{ "architecture": { "base_width": 16, "depth": 56, "family": "CIFAR 6n+2 ResNet, option-A shortcuts", "forward_parameter_count": 853018, "normalization": "batchnorm", "residual_scale": 1.0 }, "args": { "augment_train": 1, "batch_size": 128, "data_dir": "/orion/u/oscarwan/scratch/sdil-data", "depth": 56, "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-d56-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": 0.1049093120151096, "train_seconds": 470.40172600746155, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09854823455810546 }, "validation_seconds": 42.355618476867676 }, { "epoch": 2, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09679424016343223, "train_seconds": 466.0533540248871, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09679457397460937 }, "validation_seconds": 42.082069396972656 }, { "epoch": 3, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09715077778763241, "train_seconds": 465.50510478019714, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09756388244628907 }, "validation_seconds": 41.84662628173828 }, { "epoch": 4, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0970374315155877, "train_seconds": 464.9111385345459, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0956327865600586 }, "validation_seconds": 42.19015383720398 }, { "epoch": 5, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09693956523074045, "train_seconds": 465.1891815662384, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09490829162597657 }, "validation_seconds": 41.88029718399048 }, { "epoch": 6, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09654380642440584, "train_seconds": 465.1731870174408, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0966841812133789 }, "validation_seconds": 42.02381706237793 }, { "epoch": 7, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09670941620535321, "train_seconds": 465.4678430557251, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09708750610351563 }, "validation_seconds": 41.63112282752991 }, { "epoch": 8, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09702321206993526, "train_seconds": 464.5645751953125, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09724942169189453 }, "validation_seconds": 41.95182228088379 }, { "epoch": 9, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0974015151196056, "train_seconds": 464.3173403739929, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0945250732421875 }, "validation_seconds": 41.758512020111084 }, { "epoch": 10, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09567246304617988, "train_seconds": 464.4738943576813, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09481480102539062 }, "validation_seconds": 41.882903814315796 }, { "epoch": 11, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09578449822399351, "train_seconds": 464.7627408504486, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09709003295898437 }, "validation_seconds": 41.83140850067139 }, { "epoch": 12, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09586407614946366, "train_seconds": 464.4240849018097, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0982633773803711 }, "validation_seconds": 41.97194838523865 }, { "epoch": 13, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09627521215942171, "train_seconds": 464.32663130760193, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09660293426513672 }, "validation_seconds": 41.93223762512207 }, { "epoch": 14, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09617627680831485, "train_seconds": 464.46084427833557, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09428795776367188 }, "validation_seconds": 42.017765522003174 }, { "epoch": 15, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09564328297244178, "train_seconds": 465.19329810142517, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09600251922607422 }, "validation_seconds": 42.234293937683105 }, { "epoch": 16, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09563761240641276, "train_seconds": 464.90938997268677, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09587730102539062 }, "validation_seconds": 42.10805058479309 }, { "epoch": 17, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0960964433060752, "train_seconds": 465.1038143634796, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09685572204589844 }, "validation_seconds": 42.27157926559448 }, { "epoch": 18, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09609683181047439, "train_seconds": 466.1774206161499, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09549898681640626 }, "validation_seconds": 42.174476623535156 }, { "epoch": 19, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09612310879892773, "train_seconds": 466.1518986225128, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09561274871826173 }, "validation_seconds": 42.16247892379761 }, { "epoch": 20, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09647270717223486, "train_seconds": 465.53555250167847, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09512040252685547 }, "validation_seconds": 42.129403829574585 }, { "epoch": 21, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09616433023346795, "train_seconds": 465.08661556243896, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0983990463256836 }, "validation_seconds": 42.04495024681091 }, { "epoch": 22, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09624761987659666, "train_seconds": 464.83717226982117, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0956104736328125 }, "validation_seconds": 42.02618384361267 }, { "epoch": 23, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09572103581825891, "train_seconds": 465.7274680137634, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0967745132446289 }, "validation_seconds": 42.163888931274414 }, { "epoch": 24, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09640631458759308, "train_seconds": 465.5332636833191, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09754253997802734 }, "validation_seconds": 42.26589584350586 }, { "epoch": 25, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09652167629798253, "train_seconds": 465.39615631103516, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09833413848876953 }, "validation_seconds": 42.07328724861145 }, { "epoch": 26, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09637802203363842, "train_seconds": 465.0871801376343, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09762132720947266 }, "validation_seconds": 41.8577835559845 }, { "epoch": 27, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09588909540838665, "train_seconds": 465.2824354171753, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09559941711425782 }, "validation_seconds": 42.20577144622803 }, { "epoch": 28, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09600038005246056, "train_seconds": 465.70421624183655, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09474451599121093 }, "validation_seconds": 41.94756460189819 }, { "epoch": 29, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09606572427219814, "train_seconds": 465.30998396873474, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09541961975097656 }, "validation_seconds": 42.07582426071167 }, { "epoch": 30, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09589181250201331, "train_seconds": 465.0965838432312, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09653666229248047 }, "validation_seconds": 42.29110312461853 }, { "epoch": 31, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09631023094124264, "train_seconds": 465.18078541755676, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0960676986694336 }, "validation_seconds": 42.05308723449707 }, { "epoch": 32, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09603364160723156, "train_seconds": 464.56597423553467, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09548921508789063 }, "validation_seconds": 41.80490756034851 }, { "epoch": 33, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09560389386018117, "train_seconds": 464.77473068237305, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09516982727050781 }, "validation_seconds": 42.142924308776855 }, { "epoch": 34, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09607277266846763, "train_seconds": 464.5014383792877, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09546457977294921 }, "validation_seconds": 42.0104558467865 }, { "epoch": 35, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09576119128863017, "train_seconds": 464.63925290107727, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09528033142089844 }, "validation_seconds": 42.067020416259766 }, { "epoch": 36, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09611127299070359, "train_seconds": 465.3532955646515, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09493963012695313 }, "validation_seconds": 41.93064761161804 }, { "epoch": 37, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09682440170579486, "train_seconds": 464.56911444664, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09636844940185547 }, "validation_seconds": 41.72228002548218 }, { "epoch": 38, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09608428453736835, "train_seconds": 463.5885651111603, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09487273864746094 }, "validation_seconds": 41.65004539489746 }, { "epoch": 39, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0963151536266009, "train_seconds": 464.0708773136139, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09748384704589844 }, "validation_seconds": 41.8119695186615 }, { "epoch": 40, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09639640080663893, "train_seconds": 464.3524947166443, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0965095932006836 }, "validation_seconds": 41.94960570335388 }, { "epoch": 41, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0956977930717998, "train_seconds": 464.94561409950256, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0935846435546875 }, "validation_seconds": 42.05508899688721 }, { "epoch": 42, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09583505643606186, "train_seconds": 465.431129693985, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09561628875732422 }, "validation_seconds": 41.988160371780396 }, { "epoch": 43, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09605752380688985, "train_seconds": 465.50451612472534, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09550503692626953 }, "validation_seconds": 42.18235206604004 }, { "epoch": 44, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0956310129072931, "train_seconds": 465.6041624546051, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09446280822753907 }, "validation_seconds": 42.08299422264099 }, { "epoch": 45, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09536773497660955, "train_seconds": 465.80098724365234, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09453959808349609 }, "validation_seconds": 42.09350252151489 }, { "epoch": 46, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0958818547460768, "train_seconds": 466.4351236820221, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09704439849853516 }, "validation_seconds": 42.08637857437134 }, { "epoch": 47, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09645931057135264, "train_seconds": 466.09434366226196, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0961579849243164 }, "validation_seconds": 42.06746745109558 }, { "epoch": 48, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0954085854821735, "train_seconds": 465.25619769096375, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0969421142578125 }, "validation_seconds": 41.731086015701294 }, { "epoch": 49, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09619432701004876, "train_seconds": 465.3983533382416, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0965331298828125 }, "validation_seconds": 41.816789865493774 }, { "epoch": 50, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09600847476058536, "train_seconds": 465.3787190914154, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09710611572265625 }, "validation_seconds": 42.01187825202942 }, { "epoch": 51, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09619079148901834, "train_seconds": 465.54860496520996, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0951966293334961 }, "validation_seconds": 42.02596402168274 }, { "epoch": 52, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09627778605222702, "train_seconds": 465.53260564804077, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09787473602294922 }, "validation_seconds": 41.86947846412659 }, { "epoch": 53, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09646207979785071, "train_seconds": 465.1533234119415, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09574561614990235 }, "validation_seconds": 41.8131160736084 }, { "epoch": 54, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09629596228996913, "train_seconds": 465.05003571510315, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0952306869506836 }, "validation_seconds": 41.844584703445435 }, { "epoch": 55, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09640434854692884, "train_seconds": 465.1228246688843, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09881563262939454 }, "validation_seconds": 41.718056440353394 }, { "epoch": 56, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0962243594262335, "train_seconds": 464.6118927001953, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09332275543212891 }, "validation_seconds": 41.883305311203 }, { "epoch": 57, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09593700580994288, "train_seconds": 464.6503734588623, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0978786376953125 }, "validation_seconds": 42.113295555114746 }, { "epoch": 58, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09609533179865944, "train_seconds": 464.24342918395996, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09458008117675781 }, "validation_seconds": 41.631030797958374 }, { "epoch": 59, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09571239232884513, "train_seconds": 464.2138133049011, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09771599426269531 }, "validation_seconds": 41.696253299713135 }, { "epoch": 60, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09557220001882977, "train_seconds": 464.86264300346375, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09608076782226563 }, "validation_seconds": 41.83806109428406 }, { "epoch": 61, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09603700024021997, "train_seconds": 464.84575176239014, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09600220947265625 }, "validation_seconds": 41.66805815696716 }, { "epoch": 62, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09620499748918745, "train_seconds": 464.15450048446655, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09603590545654297 }, "validation_seconds": 41.74782752990723 }, { "epoch": 63, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0962453338013755, "train_seconds": 464.59076714515686, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09555384216308593 }, "validation_seconds": 41.62419104576111 }, { "epoch": 64, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09611023871501287, "train_seconds": 464.39524030685425, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09319230194091797 }, "validation_seconds": 41.61292552947998 }, { "epoch": 65, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09617606501976649, "train_seconds": 464.5210211277008, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09480837554931641 }, "validation_seconds": 41.89604687690735 }, { "epoch": 66, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0961561086707645, "train_seconds": 464.26877069473267, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09668070220947265 }, "validation_seconds": 41.73206186294556 }, { "epoch": 67, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09578220612340503, "train_seconds": 464.26437520980835, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09633714294433594 }, "validation_seconds": 41.749727964401245 }, { "epoch": 68, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09601921813752916, "train_seconds": 464.677282333374, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0954210693359375 }, "validation_seconds": 41.653220653533936 }, { "epoch": 69, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0958134253554874, "train_seconds": 464.6186845302582, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09663692932128906 }, "validation_seconds": 41.856159687042236 }, { "epoch": 70, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09630409976376428, "train_seconds": 463.54873037338257, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09525140838623047 }, "validation_seconds": 41.317835092544556 }, { "epoch": 71, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09535665365589989, "train_seconds": 462.5900402069092, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09930306091308594 }, "validation_seconds": 41.25684332847595 }, { "epoch": 72, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09590638970666461, "train_seconds": 462.9886476993561, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09919052581787109 }, "validation_seconds": 41.03996682167053 }, { "epoch": 73, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09605056209696664, "train_seconds": 462.34330105781555, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09603641967773438 }, "validation_seconds": 41.09140729904175 }, { "epoch": 74, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09621421782573064, "train_seconds": 462.92142152786255, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09542268829345703 }, "validation_seconds": 41.37644982337952 }, { "epoch": 75, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09559676223993302, "train_seconds": 462.5050027370453, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.095793212890625 }, "validation_seconds": 41.256924629211426 }, { "epoch": 76, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09604653212229411, "train_seconds": 462.49127101898193, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09619957275390625 }, "validation_seconds": 41.06153750419617 }, { "epoch": 77, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09564897516965866, "train_seconds": 462.3650414943695, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09792278289794921 }, "validation_seconds": 41.19801068305969 }, { "epoch": 78, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09598682165808148, "train_seconds": 462.66747975349426, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09466236419677734 }, "validation_seconds": 41.33188843727112 }, { "epoch": 79, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09604800218343734, "train_seconds": 463.091757774353, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09714752655029296 }, "validation_seconds": 41.29967379570007 }, { "epoch": 80, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0963094130622016, "train_seconds": 462.95998096466064, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09555538482666015 }, "validation_seconds": 41.277440547943115 }, { "epoch": 81, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09557393687168757, "train_seconds": 462.97270131111145, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09269241485595703 }, "validation_seconds": 41.38614511489868 }, { "epoch": 82, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09606576175159878, "train_seconds": 462.94847536087036, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09701722106933594 }, "validation_seconds": 41.294069051742554 }, { "epoch": 83, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09575901180903117, "train_seconds": 463.94762802124023, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09515934143066407 }, "validation_seconds": 41.48845863342285 }, { "epoch": 84, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09559025944206449, "train_seconds": 463.4072983264923, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09719501342773437 }, "validation_seconds": 41.55417251586914 }, { "epoch": 85, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09569342220359378, "train_seconds": 462.94681215286255, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09763941040039062 }, "validation_seconds": 41.40012240409851 }, { "epoch": 86, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09554593718979094, "train_seconds": 463.18996381759644, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09514545745849609 }, "validation_seconds": 41.31304478645325 }, { "epoch": 87, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09627738233407339, "train_seconds": 463.1316273212433, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.0954114501953125 }, "validation_seconds": 41.61716270446777 }, { "epoch": 88, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09551246789958742, "train_seconds": 463.3723156452179, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09582576599121094 }, "validation_seconds": 41.4737389087677 }, { "epoch": 89, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09581275145345264, "train_seconds": 463.5044505596161, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09365298461914062 }, "validation_seconds": 41.32333970069885 }, { "epoch": 90, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09594039207034641, "train_seconds": 463.22664976119995, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09755116119384766 }, "validation_seconds": 41.452555894851685 }, { "epoch": 91, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09589890376726787, "train_seconds": 463.3689317703247, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09362464294433594 }, "validation_seconds": 41.49372625350952 }, { "epoch": 92, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09604071430762609, "train_seconds": 463.23748564720154, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09576106414794922 }, "validation_seconds": 41.29102540016174 }, { "epoch": 93, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0958403738392724, "train_seconds": 462.9280080795288, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09767513275146485 }, "validation_seconds": 41.327802896499634 }, { "epoch": 94, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09607556787464354, "train_seconds": 463.41283416748047, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09459663238525391 }, "validation_seconds": 41.50932788848877 }, { "epoch": 95, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09609565815925598, "train_seconds": 463.442754983902, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09667721099853516 }, "validation_seconds": 41.346176624298096 }, { "epoch": 96, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09569309736887614, "train_seconds": 462.8477804660797, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09482781524658203 }, "validation_seconds": 41.56019306182861 }, { "epoch": 97, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09579816060596043, "train_seconds": 463.14385962486267, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09566875915527344 }, "validation_seconds": 41.318729877471924 }, { "epoch": 98, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.0960154987997479, "train_seconds": 463.26160764694214, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09674229278564453 }, "validation_seconds": 41.44693160057068 }, { "epoch": 99, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09639488592677646, "train_seconds": 462.9610106945038, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09510955657958985 }, "validation_seconds": 41.40132117271423 }, { "epoch": 100, "lr": 0.0003, "output_lr": 0.0003, "train_examples": 45000, "train_loss": 0.09564637204408645, "train_seconds": 463.3234283924103, "validation": { "accuracy": 0.1, "examples": 5000, "loss": 0.09450665283203125 }, "validation_seconds": 41.3101544380188 } ], "evaluation_protocol": { "split": "validation", "test_evaluations": 0, "test_used_for_selection": false }, "final": { "accuracy": 0.1, "evaluation_seconds": 41.240410566329956, "examples": 5000, "finite": true, "loss": 0.09450665283203125 }, "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": 11683346432, "peak_memory_reserved_bytes": 13419675648, "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": 50663.67151045799, "work": { "candidate_label_evaluation_presentations": 0, "completed_global_epochs": 100, "feedforward_example_passes": 5005000, "forward_macs_per_example": 125485696, "forward_parameter_count": 853018, "local_target_backward_example_evaluations": 0, "local_vjp_example_evaluations": 118100000, "logical_task_loss_queries": 0, "num_trainable_layers": 56, "ordinary_training_examples": 4500000, "ordinary_validation_examples": 505000, "relaxation_example_passes": 118100000, "training_example_presentations": 4500000 } }