{ "architecture": { "base_width": 16, "depth": 20, "family": "CIFAR 6n+2 ResNet, option-A shortcuts", "forward_parameter_count": 269722, "normalization": "batchnorm", "residual_scale": 1.0 }, "args": { "augment_train": 1, "batch_size": 128, "data_dir": "/orion/u/oscarwan/scratch/sdil-data", "depth": 20, "device": "cuda", "dp_alpha": 0.0, "dp_beta": 0.1, "dp_inference_passes": 16, "ep_beta": 0.5, "ep_dt": 0.5, "ep_free_steps": 20, "ep_nudge_steps": 4, "epochs": 100, "eval_every": 1, "eval_split": "validation", "feedback_seed": 1729, "ff_score_from_layer": 1, "ff_threshold": 2.0, "loader_seed": 0, "lr": 0.0003, "lr_gamma": 0.1, "lr_milestones": "100,150", "lr_schedule": "constant", "max_steps": 0, "method": "ep", "momentum": 0.9, "out": "/orion/u/oscarwan/scratch/sdil-work/sdil-a6000-full-rerun/results/resnet_crossover/r2/resnet-r2-ep-d20-lr0p0003.json", "output_lr": 0.0003, "pepita_projection_scale": 0.05, "seed": 0, "split_seed": 2027, "train_limit": 0, "val_examples": 5000, "weight_decay": 0.0001, "width": 16 }, "epochs": [ { "epoch": 1, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.10000092970662647, "train_seconds": 139.11669993400574, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09817654571533203 }, "validation_seconds": 11.159880876541138 }, { "epoch": 2, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.096760009001361, "train_seconds": 138.82202982902527, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09676226196289063 }, "validation_seconds": 11.14545726776123 }, { "epoch": 3, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09596625119845073, "train_seconds": 138.83978295326233, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0960817398071289 }, "validation_seconds": 11.141392230987549 }, { "epoch": 4, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09608556209140354, "train_seconds": 138.88129901885986, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09456661682128906 }, "validation_seconds": 11.150422811508179 }, { "epoch": 5, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.096144191037284, "train_seconds": 138.87270593643188, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09416954193115235 }, "validation_seconds": 11.146359920501709 }, { "epoch": 6, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0957507728377978, "train_seconds": 138.86185312271118, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09589024200439453 }, "validation_seconds": 11.14265751838684 }, { "epoch": 7, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09609245043992996, "train_seconds": 138.87552165985107, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0969051010131836 }, "validation_seconds": 11.147951364517212 }, { "epoch": 8, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09637515972455342, "train_seconds": 139.2255883216858, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0969967056274414 }, "validation_seconds": 11.152616500854492 }, { "epoch": 9, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09704533698028989, "train_seconds": 139.07272458076477, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09479008636474609 }, "validation_seconds": 11.14713978767395 }, { "epoch": 10, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0958041417002678, "train_seconds": 138.8842315673828, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09533480987548829 }, "validation_seconds": 11.152060270309448 }, { "epoch": 11, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09585828256342147, "train_seconds": 138.86268758773804, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.096821923828125 }, "validation_seconds": 11.150679111480713 }, { "epoch": 12, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.095636143067148, "train_seconds": 138.91375017166138, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09711122283935547 }, "validation_seconds": 11.14675760269165 }, { "epoch": 13, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09573306566079458, "train_seconds": 138.89357042312622, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09591180725097656 }, "validation_seconds": 11.144380331039429 }, { "epoch": 14, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09582586610449685, "train_seconds": 138.9151952266693, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09458838500976563 }, "validation_seconds": 11.150224208831787 }, { "epoch": 15, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.095475492699941, "train_seconds": 138.97902250289917, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0952644287109375 }, "validation_seconds": 11.146376609802246 }, { "epoch": 16, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09557586745686002, "train_seconds": 138.93276834487915, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09532229919433594 }, "validation_seconds": 11.151374340057373 }, { "epoch": 17, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09595397739410401, "train_seconds": 138.96198630332947, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09608304290771484 }, "validation_seconds": 11.14755392074585 }, { "epoch": 18, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09530874093108707, "train_seconds": 138.92328310012817, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09483186340332031 }, "validation_seconds": 11.149696826934814 }, { "epoch": 19, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09545710382726458, "train_seconds": 138.9205515384674, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09437860565185546 }, "validation_seconds": 11.14361047744751 }, { "epoch": 20, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09598765107393264, "train_seconds": 138.96228528022766, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09477271881103516 }, "validation_seconds": 11.148051023483276 }, { "epoch": 21, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09561405028104782, "train_seconds": 138.91624879837036, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09785403137207031 }, "validation_seconds": 11.150966167449951 }, { "epoch": 22, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09561793030500412, "train_seconds": 138.9595160484314, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09515677032470703 }, "validation_seconds": 11.145541429519653 }, { "epoch": 23, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09509115369982189, "train_seconds": 138.9069902896881, "validation": { "accuracy": 0.0922, "examples": 5000, "loss": 0.09626204528808593 }, "validation_seconds": 11.147856712341309 }, { "epoch": 24, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09560073957708147, "train_seconds": 138.90601563453674, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09640469055175781 }, "validation_seconds": 11.146676540374756 }, { "epoch": 25, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09574450379742516, "train_seconds": 138.99443531036377, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09708086853027344 }, "validation_seconds": 11.152676582336426 }, { "epoch": 26, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09567417042917675, "train_seconds": 139.10978770256042, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09664617004394531 }, "validation_seconds": 11.15486764907837 }, { "epoch": 27, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09516358087592655, "train_seconds": 139.1539888381958, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09485475616455079 }, "validation_seconds": 11.144922256469727 }, { "epoch": 28, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09542587406900194, "train_seconds": 138.9068100452423, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.094023095703125 }, "validation_seconds": 11.144463777542114 }, { "epoch": 29, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09558091937568453, "train_seconds": 138.85909581184387, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0952291488647461 }, "validation_seconds": 11.140281677246094 }, { "epoch": 30, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09604495452377532, "train_seconds": 138.83931016921997, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09689130554199218 }, "validation_seconds": 11.138436555862427 }, { "epoch": 31, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09616093474494086, "train_seconds": 138.60884046554565, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09545914764404297 }, "validation_seconds": 11.133517980575562 }, { "epoch": 32, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0956987985306316, "train_seconds": 138.62059688568115, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09510513305664063 }, "validation_seconds": 11.137269735336304 }, { "epoch": 33, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09554892805814744, "train_seconds": 138.61837697029114, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0953763916015625 }, "validation_seconds": 11.135218620300293 }, { "epoch": 34, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09605118489927715, "train_seconds": 138.62338376045227, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09536750946044922 }, "validation_seconds": 11.138956546783447 }, { "epoch": 35, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09561948593987359, "train_seconds": 138.6473937034607, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09514895172119141 }, "validation_seconds": 11.136772155761719 }, { "epoch": 36, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09602282423973084, "train_seconds": 138.74485182762146, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09477901153564453 }, "validation_seconds": 11.140544414520264 }, { "epoch": 37, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09655213383701113, "train_seconds": 138.73414587974548, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09620673370361328 }, "validation_seconds": 11.137867212295532 }, { "epoch": 38, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09576946388880411, "train_seconds": 138.76311469078064, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09446058502197266 }, "validation_seconds": 11.140277862548828 }, { "epoch": 39, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.095974667693509, "train_seconds": 138.73441815376282, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09710462036132812 }, "validation_seconds": 11.13795518875122 }, { "epoch": 40, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09609547713862525, "train_seconds": 138.68924736976624, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09623699340820313 }, "validation_seconds": 11.141454458236694 }, { "epoch": 41, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09539313185214997, "train_seconds": 138.66758728027344, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09325079956054687 }, "validation_seconds": 11.137190341949463 }, { "epoch": 42, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0956763015402688, "train_seconds": 138.69396257400513, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0959182632446289 }, "validation_seconds": 11.139819622039795 }, { "epoch": 43, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09580025514364243, "train_seconds": 138.75355863571167, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09510519409179688 }, "validation_seconds": 11.140304565429688 }, { "epoch": 44, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09580142854981952, "train_seconds": 138.7804079055786, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09460178375244141 }, "validation_seconds": 11.142134666442871 }, { "epoch": 45, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0952616327073839, "train_seconds": 138.74207544326782, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09451842498779296 }, "validation_seconds": 11.138129711151123 }, { "epoch": 46, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09581543866263495, "train_seconds": 138.78086709976196, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09661761474609375 }, "validation_seconds": 11.14474630355835 }, { "epoch": 47, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09611166249513627, "train_seconds": 138.81940126419067, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09566448211669921 }, "validation_seconds": 11.141072750091553 }, { "epoch": 48, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09542014650636249, "train_seconds": 138.77664518356323, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09565098419189454 }, "validation_seconds": 11.144866228103638 }, { "epoch": 49, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09609724575148688, "train_seconds": 138.78950238227844, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09689692687988281 }, "validation_seconds": 11.142502546310425 }, { "epoch": 50, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09599001576503118, "train_seconds": 138.87918281555176, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09705646209716796 }, "validation_seconds": 11.141917705535889 }, { "epoch": 51, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09611067996554905, "train_seconds": 138.81238508224487, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09507562713623047 }, "validation_seconds": 11.139949560165405 }, { "epoch": 52, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09620131080547968, "train_seconds": 138.90105414390564, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09800862884521484 }, "validation_seconds": 11.146653175354004 }, { "epoch": 53, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09654972725576824, "train_seconds": 138.98731064796448, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09603324127197266 }, "validation_seconds": 11.144920110702515 }, { "epoch": 54, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09627177377939224, "train_seconds": 138.9740378856659, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0954896240234375 }, "validation_seconds": 11.148634433746338 }, { "epoch": 55, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09626585129234526, "train_seconds": 138.9596757888794, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09900887756347657 }, "validation_seconds": 11.143959760665894 }, { "epoch": 56, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09648386177884208, "train_seconds": 138.9568064212799, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09393155670166016 }, "validation_seconds": 11.14871096611023 }, { "epoch": 57, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09645076555146112, "train_seconds": 138.95467948913574, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0980503646850586 }, "validation_seconds": 11.14472484588623 }, { "epoch": 58, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09627769343058268, "train_seconds": 139.03842067718506, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09474974517822266 }, "validation_seconds": 11.150963068008423 }, { "epoch": 59, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09605537111361821, "train_seconds": 139.1419117450714, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09802635192871094 }, "validation_seconds": 11.146832466125488 }, { "epoch": 60, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09588881915277905, "train_seconds": 139.09739804267883, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0964722396850586 }, "validation_seconds": 11.148901224136353 }, { "epoch": 61, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09624453692171309, "train_seconds": 139.0157241821289, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09603090057373047 }, "validation_seconds": 11.145693302154541 }, { "epoch": 62, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09638927124738693, "train_seconds": 139.01228952407837, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09630657653808594 }, "validation_seconds": 11.15030837059021 }, { "epoch": 63, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09640581868754493, "train_seconds": 139.06430864334106, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0956555206298828 }, "validation_seconds": 11.145128965377808 }, { "epoch": 64, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09614854072729746, "train_seconds": 139.0981409549713, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09355887908935547 }, "validation_seconds": 11.149465322494507 }, { "epoch": 65, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09623182185226016, "train_seconds": 139.18455290794373, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09463775634765625 }, "validation_seconds": 11.150774478912354 }, { "epoch": 66, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09593160262770123, "train_seconds": 139.1528673171997, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0963241683959961 }, "validation_seconds": 11.157290697097778 }, { "epoch": 67, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09562942595614328, "train_seconds": 139.32904410362244, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09660213775634766 }, "validation_seconds": 11.155967235565186 }, { "epoch": 68, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09585837714936998, "train_seconds": 139.3654706478119, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09532363433837891 }, "validation_seconds": 11.157169580459595 }, { "epoch": 69, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09573291099998686, "train_seconds": 139.34520745277405, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09697261047363281 }, "validation_seconds": 11.158665418624878 }, { "epoch": 70, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09632284501658546, "train_seconds": 139.34444737434387, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09541180267333985 }, "validation_seconds": 11.157097578048706 }, { "epoch": 71, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09558174745374255, "train_seconds": 139.3179748058319, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09908675689697266 }, "validation_seconds": 11.1538667678833 }, { "epoch": 72, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09581099202367994, "train_seconds": 139.2078824043274, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0985763900756836 }, "validation_seconds": 11.154834270477295 }, { "epoch": 73, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09601372849543889, "train_seconds": 139.2479133605957, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09627315826416015 }, "validation_seconds": 11.15363883972168 }, { "epoch": 74, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09621802689234416, "train_seconds": 139.25913310050964, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0955810287475586 }, "validation_seconds": 11.156188011169434 }, { "epoch": 75, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09564000614616606, "train_seconds": 139.2822630405426, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09619723358154297 }, "validation_seconds": 11.155544519424438 }, { "epoch": 76, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0960162280175421, "train_seconds": 139.34670281410217, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09646122741699219 }, "validation_seconds": 11.155498504638672 }, { "epoch": 77, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09561879307958815, "train_seconds": 139.26200103759766, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09832343292236329 }, "validation_seconds": 11.154755353927612 }, { "epoch": 78, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09588940723737081, "train_seconds": 139.21251392364502, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09473636779785156 }, "validation_seconds": 11.15525221824646 }, { "epoch": 79, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09603877027299669, "train_seconds": 139.1633071899414, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09692667694091797 }, "validation_seconds": 11.1514732837677 }, { "epoch": 80, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09629039370351368, "train_seconds": 139.21222305297852, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0954868438720703 }, "validation_seconds": 11.154970169067383 }, { "epoch": 81, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09558552947839101, "train_seconds": 139.18993878364563, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09273733062744141 }, "validation_seconds": 11.155423641204834 }, { "epoch": 82, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09602079207499822, "train_seconds": 139.26600408554077, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09672743530273438 }, "validation_seconds": 11.157419919967651 }, { "epoch": 83, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09581270765331057, "train_seconds": 139.23123788833618, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09539513549804687 }, "validation_seconds": 11.153031349182129 }, { "epoch": 84, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09577875078784095, "train_seconds": 139.1194291114807, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09731991729736328 }, "validation_seconds": 11.154894828796387 }, { "epoch": 85, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09578391640848584, "train_seconds": 139.0757772922516, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09821551055908204 }, "validation_seconds": 11.155132532119751 }, { "epoch": 86, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09557791891230477, "train_seconds": 139.0469081401825, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09509595336914063 }, "validation_seconds": 11.150206327438354 }, { "epoch": 87, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09617045336696836, "train_seconds": 138.9863407611847, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0955680892944336 }, "validation_seconds": 11.152897119522095 }, { "epoch": 88, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09557901961538527, "train_seconds": 139.05934143066406, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09560486907958984 }, "validation_seconds": 11.151857137680054 }, { "epoch": 89, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09592116408877903, "train_seconds": 139.1188690662384, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0936294677734375 }, "validation_seconds": 11.1571364402771 }, { "epoch": 90, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09588112912840313, "train_seconds": 139.1143081188202, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09687020568847657 }, "validation_seconds": 11.151924133300781 }, { "epoch": 91, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09571159390211105, "train_seconds": 139.09063410758972, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09342134704589844 }, "validation_seconds": 11.154841661453247 }, { "epoch": 92, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09590040313800176, "train_seconds": 139.09315967559814, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0954732666015625 }, "validation_seconds": 11.153255462646484 }, { "epoch": 93, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09588631826241811, "train_seconds": 139.12650752067566, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09767638397216796 }, "validation_seconds": 11.1552095413208 }, { "epoch": 94, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0961217908249961, "train_seconds": 139.14415001869202, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09442243499755859 }, "validation_seconds": 11.151734828948975 }, { "epoch": 95, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09611847210062875, "train_seconds": 139.14410185813904, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09692444000244141 }, "validation_seconds": 11.152769804000854 }, { "epoch": 96, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0958005420896742, "train_seconds": 139.11920261383057, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09459645385742188 }, "validation_seconds": 11.155317544937134 }, { "epoch": 97, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09605646768146091, "train_seconds": 139.06602239608765, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09579430389404296 }, "validation_seconds": 11.153225183486938 }, { "epoch": 98, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09623256181875865, "train_seconds": 139.11849069595337, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09721367797851563 }, "validation_seconds": 11.151837587356567 }, { "epoch": 99, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09644517441458172, "train_seconds": 139.11283159255981, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09481856689453125 }, "validation_seconds": 11.155541896820068 }, { "epoch": 100, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0958042881541782, "train_seconds": 139.26254963874817, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09437472686767578 }, "validation_seconds": 11.155186176300049 } ], "evaluation_protocol": { "split": "validation", "test_evaluations": 0, "test_used_for_selection": false }, "final": { "accuracy": 0.1, "evaluation_seconds": 11.155261516571045, "examples": 5000, "finite": true, "loss": 0.09437472686767578 }, "first_nonfinite_step": null, "hardware": { "cuda_device_name": "NVIDIA RTX A6000", "cuda_visible_devices": "0", "device": "cuda", "device_total_memory_bytes": 50896568320, "peak_memory_allocated_bytes": 4845611008, "peak_memory_reserved_bytes": 5939134464, "torch_version": "2.3.1+cu118" }, "layers": [], "protocol_family": "resnet_local_learning_crossover", "provenance": { "git_commit": "f55872beaee658ff76ac15fd421b73017820c320", "git_tracked_dirty": false }, "schema_version": 1, "split": { "cifar_source_files": [ { "bytes": 31035704, "path": "/orion/u/oscarwan/scratch/sdil-data/cifar-10-batches-py/data_batch_1", "sha256": "54636561a3ce25bd3e19253c6b0d8538147b0ae398331ac4a2d86c6d987368cd" }, { "bytes": 31035320, "path": "/orion/u/oscarwan/scratch/sdil-data/cifar-10-batches-py/data_batch_2", "sha256": "766b2cef9fbc745cf056b3152224f7cf77163b330ea9a15f9392beb8b89bc5a8" }, { "bytes": 31035999, "path": "/orion/u/oscarwan/scratch/sdil-data/cifar-10-batches-py/data_batch_3", "sha256": "0f00d98ebfb30b3ec0ad19f9756dc2630b89003e10525f5e148445e82aa6a1f9" }, { "bytes": 31035696, "path": "/orion/u/oscarwan/scratch/sdil-data/cifar-10-batches-py/data_batch_4", "sha256": "3f7bb240661948b8f4d53e36ec720d8306f5668bd0071dcb4e6c947f78e9682b" }, { "bytes": 31035623, "path": "/orion/u/oscarwan/scratch/sdil-data/cifar-10-batches-py/data_batch_5", "sha256": "d91802434d8376bbaeeadf58a737e3a1b12ac839077e931237e0dcd43adcb154" }, { "bytes": 31035526, "path": "/orion/u/oscarwan/scratch/sdil-data/cifar-10-batches-py/test_batch", "sha256": "f53d8d457504f7cff4ea9e021afcf0e0ad8e24a91f3fc42091b8adef61157831" } ], "dataset": "cifar10", "input_layout": "NCHW", "input_shape": [ 3, 32, 32 ], "loader_seed": 0, "normalization_mean": [ 0.49140000343322754, 0.4821999967098236, 0.4465000033378601 ], "normalization_std": [ 0.24699999392032623, 0.2434999942779541, 0.26159998774528503 ], "split_from_training_only": true, "split_seed": 2027, "test_examples": 10000, "train_examples": 45000, "training_augmentation": "random_crop_32_padding_4_zero_then_horizontal_flip_p0.5", "validation_class_counts": { "0": 500, "1": 500, "2": 500, "3": 500, "4": 500, "5": 500, "6": 500, "7": 500, "8": 500, "9": 500 }, "validation_examples": 5000, "validation_index_sha256": "8328b206a97c420e49e54e3eca4abe3274c4756b084355784ea3fb8059e4515b" }, "total_wall_seconds": 15025.534816026688, "work": { "candidate_label_evaluation_presentations": 0, "completed_global_epochs": 100, "feedforward_example_passes": 5005000, "forward_macs_per_example": 40551040, "forward_parameter_count": 269722, "local_target_backward_example_evaluations": 0, "local_vjp_example_evaluations": 118100000, "logical_task_loss_queries": 0, "num_trainable_layers": 20, "ordinary_training_examples": 4500000, "ordinary_validation_examples": 505000, "relaxation_example_passes": 118100000, "training_example_presentations": 4500000 } }