summaryrefslogtreecommitdiff
path: root/results/ep_depth2_v1_mnist_bp_w500_d2_s0.json
blob: fe5470844b8d168a662129538b06f2893b3bb6a2 (plain)
1
{"args": {"mode": "bp", "dataset": "mnist", "depth": 2, "width": 500, "act": "tanh", "residual": 0, "epochs": 60, "batch_size": 128, "train_examples": 50000, "eta": 0.05, "eta_A": 0.02, "eta_P": 0.002, "momentum": 0.9, "w_scale": 1.0, "a_scale": 1.0, "pert_sigma": 0.01, "pert_every": 4, "pert_ndirs": 16, "pert_mode": "simultaneous", "use_residual": 1, "raw_scale_control": "none", "learn_A": 1, "learn_P": 1, "p_neutral": 1, "p_warmup_steps": 200, "p_warmup_eta": 0.05, "nuis_rho": 0.0, "predictor_mode": "diagonal", "normalize_delta": 0, "settle_steps": 0, "kappa": 0.0, "feedback": "error", "seed": 0, "device": "cuda", "log_every": 50, "max_steps": 0, "probe_bs": 512, "outdir": "results", "tag": "ep_depth2_v1_mnist_bp_w500_d2_s0", "n_in": 784}, "provenance": {"git_commit": "f2b6107e20975aa649c4ab36b3faab21d1e3c281", "git_dirty": false}, "steps": [{"step": 0, "epoch": 0, "train_loss": 2.455000400543213}, {"step": 50, "epoch": 0, "train_loss": 0.4258231222629547}, {"step": 100, "epoch": 0, "train_loss": 0.1558641940355301}, {"step": 150, "epoch": 0, "train_loss": 0.19458484649658203}, {"step": 200, "epoch": 0, "train_loss": 0.22008219361305237}, {"step": 250, "epoch": 0, "train_loss": 0.24774634838104248}, {"step": 300, "epoch": 0, "train_loss": 0.13055270910263062}, {"step": 350, "epoch": 0, "train_loss": 0.18709483742713928}, {"epoch_end": 0, "step": 391, "test_acc": 0.9631, "test_loss": 0.12574152221679688}, {"step": 400, "epoch": 1, "train_loss": 0.08105223625898361}, {"step": 450, "epoch": 1, "train_loss": 0.1096838116645813}, {"step": 500, "epoch": 1, "train_loss": 0.15618440508842468}, {"step": 550, "epoch": 1, "train_loss": 0.04939229413866997}, {"step": 600, "epoch": 1, "train_loss": 0.051858898252248764}, {"step": 650, "epoch": 1, "train_loss": 0.05193959176540375}, {"step": 700, "epoch": 1, "train_loss": 0.06250821053981781}, {"step": 750, "epoch": 1, "train_loss": 0.10700735449790955}, {"epoch_end": 1, "step": 782, "test_acc": 0.9635, "test_loss": 0.11929717254638672}, {"step": 800, "epoch": 2, "train_loss": 0.046721894294023514}, {"step": 850, "epoch": 2, "train_loss": 0.03415483981370926}, {"step": 900, "epoch": 2, "train_loss": 0.02530835196375847}, {"step": 950, "epoch": 2, "train_loss": 0.0893852710723877}, {"step": 1000, "epoch": 2, "train_loss": 0.04095539078116417}, {"step": 1050, "epoch": 2, "train_loss": 0.03202667459845543}, {"step": 1100, "epoch": 2, "train_loss": 0.08770047873258591}, {"step": 1150, "epoch": 2, "train_loss": 0.08304520696401596}, {"epoch_end": 2, "step": 1173, "test_acc": 0.9728, "test_loss": 0.08466080265045166}, {"step": 1200, "epoch": 3, "train_loss": 0.02037188783288002}, {"step": 1250, "epoch": 3, "train_loss": 0.03868574649095535}, {"step": 1300, "epoch": 3, "train_loss": 0.004869620315730572}, {"step": 1350, "epoch": 3, "train_loss": 0.022573422640562057}, {"step": 1400, "epoch": 3, "train_loss": 0.0182911679148674}, {"step": 1450, "epoch": 3, "train_loss": 0.02544417604804039}, {"step": 1500, "epoch": 3, "train_loss": 0.06943915039300919}, {"step": 1550, "epoch": 3, "train_loss": 0.012584048323333263}, {"epoch_end": 3, "step": 1564, "test_acc": 0.9741, "test_loss": 0.09150042858123779}, {"step": 1600, "epoch": 4, "train_loss": 0.03463534265756607}, {"step": 1650, "epoch": 4, "train_loss": 0.033722925931215286}, {"step": 1700, "epoch": 4, "train_loss": 0.04138275235891342}, {"step": 1750, "epoch": 4, "train_loss": 0.005293062422424555}, {"step": 1800, "epoch": 4, "train_loss": 0.007497875951230526}, {"step": 1850, "epoch": 4, "train_loss": 0.03764200210571289}, {"step": 1900, "epoch": 4, "train_loss": 0.007082010619342327}, {"step": 1950, "epoch": 4, "train_loss": 0.009001076221466064}, {"epoch_end": 4, "step": 1955, "test_acc": 0.9787, "test_loss": 0.06898830814361573}, {"step": 2000, "epoch": 5, "train_loss": 0.012570385821163654}, {"step": 2050, "epoch": 5, "train_loss": 0.0037833440583199263}, {"step": 2100, "epoch": 5, "train_loss": 0.005348308477550745}, {"step": 2150, "epoch": 5, "train_loss": 0.0012415433302521706}, {"step": 2200, "epoch": 5, "train_loss": 0.01940673030912876}, {"step": 2250, "epoch": 5, "train_loss": 0.0038129370659589767}, {"step": 2300, "epoch": 5, "train_loss": 0.004817248787730932}, {"epoch_end": 5, "step": 2346, "test_acc": 0.9811, "test_loss": 0.06667604627609253}, {"step": 2350, "epoch": 6, "train_loss": 0.002647221088409424}, {"step": 2400, "epoch": 6, "train_loss": 0.000999102951027453}, {"step": 2450, "epoch": 6, "train_loss": 0.0017519164830446243}, {"step": 2500, "epoch": 6, "train_loss": 0.0030223140493035316}, {"step": 2550, "epoch": 6, "train_loss": 0.004342074505984783}, {"step": 2600, "epoch": 6, "train_loss": 0.0016586732817813754}, {"step": 2650, "epoch": 6, "train_loss": 0.001031461637467146}, {"step": 2700, "epoch": 6, "train_loss": 0.0011721968185156584}, {"epoch_end": 6, "step": 2737, "test_acc": 0.983, "test_loss": 0.06367138085365295}, {"step": 2750, "epoch": 7, "train_loss": 0.0004316671402193606}, {"step": 2800, "epoch": 7, "train_loss": 0.0011546170571818948}, {"step": 2850, "epoch": 7, "train_loss": 0.0017482187831774354}, {"step": 2900, "epoch": 7, "train_loss": 0.0014804452657699585}, {"step": 2950, "epoch": 7, "train_loss": 0.0013923580991104245}, {"step": 3000, "epoch": 7, "train_loss": 0.0004969580331817269}, {"step": 3050, "epoch": 7, "train_loss": 0.00102445506490767}, {"step": 3100, "epoch": 7, "train_loss": 0.0007330214721150696}, {"epoch_end": 7, "step": 3128, "test_acc": 0.9817, "test_loss": 0.06497008666992188}, {"step": 3150, "epoch": 8, "train_loss": 0.0007266164175234735}, {"step": 3200, "epoch": 8, "train_loss": 0.0005904819117859006}, {"step": 3250, "epoch": 8, "train_loss": 0.00035752521944232285}, {"step": 3300, "epoch": 8, "train_loss": 0.0007480670465156436}, {"step": 3350, "epoch": 8, "train_loss": 0.0010110170114785433}, {"step": 3400, "epoch": 8, "train_loss": 0.00046568087418563664}, {"step": 3450, "epoch": 8, "train_loss": 0.0020568638574332}, {"step": 3500, "epoch": 8, "train_loss": 0.0020604399032890797}, {"epoch_end": 8, "step": 3519, "test_acc": 0.9834, "test_loss": 0.06305290551185608}, {"step": 3550, "epoch": 9, "train_loss": 0.0011258182348683476}, {"step": 3600, "epoch": 9, "train_loss": 0.000561850902158767}, {"step": 3650, "epoch": 9, "train_loss": 0.0007293248199857771}, {"step": 3700, "epoch": 9, "train_loss": 0.0008180361473932862}, {"step": 3750, "epoch": 9, "train_loss": 0.0008131786016747355}, {"step": 3800, "epoch": 9, "train_loss": 0.00045516283717006445}, {"step": 3850, "epoch": 9, "train_loss": 0.0009639261988922954}, {"step": 3900, "epoch": 9, "train_loss": 0.00044355198042467237}, {"epoch_end": 9, "step": 3910, "test_acc": 0.9825, "test_loss": 0.06331720509529114}, {"step": 3950, "epoch": 10, "train_loss": 0.0003410671779420227}, {"step": 4000, "epoch": 10, "train_loss": 0.00045174304977990687}, {"step": 4050, "epoch": 10, "train_loss": 0.0005011526518501341}, {"step": 4100, "epoch": 10, "train_loss": 0.0005138218402862549}, {"step": 4150, "epoch": 10, "train_loss": 0.00018292287131771445}, {"step": 4200, "epoch": 10, "train_loss": 0.0005085116135887802}, {"step": 4250, "epoch": 10, "train_loss": 0.0015549617819488049}, {"step": 4300, "epoch": 10, "train_loss": 0.0007714828243479133}, {"epoch_end": 10, "step": 4301, "test_acc": 0.983, "test_loss": 0.06405115609169007}, {"step": 4350, "epoch": 11, "train_loss": 0.00043703374103643}, {"step": 4400, "epoch": 11, "train_loss": 0.0005781386862508953}, {"step": 4450, "epoch": 11, "train_loss": 0.0004706254112534225}, {"step": 4500, "epoch": 11, "train_loss": 0.00012892181985080242}, {"step": 4550, "epoch": 11, "train_loss": 0.0005865339771844447}, {"step": 4600, "epoch": 11, "train_loss": 0.00030653938301838934}, {"step": 4650, "epoch": 11, "train_loss": 0.0003076840948779136}, {"epoch_end": 11, "step": 4692, "test_acc": 0.9834, "test_loss": 0.06431370944976807}, {"step": 4700, "epoch": 12, "train_loss": 0.0006583298090845346}, {"step": 4750, "epoch": 12, "train_loss": 0.000330798706272617}, {"step": 4800, "epoch": 12, "train_loss": 0.00031928985845297575}, {"step": 4850, "epoch": 12, "train_loss": 0.0004050260758958757}, {"step": 4900, "epoch": 12, "train_loss": 0.0003135697916150093}, {"step": 4950, "epoch": 12, "train_loss": 0.00023153075017035007}, {"step": 5000, "epoch": 12, "train_loss": 0.00022012891713529825}, {"step": 5050, "epoch": 12, "train_loss": 0.00036280619679018855}, {"epoch_end": 12, "step": 5083, "test_acc": 0.9831, "test_loss": 0.0650487340927124}, {"step": 5100, "epoch": 13, "train_loss": 0.00020734767895191908}, {"step": 5150, "epoch": 13, "train_loss": 0.00024115675478242338}, {"step": 5200, "epoch": 13, "train_loss": 0.00032544726855121553}, {"step": 5250, "epoch": 13, "train_loss": 0.0002673465642146766}, {"step": 5300, "epoch": 13, "train_loss": 0.0005078844260424376}, {"step": 5350, "epoch": 13, "train_loss": 0.0004899374907836318}, {"step": 5400, "epoch": 13, "train_loss": 0.0002839984081219882}, {"step": 5450, "epoch": 13, "train_loss": 0.00032767534139566123}, {"epoch_end": 13, "step": 5474, "test_acc": 0.9828, "test_loss": 0.06566979150772095}, {"step": 5500, "epoch": 14, "train_loss": 0.00017136841779574752}, {"step": 5550, "epoch": 14, "train_loss": 0.00016080682689789683}, {"step": 5600, "epoch": 14, "train_loss": 0.0003217956400476396}, {"step": 5650, "epoch": 14, "train_loss": 0.00024880957789719105}, {"step": 5700, "epoch": 14, "train_loss": 0.00010403380292700604}, {"step": 5750, "epoch": 14, "train_loss": 0.00034919678000733256}, {"step": 5800, "epoch": 14, "train_loss": 0.00044214047375135124}, {"step": 5850, "epoch": 14, "train_loss": 0.0003756832447834313}, {"epoch_end": 14, "step": 5865, "test_acc": 0.9829, "test_loss": 0.06549515943527222}, {"step": 5900, "epoch": 15, "train_loss": 0.0002283672511111945}, {"step": 5950, "epoch": 15, "train_loss": 0.0004711742512881756}, {"step": 6000, "epoch": 15, "train_loss": 0.0005615975824184716}, {"step": 6050, "epoch": 15, "train_loss": 0.0005116185056976974}, {"step": 6100, "epoch": 15, "train_loss": 0.0003569346445146948}, {"step": 6150, "epoch": 15, "train_loss": 0.00026347232051193714}, {"step": 6200, "epoch": 15, "train_loss": 0.00021341886895243078}, {"step": 6250, "epoch": 15, "train_loss": 0.00023101814440451562}, {"epoch_end": 15, "step": 6256, "test_acc": 0.9833, "test_loss": 0.06582820467948913}, {"step": 6300, "epoch": 16, "train_loss": 0.00037468253867700696}, {"step": 6350, "epoch": 16, "train_loss": 0.0002497027744539082}, {"step": 6400, "epoch": 16, "train_loss": 0.0002483805874362588}, {"step": 6450, "epoch": 16, "train_loss": 0.00016304900054819882}, {"step": 6500, "epoch": 16, "train_loss": 0.00022628418810199946}, {"step": 6550, "epoch": 16, "train_loss": 0.00033413682831451297}, {"step": 6600, "epoch": 16, "train_loss": 0.0001534582261228934}, {"epoch_end": 16, "step": 6647, "test_acc": 0.9832, "test_loss": 0.06650953860282897}, {"step": 6650, "epoch": 17, "train_loss": 0.00013430760009214282}, {"step": 6700, "epoch": 17, "train_loss": 8.072109631029889e-05}, {"step": 6750, "epoch": 17, "train_loss": 0.00042716783354990184}, {"step": 6800, "epoch": 17, "train_loss": 0.0004267847689334303}, {"step": 6850, "epoch": 17, "train_loss": 0.00024257028417196125}, {"step": 6900, "epoch": 17, "train_loss": 0.0004263312148395926}, {"step": 6950, "epoch": 17, "train_loss": 0.0004024652298539877}, {"step": 7000, "epoch": 17, "train_loss": 0.00026760788750834763}, {"epoch_end": 17, "step": 7038, "test_acc": 0.9827, "test_loss": 0.06678133344650268}, {"step": 7050, "epoch": 18, "train_loss": 7.28661980247125e-05}, {"step": 7100, "epoch": 18, "train_loss": 0.00043220294173806906}, {"step": 7150, "epoch": 18, "train_loss": 0.00010792810644488782}, {"step": 7200, "epoch": 18, "train_loss": 0.00026146828895434737}, {"step": 7250, "epoch": 18, "train_loss": 0.0002680715697351843}, {"step": 7300, "epoch": 18, "train_loss": 0.00015434525266755372}, {"step": 7350, "epoch": 18, "train_loss": 0.0003079807211179286}, {"step": 7400, "epoch": 18, "train_loss": 0.0001408726820955053}, {"epoch_end": 18, "step": 7429, "test_acc": 0.983, "test_loss": 0.06710904097557067}, {"step": 7450, "epoch": 19, "train_loss": 8.190803055185825e-05}, {"step": 7500, "epoch": 19, "train_loss": 0.00018964319315273315}, {"step": 7550, "epoch": 19, "train_loss": 0.0001258395641343668}, {"step": 7600, "epoch": 19, "train_loss": 0.00011746882955776528}, {"step": 7650, "epoch": 19, "train_loss": 0.0002915307122748345}, {"step": 7700, "epoch": 19, "train_loss": 0.0003669032594189048}, {"step": 7750, "epoch": 19, "train_loss": 0.00035408782423473895}, {"step": 7800, "epoch": 19, "train_loss": 0.0003136630402877927}, {"epoch_end": 19, "step": 7820, "test_acc": 0.9826, "test_loss": 0.06760840563774109}, {"step": 7850, "epoch": 20, "train_loss": 0.00017070969624910504}, {"step": 7900, "epoch": 20, "train_loss": 0.000212626822758466}, {"step": 7950, "epoch": 20, "train_loss": 0.00021706342522520572}, {"step": 8000, "epoch": 20, "train_loss": 0.00014773097063880414}, {"step": 8050, "epoch": 20, "train_loss": 0.0002754788438323885}, {"step": 8100, "epoch": 20, "train_loss": 0.0002578064741101116}, {"step": 8150, "epoch": 20, "train_loss": 0.00014915165957063437}, {"step": 8200, "epoch": 20, "train_loss": 0.0001236435055034235}, {"epoch_end": 20, "step": 8211, "test_acc": 0.9829, "test_loss": 0.0674928373336792}, {"step": 8250, "epoch": 21, "train_loss": 0.0003566615632735193}, {"step": 8300, "epoch": 21, "train_loss": 0.0002463156415615231}, {"step": 8350, "epoch": 21, "train_loss": 0.00023629865609109402}, {"step": 8400, "epoch": 21, "train_loss": 0.00016280171985272318}, {"step": 8450, "epoch": 21, "train_loss": 0.0002574823156464845}, {"step": 8500, "epoch": 21, "train_loss": 0.0002660335449036211}, {"step": 8550, "epoch": 21, "train_loss": 0.0002279353211633861}, {"step": 8600, "epoch": 21, "train_loss": 0.0003212873707525432}, {"epoch_end": 21, "step": 8602, "test_acc": 0.9828, "test_loss": 0.06791936025619506}, {"step": 8650, "epoch": 22, "train_loss": 9.89121908787638e-05}, {"step": 8700, "epoch": 22, "train_loss": 0.0001375317806378007}, {"step": 8750, "epoch": 22, "train_loss": 0.00014341005589812994}, {"step": 8800, "epoch": 22, "train_loss": 0.00021778515656478703}, {"step": 8850, "epoch": 22, "train_loss": 0.00016443269851151854}, {"step": 8900, "epoch": 22, "train_loss": 0.0003277600044384599}, {"step": 8950, "epoch": 22, "train_loss": 0.0001869514089776203}, {"epoch_end": 22, "step": 8993, "test_acc": 0.983, "test_loss": 0.068073020029068}, {"step": 9000, "epoch": 23, "train_loss": 0.00016107289411593229}, {"step": 9050, "epoch": 23, "train_loss": 0.00018997443839907646}, {"step": 9100, "epoch": 23, "train_loss": 0.00011393775639589876}, {"step": 9150, "epoch": 23, "train_loss": 0.00023429698194377124}, {"step": 9200, "epoch": 23, "train_loss": 0.00035308668157085776}, {"step": 9250, "epoch": 23, "train_loss": 0.0001855173468356952}, {"step": 9300, "epoch": 23, "train_loss": 0.00016155109915416688}, {"step": 9350, "epoch": 23, "train_loss": 0.00015138505841605365}, {"epoch_end": 23, "step": 9384, "test_acc": 0.983, "test_loss": 0.06856325650215149}, {"step": 9400, "epoch": 24, "train_loss": 0.00013833236880600452}, {"step": 9450, "epoch": 24, "train_loss": 0.00012217345647513866}, {"step": 9500, "epoch": 24, "train_loss": 0.00022099098714534193}, {"step": 9550, "epoch": 24, "train_loss": 0.00018132691911887378}, {"step": 9600, "epoch": 24, "train_loss": 0.00017276736616622657}, {"step": 9650, "epoch": 24, "train_loss": 0.0003719521046150476}, {"step": 9700, "epoch": 24, "train_loss": 0.00014987040776759386}, {"step": 9750, "epoch": 24, "train_loss": 0.0003004814498126507}, {"epoch_end": 24, "step": 9775, "test_acc": 0.9829, "test_loss": 0.06860575995445252}, {"step": 9800, "epoch": 25, "train_loss": 0.00026047040591947734}, {"step": 9850, "epoch": 25, "train_loss": 9.5279399829451e-05}, {"step": 9900, "epoch": 25, "train_loss": 0.00012814371439162642}, {"step": 9950, "epoch": 25, "train_loss": 0.0001721162407193333}, {"step": 10000, "epoch": 25, "train_loss": 0.00023732829140499234}, {"step": 10050, "epoch": 25, "train_loss": 0.0001835065777413547}, {"step": 10100, "epoch": 25, "train_loss": 0.00011126815661555156}, {"step": 10150, "epoch": 25, "train_loss": 7.807118527125567e-05}, {"epoch_end": 25, "step": 10166, "test_acc": 0.9829, "test_loss": 0.06891820969581604}, {"step": 10200, "epoch": 26, "train_loss": 0.00020898175716865808}, {"step": 10250, "epoch": 26, "train_loss": 0.0001739787549013272}, {"step": 10300, "epoch": 26, "train_loss": 0.0005611208034679294}, {"step": 10350, "epoch": 26, "train_loss": 6.420022691600025e-05}, {"step": 10400, "epoch": 26, "train_loss": 0.0002528376644477248}, {"step": 10450, "epoch": 26, "train_loss": 0.00027096158009953797}, {"step": 10500, "epoch": 26, "train_loss": 0.00028754991944879293}, {"step": 10550, "epoch": 26, "train_loss": 0.00017238639702554792}, {"epoch_end": 26, "step": 10557, "test_acc": 0.9828, "test_loss": 0.06915889058113098}, {"step": 10600, "epoch": 27, "train_loss": 0.00015910649381112307}, {"step": 10650, "epoch": 27, "train_loss": 0.00013032725837547332}, {"step": 10700, "epoch": 27, "train_loss": 0.00014026474673300982}, {"step": 10750, "epoch": 27, "train_loss": 9.386497549712658e-05}, {"step": 10800, "epoch": 27, "train_loss": 9.271098679164425e-05}, {"step": 10850, "epoch": 27, "train_loss": 0.00011809764691861346}, {"step": 10900, "epoch": 27, "train_loss": 0.0001386977091897279}, {"epoch_end": 27, "step": 10948, "test_acc": 0.983, "test_loss": 0.06935790076255799}, {"step": 10950, "epoch": 28, "train_loss": 0.00011927788727916777}, {"step": 11000, "epoch": 28, "train_loss": 0.0001234469236806035}, {"step": 11050, "epoch": 28, "train_loss": 0.00019285896269138902}, {"step": 11100, "epoch": 28, "train_loss": 0.00016927815158851445}, {"step": 11150, "epoch": 28, "train_loss": 0.0001999600208364427}, {"step": 11200, "epoch": 28, "train_loss": 0.0002457947994116694}, {"step": 11250, "epoch": 28, "train_loss": 0.0002256510779261589}, {"step": 11300, "epoch": 28, "train_loss": 0.000164991884957999}, {"epoch_end": 28, "step": 11339, "test_acc": 0.9829, "test_loss": 0.06949651594161987}, {"step": 11350, "epoch": 29, "train_loss": 8.639090083306655e-05}, {"step": 11400, "epoch": 29, "train_loss": 0.00012994823919143528}, {"step": 11450, "epoch": 29, "train_loss": 0.00014419502986129373}, {"step": 11500, "epoch": 29, "train_loss": 0.00020839109492953867}, {"step": 11550, "epoch": 29, "train_loss": 0.0001562200195621699}, {"step": 11600, "epoch": 29, "train_loss": 0.00014407116395886987}, {"step": 11650, "epoch": 29, "train_loss": 0.00017622964514885098}, {"step": 11700, "epoch": 29, "train_loss": 9.016267722472548e-05}, {"epoch_end": 29, "step": 11730, "test_acc": 0.9831, "test_loss": 0.06988456792831421}, {"step": 11750, "epoch": 30, "train_loss": 0.00019884758512489498}, {"step": 11800, "epoch": 30, "train_loss": 0.00011804633686551824}, {"step": 11850, "epoch": 30, "train_loss": 0.00010549449507379904}, {"step": 11900, "epoch": 30, "train_loss": 0.0002808020799420774}, {"step": 11950, "epoch": 30, "train_loss": 5.885396240046248e-05}, {"step": 12000, "epoch": 30, "train_loss": 0.00012905721087008715}, {"step": 12050, "epoch": 30, "train_loss": 0.0002965299354400486}, {"step": 12100, "epoch": 30, "train_loss": 0.00014367900439538062}, {"epoch_end": 30, "step": 12121, "test_acc": 0.9832, "test_loss": 0.06994562859535217}, {"step": 12150, "epoch": 31, "train_loss": 6.352635682560503e-05}, {"step": 12200, "epoch": 31, "train_loss": 0.00015131627151276916}, {"step": 12250, "epoch": 31, "train_loss": 0.000171226667589508}, {"step": 12300, "epoch": 31, "train_loss": 0.00018220422498416156}, {"step": 12350, "epoch": 31, "train_loss": 0.00015935255214571953}, {"step": 12400, "epoch": 31, "train_loss": 0.00011636834096862003}, {"step": 12450, "epoch": 31, "train_loss": 7.460927736246958e-05}, {"step": 12500, "epoch": 31, "train_loss": 8.722083293832839e-05}, {"epoch_end": 31, "step": 12512, "test_acc": 0.9828, "test_loss": 0.07018799428939819}, {"step": 12550, "epoch": 32, "train_loss": 0.00013992670574225485}, {"step": 12600, "epoch": 32, "train_loss": 5.382238668971695e-05}, {"step": 12650, "epoch": 32, "train_loss": 0.00014702172484248877}, {"step": 12700, "epoch": 32, "train_loss": 0.0001503830135334283}, {"step": 12750, "epoch": 32, "train_loss": 6.947641668375582e-05}, {"step": 12800, "epoch": 32, "train_loss": 0.00017082021804526448}, {"step": 12850, "epoch": 32, "train_loss": 6.839827256044373e-05}, {"step": 12900, "epoch": 32, "train_loss": 0.00015171620179899037}, {"epoch_end": 32, "step": 12903, "test_acc": 0.9831, "test_loss": 0.07034191145896912}, {"step": 12950, "epoch": 33, "train_loss": 9.353787027066574e-05}, {"step": 13000, "epoch": 33, "train_loss": 2.6175384846283123e-05}, {"step": 13050, "epoch": 33, "train_loss": 0.00010014730651164427}, {"step": 13100, "epoch": 33, "train_loss": 0.00010122975800186396}, {"step": 13150, "epoch": 33, "train_loss": 8.986737520899624e-05}, {"step": 13200, "epoch": 33, "train_loss": 0.00010119527723873034}, {"step": 13250, "epoch": 33, "train_loss": 0.0001268271735170856}, {"epoch_end": 33, "step": 13294, "test_acc": 0.9829, "test_loss": 0.07056150422096252}, {"step": 13300, "epoch": 34, "train_loss": 0.00010624496644595638}, {"step": 13350, "epoch": 34, "train_loss": 0.00010990109876729548}, {"step": 13400, "epoch": 34, "train_loss": 5.600217991741374e-05}, {"step": 13450, "epoch": 34, "train_loss": 0.00016930401034187526}, {"step": 13500, "epoch": 34, "train_loss": 0.00011057667870773003}, {"step": 13550, "epoch": 34, "train_loss": 0.00016492471331730485}, {"step": 13600, "epoch": 34, "train_loss": 7.369748345809057e-05}, {"step": 13650, "epoch": 34, "train_loss": 0.0001396914740325883}, {"epoch_end": 34, "step": 13685, "test_acc": 0.983, "test_loss": 0.07071907658576965}, {"step": 13700, "epoch": 35, "train_loss": 0.0001093342580134049}, {"step": 13750, "epoch": 35, "train_loss": 0.00010902685608016327}, {"step": 13800, "epoch": 35, "train_loss": 5.917489397688769e-05}, {"step": 13850, "epoch": 35, "train_loss": 0.00012219726340845227}, {"step": 13900, "epoch": 35, "train_loss": 7.114122126949951e-05}, {"step": 13950, "epoch": 35, "train_loss": 0.00012680963845923543}, {"step": 14000, "epoch": 35, "train_loss": 0.00010459933400852606}, {"step": 14050, "epoch": 35, "train_loss": 7.619381358381361e-05}, {"epoch_end": 35, "step": 14076, "test_acc": 0.9829, "test_loss": 0.07084052271842957}, {"step": 14100, "epoch": 36, "train_loss": 0.00010069071140605956}, {"step": 14150, "epoch": 36, "train_loss": 0.00010986677807522938}, {"step": 14200, "epoch": 36, "train_loss": 0.00011096067464677617}, {"step": 14250, "epoch": 36, "train_loss": 0.00011380804789951071}, {"step": 14300, "epoch": 36, "train_loss": 7.243134314194322e-05}, {"step": 14350, "epoch": 36, "train_loss": 7.661034032935277e-05}, {"step": 14400, "epoch": 36, "train_loss": 5.5071453971322626e-05}, {"step": 14450, "epoch": 36, "train_loss": 8.944510045694187e-05}, {"epoch_end": 36, "step": 14467, "test_acc": 0.983, "test_loss": 0.07114657344818115}, {"step": 14500, "epoch": 37, "train_loss": 0.000153565953951329}, {"step": 14550, "epoch": 37, "train_loss": 0.0001421974739059806}, {"step": 14600, "epoch": 37, "train_loss": 9.206886170431972e-05}, {"step": 14650, "epoch": 37, "train_loss": 0.00011665125202853233}, {"step": 14700, "epoch": 37, "train_loss": 0.00012141518527641892}, {"step": 14750, "epoch": 37, "train_loss": 7.977598579600453e-05}, {"step": 14800, "epoch": 37, "train_loss": 0.0001294601388508454}, {"step": 14850, "epoch": 37, "train_loss": 0.0001155207137344405}, {"epoch_end": 37, "step": 14858, "test_acc": 0.983, "test_loss": 0.0712730592250824}, {"step": 14900, "epoch": 38, "train_loss": 0.0001087886412278749}, {"step": 14950, "epoch": 38, "train_loss": 5.6642886193003505e-05}, {"step": 15000, "epoch": 38, "train_loss": 0.00011933179484913126}, {"step": 15050, "epoch": 38, "train_loss": 0.00010683265281841159}, {"step": 15100, "epoch": 38, "train_loss": 6.998535536695272e-05}, {"step": 15150, "epoch": 38, "train_loss": 0.00012313142360653728}, {"step": 15200, "epoch": 38, "train_loss": 0.0001282297307625413}, {"epoch_end": 38, "step": 15249, "test_acc": 0.983, "test_loss": 0.07151314754486084}, {"step": 15250, "epoch": 39, "train_loss": 4.036633617943153e-05}, {"step": 15300, "epoch": 39, "train_loss": 9.963897900888696e-05}, {"step": 15350, "epoch": 39, "train_loss": 0.00011684267519740388}, {"step": 15400, "epoch": 39, "train_loss": 6.338656385196373e-05}, {"step": 15450, "epoch": 39, "train_loss": 6.475069676525891e-05}, {"step": 15500, "epoch": 39, "train_loss": 0.00014432932948693633}, {"step": 15550, "epoch": 39, "train_loss": 9.879635035758838e-05}, {"step": 15600, "epoch": 39, "train_loss": 0.00012884431635029614}, {"epoch_end": 39, "step": 15640, "test_acc": 0.9831, "test_loss": 0.0714859881401062}, {"step": 15650, "epoch": 40, "train_loss": 0.00013576664787251502}, {"step": 15700, "epoch": 40, "train_loss": 9.61833429755643e-05}, {"step": 15750, "epoch": 40, "train_loss": 0.00012866563338320702}, {"step": 15800, "epoch": 40, "train_loss": 7.024908700259402e-05}, {"step": 15850, "epoch": 40, "train_loss": 4.849846664001234e-05}, {"step": 15900, "epoch": 40, "train_loss": 5.472193515743129e-05}, {"step": 15950, "epoch": 40, "train_loss": 8.315794548252597e-05}, {"step": 16000, "epoch": 40, "train_loss": 4.414144859765656e-05}, {"epoch_end": 40, "step": 16031, "test_acc": 0.9831, "test_loss": 0.0716530348777771}, {"step": 16050, "epoch": 41, "train_loss": 4.7670764615759254e-05}, {"step": 16100, "epoch": 41, "train_loss": 7.920087955426425e-05}, {"step": 16150, "epoch": 41, "train_loss": 9.031583613250405e-05}, {"step": 16200, "epoch": 41, "train_loss": 9.039277938427404e-05}, {"step": 16250, "epoch": 41, "train_loss": 9.72134293988347e-05}, {"step": 16300, "epoch": 41, "train_loss": 7.812162220943719e-05}, {"step": 16350, "epoch": 41, "train_loss": 8.778108167462051e-05}, {"step": 16400, "epoch": 41, "train_loss": 8.631501259515062e-05}, {"epoch_end": 41, "step": 16422, "test_acc": 0.9831, "test_loss": 0.07188808426856995}, {"step": 16450, "epoch": 42, "train_loss": 8.107461326289922e-05}, {"step": 16500, "epoch": 42, "train_loss": 8.039369276957586e-05}, {"step": 16550, "epoch": 42, "train_loss": 0.0001388849486829713}, {"step": 16600, "epoch": 42, "train_loss": 0.00012641250214073807}, {"step": 16650, "epoch": 42, "train_loss": 8.834349864628166e-05}, {"step": 16700, "epoch": 42, "train_loss": 0.00010126486449735239}, {"step": 16750, "epoch": 42, "train_loss": 6.706907151965424e-05}, {"step": 16800, "epoch": 42, "train_loss": 7.423426723107696e-05}, {"epoch_end": 42, "step": 16813, "test_acc": 0.983, "test_loss": 0.07195775680541992}, {"step": 16850, "epoch": 43, "train_loss": 0.0001795074640540406}, {"step": 16900, "epoch": 43, "train_loss": 5.1445204007904977e-05}, {"step": 16950, "epoch": 43, "train_loss": 0.00012657896149903536}, {"step": 17000, "epoch": 43, "train_loss": 7.780250598443672e-05}, {"step": 17050, "epoch": 43, "train_loss": 9.802974091144279e-05}, {"step": 17100, "epoch": 43, "train_loss": 7.749791257083416e-05}, {"step": 17150, "epoch": 43, "train_loss": 8.950717892730609e-05}, {"step": 17200, "epoch": 43, "train_loss": 9.735610365169123e-05}, {"epoch_end": 43, "step": 17204, "test_acc": 0.9831, "test_loss": 0.07209037795066833}, {"step": 17250, "epoch": 44, "train_loss": 0.00011830923904199153}, {"step": 17300, "epoch": 44, "train_loss": 8.741291821934283e-05}, {"step": 17350, "epoch": 44, "train_loss": 9.098831651499495e-05}, {"step": 17400, "epoch": 44, "train_loss": 6.396423850674182e-05}, {"step": 17450, "epoch": 44, "train_loss": 0.0001114930200856179}, {"step": 17500, "epoch": 44, "train_loss": 8.885953866411e-05}, {"step": 17550, "epoch": 44, "train_loss": 3.898189606843516e-05}, {"epoch_end": 44, "step": 17595, "test_acc": 0.983, "test_loss": 0.07221249461174012}, {"step": 17600, "epoch": 45, "train_loss": 5.3347950597526506e-05}, {"step": 17650, "epoch": 45, "train_loss": 3.632425796240568e-05}, {"step": 17700, "epoch": 45, "train_loss": 5.11020261910744e-05}, {"step": 17750, "epoch": 45, "train_loss": 0.00010304878378519788}, {"step": 17800, "epoch": 45, "train_loss": 0.00015219459601212293}, {"step": 17850, "epoch": 45, "train_loss": 3.2543157431064174e-05}, {"step": 17900, "epoch": 45, "train_loss": 0.00011294415890006348}, {"step": 17950, "epoch": 45, "train_loss": 9.968929953174666e-05}, {"epoch_end": 45, "step": 17986, "test_acc": 0.983, "test_loss": 0.07232235541343689}, {"step": 18000, "epoch": 46, "train_loss": 7.751960220048204e-05}, {"step": 18050, "epoch": 46, "train_loss": 8.611305383965373e-05}, {"step": 18100, "epoch": 46, "train_loss": 0.00014589772035833448}, {"step": 18150, "epoch": 46, "train_loss": 4.454571535461582e-05}, {"step": 18200, "epoch": 46, "train_loss": 7.942465890664607e-05}, {"step": 18250, "epoch": 46, "train_loss": 0.0001393747079418972}, {"step": 18300, "epoch": 46, "train_loss": 9.753733320394531e-05}, {"step": 18350, "epoch": 46, "train_loss": 8.12069047242403e-05}, {"epoch_end": 46, "step": 18377, "test_acc": 0.983, "test_loss": 0.07255132608413696}, {"step": 18400, "epoch": 47, "train_loss": 0.0001353989209746942}, {"step": 18450, "epoch": 47, "train_loss": 6.589489203179255e-05}, {"step": 18500, "epoch": 47, "train_loss": 8.964935841504484e-05}, {"step": 18550, "epoch": 47, "train_loss": 8.154760143952444e-05}, {"step": 18600, "epoch": 47, "train_loss": 6.491383828688413e-05}, {"step": 18650, "epoch": 47, "train_loss": 6.248216959647834e-05}, {"step": 18700, "epoch": 47, "train_loss": 0.00010665922309271991}, {"step": 18750, "epoch": 47, "train_loss": 4.791575702256523e-05}, {"epoch_end": 47, "step": 18768, "test_acc": 0.9829, "test_loss": 0.07266989817619324}, {"step": 18800, "epoch": 48, "train_loss": 7.327687490032986e-05}, {"step": 18850, "epoch": 48, "train_loss": 0.00012444551975931972}, {"step": 18900, "epoch": 48, "train_loss": 5.103684088680893e-05}, {"step": 18950, "epoch": 48, "train_loss": 0.00014394389290828258}, {"step": 19000, "epoch": 48, "train_loss": 5.3911880968371406e-05}, {"step": 19050, "epoch": 48, "train_loss": 0.0001294712274102494}, {"step": 19100, "epoch": 48, "train_loss": 0.00010471369751030579}, {"step": 19150, "epoch": 48, "train_loss": 5.676844739355147e-05}, {"epoch_end": 48, "step": 19159, "test_acc": 0.9829, "test_loss": 0.0727165587425232}, {"step": 19200, "epoch": 49, "train_loss": 6.513070547953248e-05}, {"step": 19250, "epoch": 49, "train_loss": 4.111141606699675e-05}, {"step": 19300, "epoch": 49, "train_loss": 0.00011344618542352691}, {"step": 19350, "epoch": 49, "train_loss": 8.001749665709212e-05}, {"step": 19400, "epoch": 49, "train_loss": 6.229235441423953e-05}, {"step": 19450, "epoch": 49, "train_loss": 9.008408960653469e-05}, {"step": 19500, "epoch": 49, "train_loss": 3.6851095501333475e-05}, {"epoch_end": 49, "step": 19550, "test_acc": 0.9833, "test_loss": 0.0728361783027649}, {"step": 19550, "epoch": 50, "train_loss": 5.213558324612677e-05}, {"step": 19600, "epoch": 50, "train_loss": 8.03028597147204e-05}, {"step": 19650, "epoch": 50, "train_loss": 3.676122287288308e-05}, {"step": 19700, "epoch": 50, "train_loss": 9.566806693328544e-05}, {"step": 19750, "epoch": 50, "train_loss": 0.00013600170495919883}, {"step": 19800, "epoch": 50, "train_loss": 9.93319190456532e-05}, {"step": 19850, "epoch": 50, "train_loss": 0.00010889917029999197}, {"step": 19900, "epoch": 50, "train_loss": 0.00010059120540972799}, {"epoch_end": 50, "step": 19941, "test_acc": 0.983, "test_loss": 0.07301612372398376}, {"step": 19950, "epoch": 51, "train_loss": 6.54758041491732e-05}, {"step": 20000, "epoch": 51, "train_loss": 6.973770359763876e-05}, {"step": 20050, "epoch": 51, "train_loss": 3.0496768886223435e-05}, {"step": 20100, "epoch": 51, "train_loss": 9.757674706634134e-05}, {"step": 20150, "epoch": 51, "train_loss": 7.136025669751689e-05}, {"step": 20200, "epoch": 51, "train_loss": 5.870054883416742e-05}, {"step": 20250, "epoch": 51, "train_loss": 7.976159395184368e-05}, {"step": 20300, "epoch": 51, "train_loss": 6.896039849380031e-05}, {"epoch_end": 51, "step": 20332, "test_acc": 0.9832, "test_loss": 0.0730734028339386}, {"step": 20350, "epoch": 52, "train_loss": 6.836825923528522e-05}, {"step": 20400, "epoch": 52, "train_loss": 9.268646681448445e-05}, {"step": 20450, "epoch": 52, "train_loss": 7.94601219240576e-05}, {"step": 20500, "epoch": 52, "train_loss": 8.638039435027167e-05}, {"step": 20550, "epoch": 52, "train_loss": 5.0112819735659286e-05}, {"step": 20600, "epoch": 52, "train_loss": 5.1197876018704847e-05}, {"step": 20650, "epoch": 52, "train_loss": 8.299294859170914e-05}, {"step": 20700, "epoch": 52, "train_loss": 9.504847548669204e-05}, {"epoch_end": 52, "step": 20723, "test_acc": 0.9831, "test_loss": 0.07319421372413636}, {"step": 20750, "epoch": 53, "train_loss": 3.985857620136812e-05}, {"step": 20800, "epoch": 53, "train_loss": 6.953693082323298e-05}, {"step": 20850, "epoch": 53, "train_loss": 0.00013133882021065801}, {"step": 20900, "epoch": 53, "train_loss": 8.728107786737382e-05}, {"step": 20950, "epoch": 53, "train_loss": 7.214489596663043e-05}, {"step": 21000, "epoch": 53, "train_loss": 6.35834876447916e-05}, {"step": 21050, "epoch": 53, "train_loss": 5.003211845178157e-05}, {"step": 21100, "epoch": 53, "train_loss": 6.582502828678116e-05}, {"epoch_end": 53, "step": 21114, "test_acc": 0.9831, "test_loss": 0.07329045658111573}, {"step": 21150, "epoch": 54, "train_loss": 4.991004607290961e-05}, {"step": 21200, "epoch": 54, "train_loss": 5.92486685491167e-05}, {"step": 21250, "epoch": 54, "train_loss": 4.427094972925261e-05}, {"step": 21300, "epoch": 54, "train_loss": 5.3950949222780764e-05}, {"step": 21350, "epoch": 54, "train_loss": 0.00010999064397765324}, {"step": 21400, "epoch": 54, "train_loss": 0.00012724676344078034}, {"step": 21450, "epoch": 54, "train_loss": 4.31414955528453e-05}, {"step": 21500, "epoch": 54, "train_loss": 5.212068936089054e-05}, {"epoch_end": 54, "step": 21505, "test_acc": 0.9831, "test_loss": 0.07339674081802368}, {"step": 21550, "epoch": 55, "train_loss": 6.26725668553263e-05}, {"step": 21600, "epoch": 55, "train_loss": 7.476429891539738e-05}, {"step": 21650, "epoch": 55, "train_loss": 5.6112803576979786e-05}, {"step": 21700, "epoch": 55, "train_loss": 7.954514876473695e-05}, {"step": 21750, "epoch": 55, "train_loss": 0.00010007397213485092}, {"step": 21800, "epoch": 55, "train_loss": 9.046608465723693e-05}, {"step": 21850, "epoch": 55, "train_loss": 5.075294029666111e-05}, {"epoch_end": 55, "step": 21896, "test_acc": 0.983, "test_loss": 0.07352954516410827}, {"step": 21900, "epoch": 56, "train_loss": 4.9636364565230906e-05}, {"step": 21950, "epoch": 56, "train_loss": 0.00011145067401230335}, {"step": 22000, "epoch": 56, "train_loss": 6.532304541906342e-05}, {"step": 22050, "epoch": 56, "train_loss": 3.4907046938315034e-05}, {"step": 22100, "epoch": 56, "train_loss": 5.872706606169231e-05}, {"step": 22150, "epoch": 56, "train_loss": 5.879162563360296e-05}, {"step": 22200, "epoch": 56, "train_loss": 0.00011730416008504108}, {"step": 22250, "epoch": 56, "train_loss": 9.803065040614456e-05}, {"epoch_end": 56, "step": 22287, "test_acc": 0.9832, "test_loss": 0.07353231401443482}, {"step": 22300, "epoch": 57, "train_loss": 9.976865840144455e-05}, {"step": 22350, "epoch": 57, "train_loss": 9.15935161174275e-05}, {"step": 22400, "epoch": 57, "train_loss": 9.440835128771141e-05}, {"step": 22450, "epoch": 57, "train_loss": 3.2459018257213756e-05}, {"step": 22500, "epoch": 57, "train_loss": 4.049080234835856e-05}, {"step": 22550, "epoch": 57, "train_loss": 9.888019849313423e-05}, {"step": 22600, "epoch": 57, "train_loss": 4.749724030261859e-05}, {"step": 22650, "epoch": 57, "train_loss": 8.564208837924525e-05}, {"epoch_end": 57, "step": 22678, "test_acc": 0.983, "test_loss": 0.07367653722763061}, {"step": 22700, "epoch": 58, "train_loss": 4.274461025488563e-05}, {"step": 22750, "epoch": 58, "train_loss": 0.00010485320672160015}, {"step": 22800, "epoch": 58, "train_loss": 7.989434379851446e-05}, {"step": 22850, "epoch": 58, "train_loss": 6.236629269551486e-05}, {"step": 22900, "epoch": 58, "train_loss": 4.125941632082686e-05}, {"step": 22950, "epoch": 58, "train_loss": 6.938726437510923e-05}, {"step": 23000, "epoch": 58, "train_loss": 5.062501804786734e-05}, {"step": 23050, "epoch": 58, "train_loss": 6.874505197629333e-05}, {"epoch_end": 58, "step": 23069, "test_acc": 0.983, "test_loss": 0.07377329449653626}, {"step": 23100, "epoch": 59, "train_loss": 5.8147506933892146e-05}, {"step": 23150, "epoch": 59, "train_loss": 4.709028507932089e-05}, {"step": 23200, "epoch": 59, "train_loss": 2.0986286472179927e-05}, {"step": 23250, "epoch": 59, "train_loss": 7.304285827558488e-05}, {"step": 23300, "epoch": 59, "train_loss": 5.322141805663705e-05}, {"step": 23350, "epoch": 59, "train_loss": 5.1514431106625125e-05}, {"step": 23400, "epoch": 59, "train_loss": 0.0001334085245616734}, {"step": 23450, "epoch": 59, "train_loss": 5.262482591206208e-05}, {"epoch_end": 59, "step": 23460, "test_acc": 0.983, "test_loss": 0.073947585105896}], "final": {"test_acc": 0.983, "test_loss": 0.073947585105896, "wall_s": 46.32193732261658}}