{ "architecture": { "base_width": 16, "depth": 32, "family": "CIFAR 6n+2 ResNet, option-A shortcuts", "forward_parameter_count": 464154, "normalization": "batchnorm", "residual_scale": 1.0 }, "args": { "augment_train": 1, "batch_size": 128, "data_dir": "/orion/u/oscarwan/scratch/sdil-data", "depth": 32, "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-d32-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": 0.10584368345869911, "train_seconds": 233.75673270225525, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09748240203857422 }, "validation_seconds": 18.687715530395508 }, { "epoch": 2, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09607491334146924, "train_seconds": 227.22966647148132, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09624678039550781 }, "validation_seconds": 18.22874617576599 }, { "epoch": 3, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09619754118786918, "train_seconds": 227.6295883655548, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09637385711669921 }, "validation_seconds": 18.230604887008667 }, { "epoch": 4, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.096190148557557, "train_seconds": 227.70378065109253, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09466078643798828 }, "validation_seconds": 18.22860026359558 }, { "epoch": 5, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09619633632633422, "train_seconds": 227.67504620552063, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0939787109375 }, "validation_seconds": 18.228357076644897 }, { "epoch": 6, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09575496319664849, "train_seconds": 227.9477686882019, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0957251968383789 }, "validation_seconds": 18.234267473220825 }, { "epoch": 7, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0960714242219925, "train_seconds": 227.73594498634338, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09659960479736328 }, "validation_seconds": 18.239380836486816 }, { "epoch": 8, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09630487336715063, "train_seconds": 228.03009343147278, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0964680160522461 }, "validation_seconds": 18.236380100250244 }, { "epoch": 9, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0967883091436492, "train_seconds": 228.04988503456116, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0942939926147461 }, "validation_seconds": 18.238703966140747 }, { "epoch": 10, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09595642296738095, "train_seconds": 227.94507455825806, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09557775573730469 }, "validation_seconds": 18.23998737335205 }, { "epoch": 11, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09607761137882868, "train_seconds": 227.97127604484558, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09727976989746094 }, "validation_seconds": 18.244027853012085 }, { "epoch": 12, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09592779109875361, "train_seconds": 228.01011180877686, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09815475616455079 }, "validation_seconds": 18.242377281188965 }, { "epoch": 13, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09610559370915095, "train_seconds": 228.01637840270996, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09631928253173828 }, "validation_seconds": 18.243257522583008 }, { "epoch": 14, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0962134348101086, "train_seconds": 227.9632499217987, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09476177368164063 }, "validation_seconds": 18.235589265823364 }, { "epoch": 15, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0959621706340048, "train_seconds": 228.08780241012573, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09589082336425782 }, "validation_seconds": 18.23856282234192 }, { "epoch": 16, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09578963030576707, "train_seconds": 228.17340397834778, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09558954162597656 }, "validation_seconds": 18.241756200790405 }, { "epoch": 17, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09626044083568785, "train_seconds": 228.17240142822266, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09684005889892579 }, "validation_seconds": 18.241687536239624 }, { "epoch": 18, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09598865276575089, "train_seconds": 228.0541534423828, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0955846206665039 }, "validation_seconds": 18.231510877609253 }, { "epoch": 19, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09609488620228238, "train_seconds": 227.87761616706848, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09553548736572266 }, "validation_seconds": 18.23040270805359 }, { "epoch": 20, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09648366617361705, "train_seconds": 227.81606435775757, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09536498565673829 }, "validation_seconds": 18.23584270477295 }, { "epoch": 21, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09604000683757993, "train_seconds": 227.8779010772705, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09821461181640626 }, "validation_seconds": 18.240376234054565 }, { "epoch": 22, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09619036858214272, "train_seconds": 228.08918619155884, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09566349029541016 }, "validation_seconds": 18.23198366165161 }, { "epoch": 23, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09555974514219497, "train_seconds": 227.9332685470581, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09647967376708984 }, "validation_seconds": 18.237937688827515 }, { "epoch": 24, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09629934613572226, "train_seconds": 227.74945855140686, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09690995330810546 }, "validation_seconds": 18.240371465682983 }, { "epoch": 25, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09639721555842294, "train_seconds": 227.9227864742279, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09780321807861328 }, "validation_seconds": 18.22849941253662 }, { "epoch": 26, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09636331896252102, "train_seconds": 227.874342918396, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09749327545166016 }, "validation_seconds": 18.230159044265747 }, { "epoch": 27, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0959015838517083, "train_seconds": 227.95591282844543, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09564348297119141 }, "validation_seconds": 18.233003854751587 }, { "epoch": 28, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09609772173298729, "train_seconds": 227.70959520339966, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0948007080078125 }, "validation_seconds": 18.234058380126953 }, { "epoch": 29, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09593701247109307, "train_seconds": 227.7418727874756, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09526193695068359 }, "validation_seconds": 18.231982469558716 }, { "epoch": 30, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09594838575919469, "train_seconds": 227.90904927253723, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09659895324707031 }, "validation_seconds": 18.23198628425598 }, { "epoch": 31, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09625638297796249, "train_seconds": 228.0143768787384, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09601235046386719 }, "validation_seconds": 18.23534321784973 }, { "epoch": 32, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09599113212161595, "train_seconds": 227.99561190605164, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0954147232055664 }, "validation_seconds": 18.23655915260315 }, { "epoch": 33, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09575922750896877, "train_seconds": 227.84987020492554, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09525252075195312 }, "validation_seconds": 18.23592734336853 }, { "epoch": 34, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09635789152251349, "train_seconds": 227.81707239151, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09551718444824218 }, "validation_seconds": 18.234346628189087 }, { "epoch": 35, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09587441050873863, "train_seconds": 227.86760759353638, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09489838562011718 }, "validation_seconds": 18.23371911048889 }, { "epoch": 36, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09623189973698722, "train_seconds": 227.76344990730286, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09447245178222656 }, "validation_seconds": 18.233267545700073 }, { "epoch": 37, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09652466434902615, "train_seconds": 228.0108609199524, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09592162322998046 }, "validation_seconds": 18.23678684234619 }, { "epoch": 38, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09560628210968442, "train_seconds": 228.01479268074036, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09405255126953126 }, "validation_seconds": 18.23764657974243 }, { "epoch": 39, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0959057019048267, "train_seconds": 228.06421947479248, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09734141845703125 }, "validation_seconds": 18.23655915260315 }, { "epoch": 40, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09606925669776069, "train_seconds": 228.01972579956055, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09611746368408203 }, "validation_seconds": 18.233552932739258 }, { "epoch": 41, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09539408498605093, "train_seconds": 227.97406220436096, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09368319549560547 }, "validation_seconds": 18.23234748840332 }, { "epoch": 42, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09593300796614752, "train_seconds": 227.96903944015503, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09569558410644531 }, "validation_seconds": 18.239282608032227 }, { "epoch": 43, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09585642129712635, "train_seconds": 227.84337186813354, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09548786163330078 }, "validation_seconds": 18.231594562530518 }, { "epoch": 44, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09586924339665306, "train_seconds": 227.81595945358276, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0945896484375 }, "validation_seconds": 18.23236846923828 }, { "epoch": 45, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09525853757328458, "train_seconds": 227.8766598701477, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09463276062011719 }, "validation_seconds": 18.23231053352356 }, { "epoch": 46, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09574040567874909, "train_seconds": 228.08118081092834, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09664346923828125 }, "validation_seconds": 18.23355007171631 }, { "epoch": 47, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09600410343408584, "train_seconds": 228.07431888580322, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09555250701904297 }, "validation_seconds": 18.234946966171265 }, { "epoch": 48, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09547720072666804, "train_seconds": 228.05487155914307, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0959607162475586 }, "validation_seconds": 18.235946893692017 }, { "epoch": 49, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09616076918972863, "train_seconds": 228.20362186431885, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09696863861083985 }, "validation_seconds": 18.243194818496704 }, { "epoch": 50, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09633234109481176, "train_seconds": 228.4779417514801, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09757501220703126 }, "validation_seconds": 18.253101587295532 }, { "epoch": 51, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09652728734413783, "train_seconds": 228.59612488746643, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09503995056152344 }, "validation_seconds": 18.25339436531067 }, { "epoch": 52, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09643947992589738, "train_seconds": 228.75814485549927, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09830532989501953 }, "validation_seconds": 18.267754793167114 }, { "epoch": 53, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09715808055930668, "train_seconds": 228.82663297653198, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09692757568359375 }, "validation_seconds": 18.26055932044983 }, { "epoch": 54, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09683998356130388, "train_seconds": 228.8337893486023, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0960301010131836 }, "validation_seconds": 18.252885818481445 }, { "epoch": 55, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09688982499837875, "train_seconds": 228.71393632888794, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09920754699707031 }, "validation_seconds": 18.25892972946167 }, { "epoch": 56, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09654736138449775, "train_seconds": 228.71237993240356, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0933240005493164 }, "validation_seconds": 18.253551483154297 }, { "epoch": 57, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09635696367555194, "train_seconds": 228.71023201942444, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09789161682128907 }, "validation_seconds": 18.252516269683838 }, { "epoch": 58, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09628176087935765, "train_seconds": 228.71620917320251, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09455877990722657 }, "validation_seconds": 18.252784967422485 }, { "epoch": 59, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09615534450610479, "train_seconds": 228.703430891037, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09805198974609375 }, "validation_seconds": 18.256057262420654 }, { "epoch": 60, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09615144361654918, "train_seconds": 228.6930000782013, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09711163940429687 }, "validation_seconds": 18.25197148323059 }, { "epoch": 61, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09662816899882423, "train_seconds": 228.6902298927307, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09641593322753907 }, "validation_seconds": 18.250696659088135 }, { "epoch": 62, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09674948098129696, "train_seconds": 228.68463706970215, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09669037780761719 }, "validation_seconds": 18.25190758705139 }, { "epoch": 63, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09671043863296509, "train_seconds": 228.63868236541748, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0957095199584961 }, "validation_seconds": 18.234554290771484 }, { "epoch": 64, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09636272977458106, "train_seconds": 228.15654468536377, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09310142059326172 }, "validation_seconds": 18.236981868743896 }, { "epoch": 65, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09650074934429592, "train_seconds": 228.1613838672638, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09531402130126954 }, "validation_seconds": 18.24026322364807 }, { "epoch": 66, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09634359340137906, "train_seconds": 228.17927384376526, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09655115051269532 }, "validation_seconds": 18.235257625579834 }, { "epoch": 67, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09597892744541169, "train_seconds": 228.24445581436157, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09803577880859375 }, "validation_seconds": 18.258594512939453 }, { "epoch": 68, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09632101210488213, "train_seconds": 228.73210763931274, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09556730346679687 }, "validation_seconds": 18.272592306137085 }, { "epoch": 69, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09618460634814369, "train_seconds": 228.6269886493683, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09790851898193359 }, "validation_seconds": 18.23514723777771 }, { "epoch": 70, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09652581750684314, "train_seconds": 228.1185097694397, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09527283630371093 }, "validation_seconds": 18.23189115524292 }, { "epoch": 71, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0956978734254837, "train_seconds": 228.124103307724, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09942215576171876 }, "validation_seconds": 18.23433780670166 }, { "epoch": 72, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09645447168217765, "train_seconds": 228.13148307800293, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09943707580566406 }, "validation_seconds": 18.238800525665283 }, { "epoch": 73, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09633775558339225, "train_seconds": 228.1336727142334, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09661944732666015 }, "validation_seconds": 18.240188598632812 }, { "epoch": 74, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09634426296022203, "train_seconds": 228.14839053153992, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09583654479980469 }, "validation_seconds": 18.23649311065674 }, { "epoch": 75, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09580984490447574, "train_seconds": 228.13686203956604, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09728650207519532 }, "validation_seconds": 18.236411094665527 }, { "epoch": 76, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09626505046553081, "train_seconds": 228.14012694358826, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09705371398925781 }, "validation_seconds": 18.238210678100586 }, { "epoch": 77, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09598139480352402, "train_seconds": 228.11646938323975, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09914067993164062 }, "validation_seconds": 18.237971544265747 }, { "epoch": 78, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0964213231338395, "train_seconds": 228.11607599258423, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09502608642578125 }, "validation_seconds": 18.238613843917847 }, { "epoch": 79, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09655453516244888, "train_seconds": 228.1036672592163, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09777397918701172 }, "validation_seconds": 18.237144470214844 }, { "epoch": 80, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09686183725330565, "train_seconds": 228.09968042373657, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09582723693847656 }, "validation_seconds": 18.23502254486084 }, { "epoch": 81, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09596207317643696, "train_seconds": 228.0861895084381, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09246314697265624 }, "validation_seconds": 18.233213663101196 }, { "epoch": 82, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09665712188747194, "train_seconds": 228.10383939743042, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09760987091064453 }, "validation_seconds": 18.23534655570984 }, { "epoch": 83, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09619577255249023, "train_seconds": 228.10573935508728, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09589176177978516 }, "validation_seconds": 18.236335515975952 }, { "epoch": 84, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09611746475564109, "train_seconds": 228.09352922439575, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09739808807373047 }, "validation_seconds": 18.236633777618408 }, { "epoch": 85, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09611809586021636, "train_seconds": 228.1817696094513, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09837869110107422 }, "validation_seconds": 18.24014949798584 }, { "epoch": 86, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09562978496021694, "train_seconds": 228.21705293655396, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09497322692871094 }, "validation_seconds": 18.23759365081787 }, { "epoch": 87, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09583760881291495, "train_seconds": 228.14998769760132, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0954147964477539 }, "validation_seconds": 18.23929476737976 }, { "epoch": 88, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09554384915033977, "train_seconds": 228.06396865844727, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09424874877929687 }, "validation_seconds": 18.236170053482056 }, { "epoch": 89, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09572083793216281, "train_seconds": 228.18749070167542, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09340807495117187 }, "validation_seconds": 18.244277715682983 }, { "epoch": 90, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09604131587876213, "train_seconds": 228.1858491897583, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09667099304199218 }, "validation_seconds": 18.237085342407227 }, { "epoch": 91, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09581272191471524, "train_seconds": 228.2162148952484, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09316986083984374 }, "validation_seconds": 18.237507104873657 }, { "epoch": 92, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09599215721024407, "train_seconds": 228.16782188415527, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09519173889160157 }, "validation_seconds": 18.24071955680847 }, { "epoch": 93, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09601165684858957, "train_seconds": 228.21594524383545, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09708834075927734 }, "validation_seconds": 18.237923622131348 }, { "epoch": 94, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0963773810784022, "train_seconds": 228.19517064094543, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09413018035888672 }, "validation_seconds": 18.24626135826111 }, { "epoch": 95, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09622333311769697, "train_seconds": 228.20594668388367, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09718131561279297 }, "validation_seconds": 18.239275217056274 }, { "epoch": 96, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09596092004113727, "train_seconds": 228.1629090309143, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09380335388183594 }, "validation_seconds": 18.23921275138855 }, { "epoch": 97, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0961335700088077, "train_seconds": 228.2011206150055, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09569081115722657 }, "validation_seconds": 18.25113034248352 }, { "epoch": 98, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0961867148703999, "train_seconds": 228.19716882705688, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09790823364257813 }, "validation_seconds": 18.23853349685669 }, { "epoch": 99, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09642944700850381, "train_seconds": 228.1831932067871, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09380260467529297 }, "validation_seconds": 18.241806268692017 }, { "epoch": 100, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09570178092850579, "train_seconds": 228.12436485290527, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09370457153320312 }, "validation_seconds": 18.23480200767517 } ], "evaluation_protocol": { "split": "validation", "test_evaluations": 0, "test_used_for_selection": false }, "final": { "accuracy": 0.1, "evaluation_seconds": 18.23977279663086, "examples": 5000, "finite": true, "loss": 0.09370457153320312 }, "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": 6988224512, "peak_memory_reserved_bytes": 8430551040, "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": 24661.099477529526, "work": { "candidate_label_evaluation_presentations": 0, "completed_global_epochs": 100, "feedforward_example_passes": 5005000, "forward_macs_per_example": 68862592, "forward_parameter_count": 464154, "local_target_backward_example_evaluations": 0, "local_vjp_example_evaluations": 118100000, "logical_task_loss_queries": 0, "num_trainable_layers": 32, "ordinary_training_examples": 4500000, "ordinary_validation_examples": 505000, "relaxation_example_passes": 118100000, "training_example_presentations": 4500000 } }