 2479.9
+ },
+ {
+  "step": 45750,
+  "inner_val_loss": 0.10452604324867328,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11420799046754837,
+   "imperial_eskd": 0.09805088117718697,
+   "liu2021": 0.1013192581012845
+  },
+  "train_loss": 0.10098396900296211,
+  "sec": 2493.7
+ },
+ {
+  "step": 46000,
+  "inner_val_loss": 0.1045264209078167,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11420429580741459,
+   "imperial_eskd": 0.098051901568066,
+   "liu2021": 0.10132306534796953
+  },
+  "train_loss": 0.10103354805707931,
+  "sec": 2507.4
+ },
+ {
+  "step": 46250,
+  "inner_val_loss": 0.10452590084412082,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1142048906120989,
+   "imperial_eskd": 0.09805378758094528,
+   "liu2021": 0.10131902433931828
+  },
+  "train_loss": 0.10065366372466088,
+  "sec": 2519.9
+ },
+ {
+  "step": 46500,
+  "inner_val_loss": 0.10452433037712718,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11420110613107681,
+   "imperial_eskd": 0.09805613146586852,
+   "liu2021": 0.10131575353443623
+  },
+  "train_loss": 0.10093317660689353,
+  "sec": 2532.4
+ },
+ {
+  "step": 46750,
+  "inner_val_loss": 0.1045211835823928,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11420217487547132,
+   "imperial_eskd": 0.09804927591573108,
+   "liu2021": 0.10131209995597601
+  },
+  "train_loss": 0.10078494611382484,
+  "sec": 2546.2
+ },
+ {
+  "step": 47000,
+  "inner_val_loss": 0.10451951431367634,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11419935483071539,
+   "imperial_eskd": 0.09804776243188164,
+   "liu2021": 0.10131142567843199
+  },
+  "train_loss": 0.10082750669121743,
+  "sec": 2560.1
+ },
+ {
+  "step": 47250,
+  "inner_val_loss": 0.10451842683621428,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11419861018657684,
+   "imperial_eskd": 0.0980422550981695,
+   "liu2021": 0.1013144152238965
+  },
+  "train_loss": 0.1008736689388752,
+  "sec": 2574.0
+ },
+ {
+  "step": 47500,
+  "inner_val_loss": 0.1045177624018405,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11419884198241764,
+   "imperial_eskd": 0.09803837503899228,
+   "liu2021": 0.1013160701841116
+  },
+  "train_loss": 0.1009077908694744,
+  "sec": 2587.8
+ },
+ {
+  "step": 47750,
+  "inner_val_loss": 0.1045140047598417,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11420127583874597,
+   "imperial_eskd": 0.09803415191444484,
+   "liu2021": 0.10130658652633429
+  },
+  "train_loss": 0.10066605165600777,
+  "sec": 2600.3
+ },
+ {
+  "step": 48000,
+  "inner_val_loss": 0.1045069520985739,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11419559973809454,
+   "imperial_eskd": 0.09802893041209741,
+   "liu2021": 0.10129632614552975
+  },
+  "train_loss": 0.10084091341495514,
+  "sec": 2614.0
+ },
+ {
+  "step": 48250,
+  "inner_val_loss": 0.10450281804233086,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11419298085901472,
+   "imperial_eskd": 0.09802338277751749,
+   "liu2021": 0.1012920904904604
+  },
+  "train_loss": 0.10097016444802284,
+  "sec": 2627.8
+ },
+ {
+  "step": 48500,
+  "inner_val_loss": 0.10450024243055368,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11419148660368389,
+   "imperial_eskd": 0.09802021052349698,
+   "liu2021": 0.10128903016448021
+  },
+  "train_loss": 0.10059801062941551,
+  "sec": 2641.7
+ },
+ {
+  "step": 48750,
+  "inner_val_loss": 0.10449881727496783,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11419137194752693,
+   "imperial_eskd": 0.09801710397005081,
+   "liu2021": 0.10128797590732574
+  },
+  "train_loss": 0.10058913278579712,
+  "sec": 2655.3
+ },
+ {
+  "step": 49000,
+  "inner_val_loss": 0.10450071787856745,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.114189263433218,
+   "imperial_eskd": 0.09801689840175888,
+   "liu2021": 0.10129599180072546
+  },
+  "train_loss": 0.10085332423448562,
+  "sec": 2669.0
+ },
+ {
+  "step": 49250,
+  "inner_val_loss": 0.10449561826031818,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11418121018343502,
+   "imperial_eskd": 0.09801097993146289,
+   "liu2021": 0.10129466466605663
+  },
+  "train_loss": 0.10081647732853889,
+  "sec": 2681.5
+ },
+ {
+  "step": 49500,
+  "inner_val_loss": 0.10449129240179704,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11418103219734298,
+   "imperial_eskd": 0.09800403463569554,
+   "liu2021": 0.1012888103723526
+  },
+  "train_loss": 0.10064074900746346,
+  "sec": 2695.4
+ },
+ {
+  "step": 49750,
+  "inner_val_loss": 0.10448835715526082,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11417523192034827,
+   "imperial_eskd": 0.09800371324474161,
+   "liu2021": 0.10128612630069256
+  },
+  "train_loss": 0.10082043588161468,
+  "sec": 2709.2
+ },
+ {
+  "step": 50000,
+  "inner_val_loss": 0.10448843315996305,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11417485152681668,
+   "imperial_eskd": 0.09800426391038028,
+   "liu2021": 0.10128618404269218
+  },
+  "train_loss": 0.10035441419482231,
+  "sec": 2723.1
+ },
+ {
+  "step": 50250,
+  "inner_val_loss": 0.10448348186659291,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11417116555902693,
+   "imperial_eskd": 0.09799927066672932,
+   "liu2021": 0.10128000937402248
+  },
+  "train_loss": 0.10068000656366348,
+  "sec": 2735.6
+ },
+ {
+  "step": 50500,
+  "inner_val_loss": 0.10447978666710733,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11416583756605785,
+   "imperial_eskd": 0.09799836338921027,
+   "liu2021": 0.10127515904605389
+  },
+  "train_loss": 0.10093591359257698,
+  "sec": 2749.2
+ },
+ {
+  "step": 50750,
+  "inner_val_loss": 0.1044744543781385,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11415959977441364,
+   "imperial_eskd": 0.09799557619474152,
+   "liu2021": 0.10126818716526031
+  },
+  "train_loss": 0.10043725880980492,
+  "sec": 2762.8
+ },
+ {
+  "step": 51000,
+  "inner_val_loss": 0.10447142945182343,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11415502760145399,
+   "imperial_eskd": 0.09799149734052745,
+   "liu2021": 0.10126776341348886
+  },
+  "train_loss": 0.100808160841465,
+  "sec": 2776.7
+ },
+ {
+  "step": 51250,
+  "inner_val_loss": 0.10446746071317681,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11415178370144632,
+   "imperial_eskd": 0.09798942506313324,
+   "liu2021": 0.10126117337495089
+  },
+  "train_loss": 0.10066160064935684,
+  "sec": 2790.4
+ },
+ {
+  "step": 51500,
+  "inner_val_loss": 0.1044602655219259,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1141371942228741,
+   "imperial_eskd": 0.0979885570704937,
+   "liu2021": 0.10125504527240992
+  },
+  "train_loss": 0.10068286058306694,
+  "sec": 2804.1
+ },
+ {
+  "step": 51750,
+  "inner_val_loss": 0.10445846238734846,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11413837638166216,
+   "imperial_eskd": 0.097986754368652,
+   "liu2021": 0.10125025641173124
+  },
+  "train_loss": 0.10071042996644974,
+  "sec": 2818.0
+ },
+ {
+  "step": 52000,
+  "inner_val_loss": 0.10445561741325089,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11412973494993316,
+   "imperial_eskd": 0.09798631512305954,
+   "liu2021": 0.10125080216675997
+  },
+  "train_loss": 0.10093105527758599,
+  "sec": 2831.8
+ },
+ {
+  "step": 52250,
+  "inner_val_loss": 0.10445165299569015,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11412214322222604,
+   "imperial_eskd": 0.09798161312937737,
+   "liu2021": 0.10125120263546705
+  },
+  "train_loss": 0.1007467679977417,
+  "sec": 2845.7
+ },
+ {
+  "step": 52500,
+  "inner_val_loss": 0.10444886661024909,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11411539051267836,
+   "imperial_eskd": 0.09798052636059848,
+   "liu2021": 0.10125068295747042
+  },
+  "train_loss": 0.10111938607692718,
+  "sec": 2859.4
+ },
+ {
+  "step": 52750,
+  "inner_val_loss": 0.10444841166387503,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11411241897278362,
+   "imperial_eskd": 0.09797675839879295,
+   "liu2021": 0.10125605762004852
+  },
+  "train_loss": 0.10081852629780769,
+  "sec": 2873.2
+ },
+ {
+  "step": 53000,
+  "inner_val_loss": 0.10444980474370798,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11411745308174027,
+   "imperial_eskd": 0.09797522832046855,
+   "liu2021": 0.10125673282891512
+  },
+  "train_loss": 0.10056555798649788,
+  "sec": 2885.7
+ },
+ {
+  "step": 53250,
+  "inner_val_loss": 0.10445170947397597,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11412215729554494,
+   "imperial_eskd": 0.09797493910247629,
+   "liu2021": 0.10125803202390671
+  },
+  "train_loss": 0.10079877761006355,
+  "sec": 2898.2
+ },
+ {
+  "step": 53500,
+  "inner_val_loss": 0.1044521587477489,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11412596330046654,
+   "imperial_eskd": 0.09797460619698871,
+   "liu2021": 0.10125590674579144
+  },
+  "train_loss": 0.10049946105480194,
+  "sec": 2910.6
+ },
+ {
+  "step": 53750,
+  "inner_val_loss": 0.10445056215982244,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11412710572282474,
+   "imperial_eskd": 0.09797230803153732,
+   "liu2021": 0.10125227272510529
+  },
+  "train_loss": 0.10083187559247017,
+  "sec": 2923.0
+ },
+ {
+  "step": 54000,
+  "inner_val_loss": 0.10444604412273124,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11412058936225043,
+   "imperial_eskd": 0.0979718416929245,
+   "liu2021": 0.1012457013130188
+  },
+  "train_loss": 0.10057690969109535,
+  "sec": 2935.4
+ },
+ {
+  "step": 54250,
+  "inner_val_loss": 0.10444388690997253,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11412030085921288,
+   "imperial_eskd": 0.09796929630366238,
+   "liu2021": 0.10124206356704235
+  },
+  "train_loss": 0.10090242931246758,
+  "sec": 2949.1
+ },
+ {
+  "step": 54500,
+  "inner_val_loss": 0.10444020304628193,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11411320832040575,
+   "imperial_eskd": 0.09796945276585492,
+   "liu2021": 0.10123794805258512
+  },
+  "train_loss": 0.10090321350097656,
+  "sec": 2962.9
+ },
+ {
+  "step": 54750,
+  "inner_val_loss": 0.10443976382577559,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11411894195609623,
+   "imperial_eskd": 0.09796730767596852,
+   "liu2021": 0.10123304184526205
+  },
+  "train_loss": 0.10038459491729736,
+  "sec": 2976.7
+ },
+ {
+  "step": 55000,
+  "inner_val_loss": 0.10443687973926565,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11410874293910132,
+   "imperial_eskd": 0.09796701134605841,
+   "liu2021": 0.10123488493263721
+  },
+  "train_loss": 0.10089713472127915,
+  "sec": 2989.1
+ },
+ {
+  "step": 55250,
+  "inner_val_loss": 0.10443456599532756,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11410935016141997,
+   "imperial_eskd": 0.0979637432504784,
+   "liu2021": 0.10123060457408428
+  },
+  "train_loss": 0.10053311419487,
+  "sec": 3003.0
+ },
+ {
+  "step": 55500,
+  "inner_val_loss": 0.10443420883782384,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11410751856035656,
+   "imperial_eskd": 0.09795845537023111,
+   "liu2021": 0.10123665258288383
+  },
+  "train_loss": 0.10083240208029746,
+  "sec": 3016.9
+ },
+ {
+  "step": 55750,
+  "inner_val_loss": 0.10443209780235885,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11410469975736406,
+   "imperial_eskd": 0.09795822304758159,
+   "liu2021": 0.10123337060213089
+  },
+  "train_loss": 0.1007587733566761,
+  "sec": 3029.4
+ },
+ {
+  "step": 56000,
+  "inner_val_loss": 0.10442913533423585,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11410131553808849,
+   "imperial_eskd": 0.0979538234797391,
+   "liu2021": 0.10123226698487997
+  },
+  "train_loss": 0.10049418112635612,
+  "sec": 3043.1
+ },
+ {
+  "step": 56250,
+  "inner_val_loss": 0.10442346822696202,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11409731954336166,
+   "imperial_eskd": 0.09795497798106888,
+   "liu2021": 0.10121810715645552
+  },
+  "train_loss": 0.1008449878692627,
+  "sec": 3056.9
+ },
+ {
+  "step": 56500,
+  "inner_val_loss": 0.10442372349401315,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11409962177276611,
+   "imperial_eskd": 0.09795508533716202,
+   "liu2021": 0.10121646337211132
+  },
+  "train_loss": 0.10023380783200264,
+  "sec": 3070.7
+ },
+ {
+  "step": 56750,
+  "inner_val_loss": 0.10441617564821885,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11409054117070304,
+   "imperial_eskd": 0.09795043515888127,
+   "liu2021": 0.10120755061507225
+  },
+  "train_loss": 0.10035135596990585,
+  "sec": 3083.1
+ },
+ {
+  "step": 57000,
+  "inner_val_loss": 0.10441502786370038,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11409730629788505,
+   "imperial_eskd": 0.09794641624797475,
+   "liu2021": 0.10120136104524136
+  },
+  "train_loss": 0.10044072920083999,
+  "sec": 3096.9
+ },
+ {
+  "step": 57250,
+  "inner_val_loss": 0.10441410258688309,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11409368076258236,
+   "imperial_eskd": 0.09794608131051064,
+   "liu2021": 0.10120254568755627
+  },
+  "train_loss": 0.10078066864609718,
+  "sec": 3110.8
+ },
+ {
+  "step": 57500,
+  "inner_val_loss": 0.10441287826144635,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11408575748403867,
+   "imperial_eskd": 0.09794382446191528,
+   "liu2021": 0.1012090528383851
+  },
+  "train_loss": 0.10070556822419166,
+  "sec": 3123.2
+ },
+ {
+  "step": 57750,
+  "inner_val_loss": 0.10440963957308218,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11407487218578656,
+   "imperial_eskd": 0.09794399717992003,
+   "liu2021": 0.10121004935353994
+  },
+  "train_loss": 0.10034150204062461,
+  "sec": 3137.0
+ },
+ {
+  "step": 58000,
+  "inner_val_loss": 0.10441087686491234,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1140727765030331,
+   "imperial_eskd": 0.09794463962316513,
+   "liu2021": 0.10121521446853876
+  },
+  "train_loss": 0.10072291618585587,
+  "sec": 3151.4
+ },
+ {
+  "step": 58250,
+  "inner_val_loss": 0.10440907391848087,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1140707785056697,
+   "imperial_eskd": 0.09794060479510915,
+   "liu2021": 0.10121583845466375
+  },
+  "train_loss": 0.10097867840528488,
+  "sec": 3164.1
+ },
+ {
+  "step": 58500,
+  "inner_val_loss": 0.10440977856905674,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1140688889556461,
+   "imperial_eskd": 0.09793567691336978,
+   "liu2021": 0.10122476983815432
+  },
+  "train_loss": 0.10085507380962372,
+  "sec": 3176.6
+ },
+ {
+  "step": 58750,
+  "inner_val_loss": 0.10441107076501806,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1140717723303371,
+   "imperial_eskd": 0.09793027659708803,
+   "liu2021": 0.10123116336762905
+  },
+  "train_loss": 0.10080719870328904,
+  "sec": 3189.0
+ },
+ {
+  "step": 59000,
+  "inner_val_loss": 0.10440879937587462,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11407179550992118,
+   "imperial_eskd": 0.0979289182207801,
+   "liu2021": 0.10122568439692259
+  },
+  "train_loss": 0.10090348866581916,
+  "sec": 3201.5
+ },
+ {
+  "step": 59250,
+  "inner_val_loss": 0.10440537953743027,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11406737648778492,
+   "imperial_eskd": 0.09792859953912822,
+   "liu2021": 0.1012201625853777
+  },
+  "train_loss": 0.1003434229195118,
+  "sec": 3214.0
+ },
+ {
+  "step": 59500,
+  "inner_val_loss": 0.10440440724293391,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11406854912638664,
+   "imperial_eskd": 0.09792927652597427,
+   "liu2021": 0.10121539607644081
+  },
+  "train_loss": 0.10067384693026543,
+  "sec": 3227.8
+ },
+ {
+  "step": 59750,
+  "inner_val_loss": 0.10440630422472352,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11407108729084332,
+   "imperial_eskd": 0.09792719713666222,
+   "liu2021": 0.101220628246665
+  },
+  "train_loss": 0.10069837829470635,
+  "sec": 3240.2
+ },
+ {
+  "step": 60000,
+  "inner_val_loss": 0.10440332228662791,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11406028932995266,
+   "imperial_eskd": 0.09792539104819298,
+   "liu2021": 0.10122428648173809
+  },
+  "train_loss": 0.10047169044613838,
+  "sec": 3252.8
+ },
+ {
+  "step": 60250,
+  "inner_val_loss": 0.10439833198746146,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11404843048916923,
+   "imperial_eskd": 0.09792292998595671,
+   "liu2021": 0.10122363548725843
+  },
+  "train_loss": 0.10072414001822472,
+  "sec": 3268.1
+ },
+ {
+  "step": 60500,
+  "inner_val_loss": 0.10439500372593451,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11404574414094289,
+   "imperial_eskd": 0.09792116826230829,
+   "liu2021": 0.10121809877455235
+  },
+  "train_loss": 0.10052234384417534,
+  "sec": 3282.3
+ },
+ {
+  "step": 60750,
+  "inner_val_loss": 0.10439151572220355,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11403664739595519,
+   "imperial_eskd": 0.09791959889910438,
+   "liu2021": 0.10121830087155104
+  },
+  "train_loss": 0.10050015240907668,
+  "sec": 3296.7
+ },
+ {
+  "step": 61000,
+  "inner_val_loss": 0.10439404207422877,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11403756216168404,
+   "imperial_eskd": 0.09791586683555083,
+   "liu2021": 0.10122869722545147
+  },
+  "train_loss": 0.10076226082444191,
+  "sec": 3310.9
+ },
+ {
+  "step": 61250,
+  "inner_val_loss": 0.10439512867994782,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11404417744941181,
+   "imperial_eskd": 0.09791638566689058,
+   "liu2021": 0.10122482292354107
+  },
+  "train_loss": 0.10091929018497467,
+  "sec": 3323.6
+ },
+ {
+  "step": 61500,
+  "inner_val_loss": 0.10439308984550061,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11404144308633274,
+   "imperial_eskd": 0.09791583567857742,
+   "liu2021": 0.10122199077159166
+  },
+  "train_loss": 0.10045726424455642,
+  "sec": 3336.4
+ },
+ {
+  "step": 61750,
+  "inner_val_loss": 0.10438836677633351,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11404245553745164,
+   "imperial_eskd": 0.09791293537074869,
+   "liu2021": 0.10120970942080021
+  },
+  "train_loss": 0.10069338873028755,
+  "sec": 3349.2
+ },
+ {
+  "step": 62000,
+  "inner_val_loss": 0.10438706585353473,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11404295555419391,
+   "imperial_eskd": 0.0979123863984238,
+   "liu2021": 0.10120585560798645
+  },
+  "train_loss": 0.10067083984613419,
+  "sec": 3363.4
+ },
+ {
+  "step": 62250,
+  "inner_val_loss": 0.10438246664399851,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11403659275836414,
+   "imperial_eskd": 0.09790844030000946,
+   "liu2021": 0.10120236687362194
+  },
+  "train_loss": 0.10039371088147163,
+  "sec": 3377.3
+ },
+ {
+  "step": 62500,
+  "inner_val_loss": 0.10437867511690968,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11403169027633137,
+   "imperial_eskd": 0.09790618616071614,
+   "liu2021": 0.10119814891368151
+  },
+  "train_loss": 0.10078647664189339,
+  "sec": 3391.0
+ },
+ {
+  "step": 62750,
+  "inner_val_loss": 0.1043718540068929,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11402543799744712,
+   "imperial_eskd": 0.09790316122499379,
+   "liu2021": 0.1011869627982378
+  },
+  "train_loss": 0.10050883036851883,
+  "sec": 3405.1
+ },
+ {
+  "step": 63000,
+  "inner_val_loss": 0.10436773711897007,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11402521323826578,
+   "imperial_eskd": 0.09789769520813768,
+   "liu2021": 0.10118030291050673
+  },
+  "train_loss": 0.10083009538054466,
+  "sec": 3419.2
+ },
+ {
+  "step": 63250,
+  "inner_val_loss": 0.1043706092237172,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11402904035316573,
+   "imperial_eskd": 0.0978968275541609,
+   "liu2021": 0.10118595976382494
+  },
+  "train_loss": 0.10083537316322326,
+  "sec": 3433.3
+ },
+ {
+  "step": 63500,
+  "inner_val_loss": 0.10437095254622024,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11402958383162816,
+   "imperial_eskd": 0.0978957438333468,
+   "liu2021": 0.10118752997368574
+  },
+  "train_loss": 0.10045717823505401,
+  "sec": 3445.8
+ },
+ {
+  "step": 63750,
+  "inner_val_loss": 0.10437227605594328,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.114032584346003,
+   "imperial_eskd": 0.09789589894088832,
+   "liu2021": 0.10118834488093853
+  },
+  "train_loss": 0.1006707364320755,
+  "sec": 3458.2
+ },
+ {
+  "step": 64000,
+  "inner_val_loss": 0.1043711246025783,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11404089546865886,
+   "imperial_eskd": 0.09788893535733223,
+   "liu2021": 0.10118354298174381
+  },
+  "train_loss": 0.10058473166823387,
+  "sec": 3470.7
+ },
+ {
+  "step": 64250,
+  "inner_val_loss": 0.1043735160539397,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.1140520249803861,
+   "imperial_eskd": 0.09788821468299086,
+   "liu2021": 0.10118030849844217
+  },
+  "train_loss": 0.10047653412818909,
+  "sec": 3483.1
+ },
+ {
+  "step": 64500,
+  "inner_val_loss": 0.10437050255749263,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11404547095298767,
+   "imperial_eskd": 0.0978840481151234,
+   "liu2021": 0.10118198860436678
+  },
+  "train_loss": 0.10047277337312699,
+  "sec": 3495.6
+ },
+ {
+  "step": 64750,
+  "inner_val_loss": 0.104369798292616,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11403557492627038,
+   "imperial_eskd": 0.09788297353820367,
+   "liu2021": 0.10119084641337395
+  },
+  "train_loss": 0.10067699885368347,
+  "sec": 3508.1
+ },
+ {
+  "step": 65000,
+  "inner_val_loss": 0.10437031920296876,
+  "inner_val_per_cohort": {
+   "ye2026_sepsis": 0.11403637338015768,
+   "imperial_eskd": 0.09788078645413573,
+   "liu2021": 0.1011937977746129
+  },
+  "train_loss": 0.10046402275562287,
+  "sec": 3520.6
+ }
+]