{ "vanilla": { "2": [ { "epoch": 1, "train_loss": 8.784176065673828, "train_acc": 0.29394, "val_loss": 3.5842281635284423, "val_acc": 0.3873, "lyapunov": null, "grad_norm": 10.000000009356935, "time_sec": 68.70752310752869 }, { "epoch": 2, "train_loss": 3.0238943502426148, "train_acc": 0.36252, "val_loss": 1.8549414070129395, "val_acc": 0.4613, "lyapunov": null, "grad_norm": 9.999999721946596, "time_sec": 66.72225046157837 }, { "epoch": 3, "train_loss": 1.6792986631774902, "train_acc": 0.43042, "val_loss": 1.3899815103530884, "val_acc": 0.5099, "lyapunov": null, "grad_norm": 9.479727781022204, "time_sec": 66.72804069519043 }, { "epoch": 4, "train_loss": 1.3923843894577026, "train_acc": 0.50764, "val_loss": 1.4540944946289063, "val_acc": 0.5021, "lyapunov": null, "grad_norm": 5.585987134055558, "time_sec": 66.7354838848114 }, { "epoch": 5, "train_loss": 1.3035008081817627, "train_acc": 0.53676, "val_loss": 1.19722311668396, "val_acc": 0.5746, "lyapunov": null, "grad_norm": 4.536164010734392, "time_sec": 66.73390054702759 }, { "epoch": 6, "train_loss": 1.2413602298355102, "train_acc": 0.55996, "val_loss": 1.2747809772491454, "val_acc": 0.5511, "lyapunov": null, "grad_norm": 3.7551640244188893, "time_sec": 66.7424738407135 }, { "epoch": 7, "train_loss": 1.1952919603729248, "train_acc": 0.57892, "val_loss": 1.3223521461486816, "val_acc": 0.5765, "lyapunov": null, "grad_norm": 3.2619059033032882, "time_sec": 66.73422837257385 }, { "epoch": 8, "train_loss": 1.155884972076416, "train_acc": 0.59154, "val_loss": 1.1085322700500488, "val_acc": 0.6163, "lyapunov": null, "grad_norm": 2.9020145937231683, "time_sec": 66.7125391960144 }, { "epoch": 9, "train_loss": 1.1166249238204955, "train_acc": 0.60418, "val_loss": 1.1168422752380371, "val_acc": 0.6199, "lyapunov": null, "grad_norm": 2.543863279566723, "time_sec": 66.71027660369873 }, { "epoch": 10, "train_loss": 1.0951375158691405, "train_acc": 0.61542, "val_loss": 1.0904820528030394, "val_acc": 0.6198, "lyapunov": null, "grad_norm": 2.3668249851916006, "time_sec": 66.71985030174255 }, { "epoch": 11, "train_loss": 1.069790382232666, "train_acc": 0.62292, "val_loss": 1.0318323984146118, "val_acc": 0.6373, "lyapunov": null, "grad_norm": 2.1919738547311467, "time_sec": 66.71968007087708 }, { "epoch": 12, "train_loss": 1.0416536513900756, "train_acc": 0.63312, "val_loss": 0.9504183568954467, "val_acc": 0.6644, "lyapunov": null, "grad_norm": 2.0726643927670843, "time_sec": 66.72264409065247 }, { "epoch": 13, "train_loss": 1.0279294053649903, "train_acc": 0.63916, "val_loss": 0.9726510414123535, "val_acc": 0.6598, "lyapunov": null, "grad_norm": 1.9937883030215755, "time_sec": 66.73272228240967 }, { "epoch": 14, "train_loss": 1.0074240830993653, "train_acc": 0.64654, "val_loss": 1.0757453815460205, "val_acc": 0.6355, "lyapunov": null, "grad_norm": 1.9051972840134976, "time_sec": 66.78505396842957 }, { "epoch": 15, "train_loss": 0.9958204155349731, "train_acc": 0.65224, "val_loss": 0.9282746883392334, "val_acc": 0.6789, "lyapunov": null, "grad_norm": 1.8406010134738577, "time_sec": 66.74679851531982 }, { "epoch": 16, "train_loss": 0.9811411659431457, "train_acc": 0.6568, "val_loss": 0.948696481513977, "val_acc": 0.6773, "lyapunov": null, "grad_norm": 1.8009035952162606, "time_sec": 66.76298546791077 }, { "epoch": 17, "train_loss": 0.9629479264068603, "train_acc": 0.66368, "val_loss": 0.9225332054138183, "val_acc": 0.6785, "lyapunov": null, "grad_norm": 1.729877169103032, "time_sec": 66.76885318756104 }, { "epoch": 18, "train_loss": 0.9539957814788819, "train_acc": 0.66876, "val_loss": 0.8840870971679687, "val_acc": 0.6934, "lyapunov": null, "grad_norm": 1.6975645226541463, "time_sec": 66.76839065551758 }, { "epoch": 19, "train_loss": 0.9348216704177856, "train_acc": 0.67396, "val_loss": 0.8788524172782898, "val_acc": 0.6974, "lyapunov": null, "grad_norm": 1.6623387209393465, "time_sec": 66.75536942481995 }, { "epoch": 20, "train_loss": 0.9298514527893066, "train_acc": 0.67744, "val_loss": 0.8792482821464539, "val_acc": 0.6879, "lyapunov": null, "grad_norm": 1.6514262668135486, "time_sec": 66.77851033210754 }, { "epoch": 21, "train_loss": 0.9187093223381042, "train_acc": 0.68172, "val_loss": 0.8565962639808655, "val_acc": 0.7052, "lyapunov": null, "grad_norm": 1.601461439120708, "time_sec": 66.76159858703613 }, { "epoch": 22, "train_loss": 0.9118761095809936, "train_acc": 0.6807, "val_loss": 0.8546041209220886, "val_acc": 0.6999, "lyapunov": null, "grad_norm": 1.5836794527231723, "time_sec": 66.78451776504517 }, { "epoch": 23, "train_loss": 0.9017199424552917, "train_acc": 0.68394, "val_loss": 0.8476145436286926, "val_acc": 0.7075, "lyapunov": null, "grad_norm": 1.5756083275546304, "time_sec": 66.76763606071472 }, { "epoch": 24, "train_loss": 0.8852598934555054, "train_acc": 0.69118, "val_loss": 0.8724938739776611, "val_acc": 0.695, "lyapunov": null, "grad_norm": 1.5497954712978341, "time_sec": 66.74203777313232 }, { "epoch": 25, "train_loss": 0.8779828417778015, "train_acc": 0.69466, "val_loss": 0.8507094660758973, "val_acc": 0.7084, "lyapunov": null, "grad_norm": 1.5587226233593017, "time_sec": 66.76736617088318 }, { "epoch": 26, "train_loss": 0.8729706660270691, "train_acc": 0.69588, "val_loss": 0.8783558692932129, "val_acc": 0.7017, "lyapunov": null, "grad_norm": 1.558806303739579, "time_sec": 66.78415656089783 }, { "epoch": 27, "train_loss": 0.8616628527641297, "train_acc": 0.69924, "val_loss": 0.8321251944541931, "val_acc": 0.7131, "lyapunov": null, "grad_norm": 1.5155312617664398, "time_sec": 66.76412272453308 }, { "epoch": 28, "train_loss": 0.853893226184845, "train_acc": 0.70296, "val_loss": 0.8290677871704102, "val_acc": 0.7202, "lyapunov": null, "grad_norm": 1.5334745326254058, "time_sec": 66.77086138725281 }, { "epoch": 29, "train_loss": 0.8483871575927734, "train_acc": 0.7044, "val_loss": 0.8194198965072632, "val_acc": 0.7177, "lyapunov": null, "grad_norm": 1.516906571895531, "time_sec": 66.76542353630066 }, { "epoch": 30, "train_loss": 0.8437307675552368, "train_acc": 0.70606, "val_loss": 0.807693408203125, "val_acc": 0.7226, "lyapunov": null, "grad_norm": 1.5177335224644324, "time_sec": 66.76670408248901 }, { "epoch": 31, "train_loss": 0.8290397611427307, "train_acc": 0.71108, "val_loss": 0.7662792547225952, "val_acc": 0.7356, "lyapunov": null, "grad_norm": 1.5206721972729154, "time_sec": 66.76865673065186 }, { "epoch": 32, "train_loss": 0.826131916179657, "train_acc": 0.71084, "val_loss": 0.7779187292098999, "val_acc": 0.7315, "lyapunov": null, "grad_norm": 1.5168697178312431, "time_sec": 66.78137564659119 }, { "epoch": 33, "train_loss": 0.8184985132217407, "train_acc": 0.71504, "val_loss": 0.7998169664382935, "val_acc": 0.7328, "lyapunov": null, "grad_norm": 1.521911359127393, "time_sec": 66.75752782821655 }, { "epoch": 34, "train_loss": 0.8095653260612488, "train_acc": 0.71712, "val_loss": 0.7914467765808105, "val_acc": 0.7279, "lyapunov": null, "grad_norm": 1.5215833614736305, "time_sec": 66.78136897087097 }, { "epoch": 35, "train_loss": 0.804902925567627, "train_acc": 0.71986, "val_loss": 0.7923053074836731, "val_acc": 0.7298, "lyapunov": null, "grad_norm": 1.5097172654341793, "time_sec": 66.77621006965637 }, { "epoch": 36, "train_loss": 0.7982923312759399, "train_acc": 0.72278, "val_loss": 0.7828660634040833, "val_acc": 0.7347, "lyapunov": null, "grad_norm": 1.5080874470718673, "time_sec": 66.78457188606262 }, { "epoch": 37, "train_loss": 0.8005728244400024, "train_acc": 0.71974, "val_loss": 0.7977071420669556, "val_acc": 0.727, "lyapunov": null, "grad_norm": 1.520762564155578, "time_sec": 66.7710018157959 }, { "epoch": 38, "train_loss": 0.7921301879501342, "train_acc": 0.7257, "val_loss": 0.7928903070449829, "val_acc": 0.7314, "lyapunov": null, "grad_norm": 1.5220549579996931, "time_sec": 66.771799325943 }, { "epoch": 39, "train_loss": 0.7908940573310852, "train_acc": 0.72564, "val_loss": 0.7704643251419068, "val_acc": 0.7389, "lyapunov": null, "grad_norm": 1.5257030534456404, "time_sec": 66.79677128791809 }, { "epoch": 40, "train_loss": 0.782756520690918, "train_acc": 0.72798, "val_loss": 0.7587965223312378, "val_acc": 0.7382, "lyapunov": null, "grad_norm": 1.5173101860758342, "time_sec": 66.7534248828888 }, { "epoch": 41, "train_loss": 0.7764551944160462, "train_acc": 0.7275, "val_loss": 0.7702336301803588, "val_acc": 0.7376, "lyapunov": null, "grad_norm": 1.5233512089255186, "time_sec": 66.72562265396118 }, { "epoch": 42, "train_loss": 0.7770634262084961, "train_acc": 0.73002, "val_loss": 0.79198386926651, "val_acc": 0.7352, "lyapunov": null, "grad_norm": 1.5256504913621811, "time_sec": 66.72624778747559 }, { "epoch": 43, "train_loss": 0.7760492914009094, "train_acc": 0.72876, "val_loss": 0.7571331614494323, "val_acc": 0.7367, "lyapunov": null, "grad_norm": 1.5305845247584131, "time_sec": 66.77268481254578 }, { "epoch": 44, "train_loss": 0.7731727365493775, "train_acc": 0.73148, "val_loss": 0.7542887442588806, "val_acc": 0.7415, "lyapunov": null, "grad_norm": 1.5154529270586274, "time_sec": 66.78488183021545 }, { "epoch": 45, "train_loss": 0.7739742170715332, "train_acc": 0.72984, "val_loss": 0.7391758235931396, "val_acc": 0.7498, "lyapunov": null, "grad_norm": 1.5251558940731724, "time_sec": 66.7819311618805 }, { "epoch": 46, "train_loss": 0.7671850444030762, "train_acc": 0.73264, "val_loss": 0.7574850060462952, "val_acc": 0.7393, "lyapunov": null, "grad_norm": 1.5170986261109833, "time_sec": 66.7733941078186 }, { "epoch": 47, "train_loss": 0.7750339817810059, "train_acc": 0.73066, "val_loss": 0.741615094947815, "val_acc": 0.7478, "lyapunov": null, "grad_norm": 1.526403632428659, "time_sec": 66.79545855522156 }, { "epoch": 48, "train_loss": 0.7732410961341858, "train_acc": 0.73054, "val_loss": 0.7477504265785218, "val_acc": 0.7463, "lyapunov": null, "grad_norm": 1.5252023193581805, "time_sec": 66.76390218734741 }, { "epoch": 49, "train_loss": 0.7645146517944336, "train_acc": 0.7362, "val_loss": 0.7431186792373657, "val_acc": 0.7469, "lyapunov": null, "grad_norm": 1.5124602849840472, "time_sec": 66.77198100090027 }, { "epoch": 50, "train_loss": 0.7630206244087219, "train_acc": 0.73456, "val_loss": 0.7702469422340393, "val_acc": 0.7372, "lyapunov": null, "grad_norm": 1.5111717198170513, "time_sec": 66.78833556175232 } ], "3": [ { "epoch": 1, "train_loss": 7.879869033813477, "train_acc": 0.29274, "val_loss": 3.8848869758605957, "val_acc": 0.4307, "lyapunov": null, "grad_norm": 10.000000052723436, "time_sec": 81.73621559143066 }, { "epoch": 2, "train_loss": 3.6673114225769043, "train_acc": 0.37732, "val_loss": 2.4374912944793703, "val_acc": 0.4641, "lyapunov": null, "grad_norm": 9.999999992064463, "time_sec": 81.64153814315796 }, { "epoch": 3, "train_loss": 2.6689771520996093, "train_acc": 0.44886, "val_loss": 2.1643002433776855, "val_acc": 0.4783, "lyapunov": null, "grad_norm": 9.99999958129824, "time_sec": 81.61646294593811 }, { "epoch": 4, "train_loss": 2.2089931160736085, "train_acc": 0.49876, "val_loss": 1.6028311902999879, "val_acc": 0.5766, "lyapunov": null, "grad_norm": 9.9479461102616, "time_sec": 81.63859295845032 }, { "epoch": 5, "train_loss": 1.3628238194465636, "train_acc": 0.56316, "val_loss": 1.166467163848877, "val_acc": 0.6153, "lyapunov": null, "grad_norm": 5.151865083183334, "time_sec": 81.61932921409607 }, { "epoch": 6, "train_loss": 1.094777367324829, "train_acc": 0.61794, "val_loss": 0.9767357675552368, "val_acc": 0.6648, "lyapunov": null, "grad_norm": 2.9141559842708555, "time_sec": 81.59153366088867 }, { "epoch": 7, "train_loss": 1.0235227158927918, "train_acc": 0.64222, "val_loss": 0.9529552062988281, "val_acc": 0.6697, "lyapunov": null, "grad_norm": 2.7307826559078365, "time_sec": 81.59089064598083 }, { "epoch": 8, "train_loss": 0.9709269950866699, "train_acc": 0.66256, "val_loss": 0.8724164709091187, "val_acc": 0.6933, "lyapunov": null, "grad_norm": 2.534221184138155, "time_sec": 81.58447670936584 }, { "epoch": 9, "train_loss": 0.9305636987686158, "train_acc": 0.67854, "val_loss": 0.9464693843841553, "val_acc": 0.6871, "lyapunov": null, "grad_norm": 2.3845503111469237, "time_sec": 81.55373859405518 }, { "epoch": 10, "train_loss": 0.884402558479309, "train_acc": 0.69342, "val_loss": 0.8181013963699341, "val_acc": 0.7151, "lyapunov": null, "grad_norm": 2.2558435355013287, "time_sec": 81.59211921691895 }, { "epoch": 11, "train_loss": 0.8623861204528809, "train_acc": 0.70352, "val_loss": 0.8547475276947022, "val_acc": 0.704, "lyapunov": null, "grad_norm": 2.1392332924582, "time_sec": 96.83045625686646 }, { "epoch": 12, "train_loss": 0.8302078694152832, "train_acc": 0.7144, "val_loss": 0.8837014081954956, "val_acc": 0.7009, "lyapunov": null, "grad_norm": 2.037610695085551, "time_sec": 81.57945966720581 }, { "epoch": 13, "train_loss": 0.8107185283660888, "train_acc": 0.71764, "val_loss": 0.8271433633804322, "val_acc": 0.7234, "lyapunov": null, "grad_norm": 1.9584004792184107, "time_sec": 91.92489910125732 }, { "epoch": 14, "train_loss": 0.7858210314559937, "train_acc": 0.7282, "val_loss": 0.7926485777854919, "val_acc": 0.7343, "lyapunov": null, "grad_norm": 1.8721183071247673, "time_sec": 81.58737134933472 }, { "epoch": 15, "train_loss": 0.7662513049697876, "train_acc": 0.7337, "val_loss": 0.8291977769851685, "val_acc": 0.7199, "lyapunov": null, "grad_norm": 1.7853534785130616, "time_sec": 81.57148289680481 }, { "epoch": 16, "train_loss": 0.7458733101463317, "train_acc": 0.74194, "val_loss": 0.8315077615737915, "val_acc": 0.7304, "lyapunov": null, "grad_norm": 1.7402805012441789, "time_sec": 91.85386824607849 }, { "epoch": 17, "train_loss": 0.7265361336135864, "train_acc": 0.7477, "val_loss": 0.8167202463150024, "val_acc": 0.7255, "lyapunov": null, "grad_norm": 1.7008211718221242, "time_sec": 81.58936834335327 }, { "epoch": 18, "train_loss": 0.7110094029045105, "train_acc": 0.7522, "val_loss": 0.7371786734580994, "val_acc": 0.7536, "lyapunov": null, "grad_norm": 1.6336862727687305, "time_sec": 81.58748602867126 }, { "epoch": 19, "train_loss": 0.6943154872894287, "train_acc": 0.75848, "val_loss": 0.8288676969528198, "val_acc": 0.7457, "lyapunov": null, "grad_norm": 1.5772527704617827, "time_sec": 81.58329463005066 }, { "epoch": 20, "train_loss": 0.6799315312194825, "train_acc": 0.76406, "val_loss": 0.7031201959133149, "val_acc": 0.7597, "lyapunov": null, "grad_norm": 1.5409222106616058, "time_sec": 81.59869694709778 }, { "epoch": 21, "train_loss": 0.6609970189476013, "train_acc": 0.76998, "val_loss": 0.7172592522621155, "val_acc": 0.757, "lyapunov": null, "grad_norm": 1.5146675171159136, "time_sec": 81.62905097007751 }, { "epoch": 22, "train_loss": 0.6413579912948608, "train_acc": 0.77642, "val_loss": 0.7587738120079041, "val_acc": 0.7564, "lyapunov": null, "grad_norm": 1.4845376317814207, "time_sec": 81.5891945362091 }, { "epoch": 23, "train_loss": 0.6361218054008484, "train_acc": 0.7808, "val_loss": 0.8577088802337647, "val_acc": 0.7302, "lyapunov": null, "grad_norm": 1.4583042056531952, "time_sec": 81.59341359138489 }, { "epoch": 24, "train_loss": 0.6177744509887695, "train_acc": 0.78602, "val_loss": 0.7803928444862366, "val_acc": 0.7525, "lyapunov": null, "grad_norm": 1.436682953672717, "time_sec": 81.62475347518921 }, { "epoch": 25, "train_loss": 0.607811340675354, "train_acc": 0.78966, "val_loss": 0.6921211493968964, "val_acc": 0.7711, "lyapunov": null, "grad_norm": 1.418159000686253, "time_sec": 81.62535405158997 }, { "epoch": 26, "train_loss": 0.59512930103302, "train_acc": 0.7951, "val_loss": 0.684100229549408, "val_acc": 0.7722, "lyapunov": null, "grad_norm": 1.3951606174178173, "time_sec": 81.63963985443115 }, { "epoch": 27, "train_loss": 0.5878567532730102, "train_acc": 0.79642, "val_loss": 0.7061413002967835, "val_acc": 0.7659, "lyapunov": null, "grad_norm": 1.3868247719185465, "time_sec": 81.6419289112091 }, { "epoch": 28, "train_loss": 0.5767165324783325, "train_acc": 0.80024, "val_loss": 0.7104957416534424, "val_acc": 0.7604, "lyapunov": null, "grad_norm": 1.3709386254089717, "time_sec": 81.66154742240906 }, { "epoch": 29, "train_loss": 0.565018659324646, "train_acc": 0.80406, "val_loss": 0.768200507259369, "val_acc": 0.7553, "lyapunov": null, "grad_norm": 1.3719882092582847, "time_sec": 81.62237930297852 }, { "epoch": 30, "train_loss": 0.5537373196792602, "train_acc": 0.80818, "val_loss": 0.7797956996917724, "val_acc": 0.754, "lyapunov": null, "grad_norm": 1.356635299189587, "time_sec": 81.6416084766388 }, { "epoch": 31, "train_loss": 0.5465290784263611, "train_acc": 0.80986, "val_loss": 0.8261515438079834, "val_acc": 0.7601, "lyapunov": null, "grad_norm": 1.3578924252088589, "time_sec": 81.63413286209106 }, { "epoch": 32, "train_loss": 0.5376561485385895, "train_acc": 0.81274, "val_loss": 0.7106016860961915, "val_acc": 0.7706, "lyapunov": null, "grad_norm": 1.3438672538902305, "time_sec": 81.63174223899841 }, { "epoch": 33, "train_loss": 0.5286657208347321, "train_acc": 0.81682, "val_loss": 0.7933181698799133, "val_acc": 0.7526, "lyapunov": null, "grad_norm": 1.3347423291514993, "time_sec": 81.64876651763916 }, { "epoch": 34, "train_loss": 0.515528777332306, "train_acc": 0.81998, "val_loss": 0.667878624010086, "val_acc": 0.7842, "lyapunov": null, "grad_norm": 1.3434716571373506, "time_sec": 81.63451099395752 }, { "epoch": 35, "train_loss": 0.517482133140564, "train_acc": 0.81894, "val_loss": 0.6637719221115113, "val_acc": 0.7788, "lyapunov": null, "grad_norm": 1.348328258826793, "time_sec": 81.64954471588135 }, { "epoch": 36, "train_loss": 0.5085288023757935, "train_acc": 0.82316, "val_loss": 0.6779749327659607, "val_acc": 0.7728, "lyapunov": null, "grad_norm": 1.3379118748369725, "time_sec": 81.63437128067017 }, { "epoch": 37, "train_loss": 0.5045176992702484, "train_acc": 0.82326, "val_loss": 0.7443327487945557, "val_acc": 0.7734, "lyapunov": null, "grad_norm": 1.351154323985601, "time_sec": 81.63439774513245 }, { "epoch": 38, "train_loss": 0.4942554948616028, "train_acc": 0.82914, "val_loss": 0.6845738918304444, "val_acc": 0.7778, "lyapunov": null, "grad_norm": 1.334483387470237, "time_sec": 81.64469575881958 }, { "epoch": 39, "train_loss": 0.4848251422119141, "train_acc": 0.8305, "val_loss": 0.7909360774993897, "val_acc": 0.7756, "lyapunov": null, "grad_norm": 1.3513076153037455, "time_sec": 81.90468525886536 }, { "epoch": 40, "train_loss": 0.4794950799369812, "train_acc": 0.83156, "val_loss": 0.6573793635368347, "val_acc": 0.7846, "lyapunov": null, "grad_norm": 1.3491734977607501, "time_sec": 81.66782331466675 }, { "epoch": 41, "train_loss": 0.4830877388286591, "train_acc": 0.83098, "val_loss": 0.7192189591407776, "val_acc": 0.7868, "lyapunov": null, "grad_norm": 1.3410571730739185, "time_sec": 81.6331217288971 }, { "epoch": 42, "train_loss": 0.47120577632904054, "train_acc": 0.83454, "val_loss": 0.7919704052448273, "val_acc": 0.7799, "lyapunov": null, "grad_norm": 1.3490152792196894, "time_sec": 81.62392020225525 }, { "epoch": 43, "train_loss": 0.47087046663284304, "train_acc": 0.8356, "val_loss": 0.7325520120620728, "val_acc": 0.775, "lyapunov": null, "grad_norm": 1.3477426194261104, "time_sec": 81.62399625778198 }, { "epoch": 44, "train_loss": 0.4646198981285095, "train_acc": 0.83768, "val_loss": 0.7522110661506652, "val_acc": 0.7794, "lyapunov": null, "grad_norm": 1.3522801846293095, "time_sec": 81.64305329322815 }, { "epoch": 45, "train_loss": 0.46615973456382753, "train_acc": 0.83716, "val_loss": 0.7075083548545837, "val_acc": 0.774, "lyapunov": null, "grad_norm": 1.3443254685283252, "time_sec": 81.6245231628418 }, { "epoch": 46, "train_loss": 0.464144686756134, "train_acc": 0.8373, "val_loss": 0.7641978782653809, "val_acc": 0.777, "lyapunov": null, "grad_norm": 1.3623767913048563, "time_sec": 81.63066077232361 }, { "epoch": 47, "train_loss": 0.45797264764785767, "train_acc": 0.83916, "val_loss": 0.6343508301258087, "val_acc": 0.7964, "lyapunov": null, "grad_norm": 1.3537053580131782, "time_sec": 81.63357996940613 }, { "epoch": 48, "train_loss": 0.45760699135780336, "train_acc": 0.8403, "val_loss": 0.7436078882217407, "val_acc": 0.7753, "lyapunov": null, "grad_norm": 1.3440105512051375, "time_sec": 81.65772676467896 }, { "epoch": 49, "train_loss": 0.4605378571224213, "train_acc": 0.83872, "val_loss": 0.724734008216858, "val_acc": 0.7765, "lyapunov": null, "grad_norm": 1.3562901018792388, "time_sec": 81.64547538757324 }, { "epoch": 50, "train_loss": 0.4620614657974243, "train_acc": 0.837, "val_loss": 0.6755017209529877, "val_acc": 0.7772, "lyapunov": null, "grad_norm": 1.3537578041931293, "time_sec": 81.64218926429749 } ], "4": [ { "epoch": 1, "train_loss": 6.681223518676758, "train_acc": 0.29266, "val_loss": 2.6236146224975587, "val_acc": 0.4208, "lyapunov": null, "grad_norm": 10.000000017509308, "time_sec": 91.20688438415527 }, { "epoch": 2, "train_loss": 2.939238514022827, "train_acc": 0.40644, "val_loss": 1.606919109725952, "val_acc": 0.5216, "lyapunov": null, "grad_norm": 9.999999716886606, "time_sec": 91.1075644493103 }, { "epoch": 3, "train_loss": 2.0392279933166506, "train_acc": 0.49038, "val_loss": 3.3730864547729493, "val_acc": 0.4914, "lyapunov": null, "grad_norm": 9.991851410151469, "time_sec": 91.08924913406372 }, { "epoch": 4, "train_loss": 1.6651515549850464, "train_acc": 0.54342, "val_loss": 1.346921759414673, "val_acc": 0.5907, "lyapunov": null, "grad_norm": 9.753475623030397, "time_sec": 91.15958094596863 }, { "epoch": 5, "train_loss": 1.392671317062378, "train_acc": 0.59812, "val_loss": 1.1599563423156738, "val_acc": 0.6353, "lyapunov": null, "grad_norm": 9.33276288903245, "time_sec": 91.0996298789978 }, { "epoch": 6, "train_loss": 1.2004287784957886, "train_acc": 0.63726, "val_loss": 1.3996731214523315, "val_acc": 0.6484, "lyapunov": null, "grad_norm": 8.337115001778015, "time_sec": 91.17090177536011 }, { "epoch": 7, "train_loss": 1.0604095679855348, "train_acc": 0.67054, "val_loss": 1.048226024055481, "val_acc": 0.6572, "lyapunov": null, "grad_norm": 7.413095027011666, "time_sec": 91.07081604003906 }, { "epoch": 8, "train_loss": 0.964676478729248, "train_acc": 0.6958, "val_loss": 0.971341968536377, "val_acc": 0.6833, "lyapunov": null, "grad_norm": 6.488820725452505, "time_sec": 91.15018796920776 }, { "epoch": 9, "train_loss": 0.9037447152328492, "train_acc": 0.71368, "val_loss": 0.9606797493934631, "val_acc": 0.6884, "lyapunov": null, "grad_norm": 6.004556556629876, "time_sec": 91.08957529067993 }, { "epoch": 10, "train_loss": 0.8311930680656433, "train_acc": 0.73054, "val_loss": 0.9408022530555725, "val_acc": 0.6857, "lyapunov": null, "grad_norm": 5.224294704950926, "time_sec": 91.09752058982849 }, { "epoch": 11, "train_loss": 0.7883032609176636, "train_acc": 0.74202, "val_loss": 0.7735380174636841, "val_acc": 0.7296, "lyapunov": null, "grad_norm": 4.817434485138619, "time_sec": 91.16273784637451 }, { "epoch": 12, "train_loss": 0.7438191238212586, "train_acc": 0.75504, "val_loss": 0.8446781181335449, "val_acc": 0.7183, "lyapunov": null, "grad_norm": 4.347280256854771, "time_sec": 91.09070873260498 }, { "epoch": 13, "train_loss": 0.711563241481781, "train_acc": 0.7632, "val_loss": 1.0206061399459838, "val_acc": 0.7067, "lyapunov": null, "grad_norm": 4.041266147302847, "time_sec": 91.15795159339905 }, { "epoch": 14, "train_loss": 0.6803020019721985, "train_acc": 0.77474, "val_loss": 0.9191598834037781, "val_acc": 0.7231, "lyapunov": null, "grad_norm": 3.8148137325221243, "time_sec": 91.06504201889038 }, { "epoch": 15, "train_loss": 0.655299462184906, "train_acc": 0.7814, "val_loss": 0.7831028885364533, "val_acc": 0.7484, "lyapunov": null, "grad_norm": 3.732384774455792, "time_sec": 91.05725073814392 }, { "epoch": 16, "train_loss": 0.6317539101028442, "train_acc": 0.7872, "val_loss": 0.984422788143158, "val_acc": 0.7296, "lyapunov": null, "grad_norm": 3.462411900383296, "time_sec": 91.17561101913452 }, { "epoch": 17, "train_loss": 0.6076911881351471, "train_acc": 0.79506, "val_loss": 1.06318473238945, "val_acc": 0.6942, "lyapunov": null, "grad_norm": 3.287738199110906, "time_sec": 91.05370211601257 }, { "epoch": 18, "train_loss": 0.5807661696720123, "train_acc": 0.80182, "val_loss": 0.8986036172389984, "val_acc": 0.7155, "lyapunov": null, "grad_norm": 2.9369074379710294, "time_sec": 91.11013340950012 }, { "epoch": 19, "train_loss": 0.5573896813774109, "train_acc": 0.80848, "val_loss": 0.7345395576477051, "val_acc": 0.7591, "lyapunov": null, "grad_norm": 2.813763525156014, "time_sec": 91.06488347053528 }, { "epoch": 20, "train_loss": 0.5408507973098755, "train_acc": 0.81528, "val_loss": 0.9427923340797424, "val_acc": 0.7514, "lyapunov": null, "grad_norm": 2.666313690069219, "time_sec": 91.17521905899048 }, { "epoch": 21, "train_loss": 0.5238163524246215, "train_acc": 0.82218, "val_loss": 0.743741114616394, "val_acc": 0.7557, "lyapunov": null, "grad_norm": 2.492968324188999, "time_sec": 91.08112859725952 }, { "epoch": 22, "train_loss": 0.5038600467681885, "train_acc": 0.8254, "val_loss": 0.9449014335632324, "val_acc": 0.7476, "lyapunov": null, "grad_norm": 2.38494268190214, "time_sec": 91.08318662643433 }, { "epoch": 23, "train_loss": 0.4813262654876709, "train_acc": 0.83484, "val_loss": 0.7643151635169982, "val_acc": 0.7431, "lyapunov": null, "grad_norm": 2.2753380735557895, "time_sec": 91.15192151069641 }, { "epoch": 24, "train_loss": 0.47195362268447877, "train_acc": 0.83722, "val_loss": 0.7273788664340973, "val_acc": 0.7618, "lyapunov": null, "grad_norm": 2.257196275035397, "time_sec": 91.06708598136902 }, { "epoch": 25, "train_loss": 0.4544234738063812, "train_acc": 0.84296, "val_loss": 0.7671098015785217, "val_acc": 0.7515, "lyapunov": null, "grad_norm": 2.1828503421488383, "time_sec": 91.08041405677795 }, { "epoch": 26, "train_loss": 0.44203910090446474, "train_acc": 0.84536, "val_loss": 0.7896852376937866, "val_acc": 0.7633, "lyapunov": null, "grad_norm": 2.1131466807169184, "time_sec": 91.05334949493408 }, { "epoch": 27, "train_loss": 0.4236212615394592, "train_acc": 0.85152, "val_loss": 0.6539630149364472, "val_acc": 0.7822, "lyapunov": null, "grad_norm": 2.0665084946802477, "time_sec": 91.07104849815369 }, { "epoch": 28, "train_loss": 0.40995263217926026, "train_acc": 0.85758, "val_loss": 1.198775110054016, "val_acc": 0.7598, "lyapunov": null, "grad_norm": 1.9610051485066777, "time_sec": 91.14437198638916 }, { "epoch": 29, "train_loss": 0.39414338116645814, "train_acc": 0.86214, "val_loss": 0.791292234325409, "val_acc": 0.7738, "lyapunov": null, "grad_norm": 1.9087488305687828, "time_sec": 91.08526229858398 }, { "epoch": 30, "train_loss": 0.3875082825088501, "train_acc": 0.86524, "val_loss": 0.8272250856399536, "val_acc": 0.7661, "lyapunov": null, "grad_norm": 1.8915918686868698, "time_sec": 91.12164497375488 }, { "epoch": 31, "train_loss": 0.3789314131355286, "train_acc": 0.86774, "val_loss": 1.0417310526847838, "val_acc": 0.7479, "lyapunov": null, "grad_norm": 1.8666795147822723, "time_sec": 91.09850406646729 }, { "epoch": 32, "train_loss": 0.366342074842453, "train_acc": 0.87018, "val_loss": 0.7929673102378845, "val_acc": 0.7758, "lyapunov": null, "grad_norm": 1.8350469642136915, "time_sec": 91.11914205551147 }, { "epoch": 33, "train_loss": 0.3532032056808472, "train_acc": 0.87628, "val_loss": 0.8825206691741944, "val_acc": 0.7629, "lyapunov": null, "grad_norm": 1.802498153794555, "time_sec": 91.08285474777222 }, { "epoch": 34, "train_loss": 0.3404730856513977, "train_acc": 0.88022, "val_loss": 0.9344773509025573, "val_acc": 0.7658, "lyapunov": null, "grad_norm": 1.7587958260631384, "time_sec": 91.06740760803223 }, { "epoch": 35, "train_loss": 0.33484573247909544, "train_acc": 0.88322, "val_loss": 1.010736496257782, "val_acc": 0.7683, "lyapunov": null, "grad_norm": 1.7304745993805748, "time_sec": 91.12027788162231 }, { "epoch": 36, "train_loss": 0.32287414850234986, "train_acc": 0.88684, "val_loss": 0.8117256899833679, "val_acc": 0.7721, "lyapunov": null, "grad_norm": 1.7313240186061953, "time_sec": 91.06619668006897 }, { "epoch": 37, "train_loss": 0.31364255765914917, "train_acc": 0.89132, "val_loss": 0.8288133618354797, "val_acc": 0.7667, "lyapunov": null, "grad_norm": 1.678412104589021, "time_sec": 91.17135667800903 }, { "epoch": 38, "train_loss": 0.3079261671066284, "train_acc": 0.8898, "val_loss": 0.9024649682044983, "val_acc": 0.7728, "lyapunov": null, "grad_norm": 1.6838346830691187, "time_sec": 91.07293057441711 }, { "epoch": 39, "train_loss": 0.3003596179485321, "train_acc": 0.89388, "val_loss": 0.9279602525711059, "val_acc": 0.7737, "lyapunov": null, "grad_norm": 1.6559655177971517, "time_sec": 91.10473394393921 }, { "epoch": 40, "train_loss": 0.2992150705242157, "train_acc": 0.89256, "val_loss": 0.8862544967651367, "val_acc": 0.7571, "lyapunov": null, "grad_norm": 1.6702414079656824, "time_sec": 91.06724452972412 }, { "epoch": 41, "train_loss": 0.288808418674469, "train_acc": 0.8989, "val_loss": 1.0894051409721375, "val_acc": 0.7731, "lyapunov": null, "grad_norm": 1.6274206248197605, "time_sec": 91.06420516967773 }, { "epoch": 42, "train_loss": 0.28119449749946596, "train_acc": 0.90048, "val_loss": 0.7861201769828796, "val_acc": 0.7868, "lyapunov": null, "grad_norm": 1.605596888674636, "time_sec": 91.12522029876709 }, { "epoch": 43, "train_loss": 0.278984568529129, "train_acc": 0.90176, "val_loss": 1.0400042490005492, "val_acc": 0.7637, "lyapunov": null, "grad_norm": 1.6082259693236154, "time_sec": 91.05233359336853 }, { "epoch": 44, "train_loss": 0.27851916655540465, "train_acc": 0.90122, "val_loss": 0.8848071938991546, "val_acc": 0.7784, "lyapunov": null, "grad_norm": 1.6122841633085097, "time_sec": 91.11059141159058 }, { "epoch": 45, "train_loss": 0.26998268226623534, "train_acc": 0.90376, "val_loss": 1.23743321685791, "val_acc": 0.7545, "lyapunov": null, "grad_norm": 1.5856940666107922, "time_sec": 91.03389286994934 }, { "epoch": 46, "train_loss": 0.2686913970518112, "train_acc": 0.905, "val_loss": 0.749934101676941, "val_acc": 0.7858, "lyapunov": null, "grad_norm": 1.6003350279819, "time_sec": 91.05149912834167 }, { "epoch": 47, "train_loss": 0.26523829220294953, "train_acc": 0.90602, "val_loss": 0.8804454656600952, "val_acc": 0.7755, "lyapunov": null, "grad_norm": 1.5346717593725685, "time_sec": 91.09767937660217 }, { "epoch": 48, "train_loss": 0.2641575988340378, "train_acc": 0.9058, "val_loss": 0.7957770586967469, "val_acc": 0.7896, "lyapunov": null, "grad_norm": 1.5696702622952918, "time_sec": 91.0563895702362 }, { "epoch": 49, "train_loss": 0.26550782563209535, "train_acc": 0.90644, "val_loss": 0.9611041374206543, "val_acc": 0.7777, "lyapunov": null, "grad_norm": 1.561546766349159, "time_sec": 91.1063346862793 }, { "epoch": 50, "train_loss": 0.26444904836654665, "train_acc": 0.90688, "val_loss": 0.7325167735099792, "val_acc": 0.7855, "lyapunov": null, "grad_norm": 1.5788265124365515, "time_sec": 91.0480043888092 } ] }, "lyapunov": { "2": [ { "epoch": 1, "train_loss": 10.00867161026001, "train_acc": 0.2901, "val_loss": 3.556594299697876, "val_acc": 0.4123, "lyapunov": 1.2662327935933457, "grad_norm": 9.999999972657708, "time_sec": 177.14422726631165 }, { "epoch": 2, "train_loss": 4.2217253616333, "train_acc": 0.36094, "val_loss": 1.9551007186889648, "val_acc": 0.4846, "lyapunov": 1.1406274261072165, "grad_norm": 9.999999984513256, "time_sec": 177.0672733783722 }, { "epoch": 3, "train_loss": 2.1374628715896606, "train_acc": 0.40514, "val_loss": 1.3729649112701416, "val_acc": 0.508, "lyapunov": 0.8079845673592804, "grad_norm": 9.999642019529409, "time_sec": 177.04155683517456 }, { "epoch": 4, "train_loss": 1.5955838970947265, "train_acc": 0.49034, "val_loss": 1.2072273414611816, "val_acc": 0.5801, "lyapunov": 0.641054344756524, "grad_norm": 5.528607621093805, "time_sec": 177.06303143501282 }, { "epoch": 5, "train_loss": 1.4756179521942139, "train_acc": 0.53236, "val_loss": 1.1856598934173583, "val_acc": 0.5878, "lyapunov": 0.616802359480992, "grad_norm": 4.052585938234476, "time_sec": 177.1198832988739 }, { "epoch": 6, "train_loss": 1.4178138249969483, "train_acc": 0.5523, "val_loss": 1.1057056358337403, "val_acc": 0.6114, "lyapunov": 0.6097407252587321, "grad_norm": 3.428489839568916, "time_sec": 177.24686312675476 }, { "epoch": 7, "train_loss": 1.3730882105255127, "train_acc": 0.5681, "val_loss": 1.0766417882919312, "val_acc": 0.632, "lyapunov": 0.6017543702479213, "grad_norm": 2.827206377968448, "time_sec": 177.08226132392883 }, { "epoch": 8, "train_loss": 1.3444231718063355, "train_acc": 0.578, "val_loss": 1.069501482772827, "val_acc": 0.6281, "lyapunov": 0.59977823114761, "grad_norm": 2.519604420727931, "time_sec": 177.0960569381714 }, { "epoch": 9, "train_loss": 1.3076832933044433, "train_acc": 0.5899, "val_loss": 1.0467708824157715, "val_acc": 0.6388, "lyapunov": 0.5927504607478676, "grad_norm": 2.2898083536939486, "time_sec": 177.06478142738342 }, { "epoch": 10, "train_loss": 1.2969296353530884, "train_acc": 0.59402, "val_loss": 1.039936413860321, "val_acc": 0.6432, "lyapunov": 0.5884103413737948, "grad_norm": 2.201081026630667, "time_sec": 177.13298535346985 }, { "epoch": 11, "train_loss": 1.2667249417114257, "train_acc": 0.60472, "val_loss": 0.986871435546875, "val_acc": 0.6622, "lyapunov": 0.5871900952685519, "grad_norm": 2.110025088226681, "time_sec": 177.09048771858215 }, { "epoch": 12, "train_loss": 1.2471018991088867, "train_acc": 0.61152, "val_loss": 1.0002752702713014, "val_acc": 0.6568, "lyapunov": 0.5852226400009507, "grad_norm": 2.0377669281134825, "time_sec": 177.14653301239014 }, { "epoch": 13, "train_loss": 1.2413101124954224, "train_acc": 0.61482, "val_loss": 0.9682174716949463, "val_acc": 0.6633, "lyapunov": 0.5857798885506438, "grad_norm": 1.996717173334976, "time_sec": 177.14922714233398 }, { "epoch": 14, "train_loss": 1.220883334350586, "train_acc": 0.6241, "val_loss": 1.050918565750122, "val_acc": 0.6424, "lyapunov": 0.5835773954001229, "grad_norm": 1.9359088620182465, "time_sec": 177.08484768867493 }, { "epoch": 15, "train_loss": 1.2058533410644532, "train_acc": 0.62666, "val_loss": 0.94383997631073, "val_acc": 0.6734, "lyapunov": 0.5801496031644094, "grad_norm": 1.9245953484666602, "time_sec": 177.1290466785431 }, { "epoch": 16, "train_loss": 1.194775874557495, "train_acc": 0.62898, "val_loss": 0.9896576827049255, "val_acc": 0.6618, "lyapunov": 0.5821160230490253, "grad_norm": 1.8946682527894054, "time_sec": 177.09172439575195 }, { "epoch": 17, "train_loss": 1.1828801498794557, "train_acc": 0.63608, "val_loss": 0.9624389289855957, "val_acc": 0.6712, "lyapunov": 0.5798683271688574, "grad_norm": 1.8806693934004701, "time_sec": 177.09124517440796 }, { "epoch": 18, "train_loss": 1.1725623610305786, "train_acc": 0.63874, "val_loss": 0.9183612252235412, "val_acc": 0.6832, "lyapunov": 0.5785246354997006, "grad_norm": 1.8650697079450238, "time_sec": 177.22030663490295 }, { "epoch": 19, "train_loss": 1.1606917097091676, "train_acc": 0.64442, "val_loss": 0.9561884815216064, "val_acc": 0.6735, "lyapunov": 0.5801093685809914, "grad_norm": 1.8235877986644256, "time_sec": 177.242351770401 }, { "epoch": 20, "train_loss": 1.1478228996658326, "train_acc": 0.6487, "val_loss": 0.913348525428772, "val_acc": 0.6865, "lyapunov": 0.581326979657878, "grad_norm": 1.7991658201234417, "time_sec": 177.21031212806702 }, { "epoch": 21, "train_loss": 1.1375225988006592, "train_acc": 0.65028, "val_loss": 0.9083258006095887, "val_acc": 0.689, "lyapunov": 0.5747417078908447, "grad_norm": 1.8037473520837901, "time_sec": 177.20658564567566 }, { "epoch": 22, "train_loss": 1.1295667491531372, "train_acc": 0.65348, "val_loss": 0.8816516056060791, "val_acc": 0.6964, "lyapunov": 0.5724963845346894, "grad_norm": 1.791938068245746, "time_sec": 177.21554136276245 }, { "epoch": 23, "train_loss": 1.1222690965652466, "train_acc": 0.65656, "val_loss": 0.8969967755317688, "val_acc": 0.6945, "lyapunov": 0.5793853938732001, "grad_norm": 1.7925037856606496, "time_sec": 177.1935474872589 }, { "epoch": 24, "train_loss": 1.104644107131958, "train_acc": 0.66278, "val_loss": 0.9318475513458252, "val_acc": 0.6834, "lyapunov": 0.5741390585899353, "grad_norm": 1.8155815877307544, "time_sec": 177.19995975494385 }, { "epoch": 25, "train_loss": 1.105108909187317, "train_acc": 0.66204, "val_loss": 0.8600421590805054, "val_acc": 0.7064, "lyapunov": 0.5767092594252828, "grad_norm": 1.7993484819343197, "time_sec": 177.18518948554993 }, { "epoch": 26, "train_loss": 1.0923445876312257, "train_acc": 0.66686, "val_loss": 0.8784162574768066, "val_acc": 0.7016, "lyapunov": 0.572128444376504, "grad_norm": 1.8097137804532373, "time_sec": 177.20949387550354 }, { "epoch": 27, "train_loss": 1.0864823266983032, "train_acc": 0.66856, "val_loss": 0.8758655055046082, "val_acc": 0.7005, "lyapunov": 0.572280927997111, "grad_norm": 1.7800939229274781, "time_sec": 177.2026491165161 }, { "epoch": 28, "train_loss": 1.0730646877288819, "train_acc": 0.67458, "val_loss": 0.8712836568832397, "val_acc": 0.7021, "lyapunov": 0.571618944863834, "grad_norm": 1.7856901159868865, "time_sec": 177.1995050907135 }, { "epoch": 29, "train_loss": 1.0670086417770386, "train_acc": 0.67514, "val_loss": 0.8878833614349365, "val_acc": 0.701, "lyapunov": 0.5699482935163981, "grad_norm": 1.7847949574151252, "time_sec": 177.19811415672302 }, { "epoch": 30, "train_loss": 1.0674501267814636, "train_acc": 0.67422, "val_loss": 0.8703726590156555, "val_acc": 0.7012, "lyapunov": 0.5725009469577419, "grad_norm": 1.8124607426755692, "time_sec": 177.18338465690613 }, { "epoch": 31, "train_loss": 1.0580872708892821, "train_acc": 0.67934, "val_loss": 0.8327074008941651, "val_acc": 0.7181, "lyapunov": 0.5730715616584738, "grad_norm": 1.7899908745094062, "time_sec": 177.24247097969055 }, { "epoch": 32, "train_loss": 1.0518890196228028, "train_acc": 0.67974, "val_loss": 0.8463589043617249, "val_acc": 0.7078, "lyapunov": 0.5694142427590801, "grad_norm": 1.7866114147321612, "time_sec": 177.19917917251587 }, { "epoch": 33, "train_loss": 1.0460617045211793, "train_acc": 0.68388, "val_loss": 0.8335932504653931, "val_acc": 0.7158, "lyapunov": 0.5693969742568863, "grad_norm": 1.7952886399086039, "time_sec": 177.15918970108032 }, { "epoch": 34, "train_loss": 1.0423357637405395, "train_acc": 0.6837, "val_loss": 0.850832084274292, "val_acc": 0.7099, "lyapunov": 0.5704684868797927, "grad_norm": 1.8065783440263816, "time_sec": 177.20710682868958 }, { "epoch": 35, "train_loss": 1.0330335846710206, "train_acc": 0.68706, "val_loss": 0.8288890453338623, "val_acc": 0.7188, "lyapunov": 0.5733764781366528, "grad_norm": 1.812065019653932, "time_sec": 177.23326182365417 }, { "epoch": 36, "train_loss": 1.0246629030609131, "train_acc": 0.6906, "val_loss": 0.8290957337379455, "val_acc": 0.7236, "lyapunov": 0.5729466845922153, "grad_norm": 1.798517271283645, "time_sec": 177.20714855194092 }, { "epoch": 37, "train_loss": 1.0212951921081543, "train_acc": 0.69156, "val_loss": 0.8780447118759155, "val_acc": 0.7009, "lyapunov": 0.5703875851600676, "grad_norm": 1.7790716245564815, "time_sec": 177.20857310295105 }, { "epoch": 38, "train_loss": 1.0166065337181092, "train_acc": 0.69364, "val_loss": 0.834115256690979, "val_acc": 0.7155, "lyapunov": 0.5691947783807965, "grad_norm": 1.7758017311029481, "time_sec": 177.24769520759583 }, { "epoch": 39, "train_loss": 1.0144381577301025, "train_acc": 0.69358, "val_loss": 0.808506533241272, "val_acc": 0.7255, "lyapunov": 0.5692081074123188, "grad_norm": 1.798700579562403, "time_sec": 177.2141478061676 }, { "epoch": 40, "train_loss": 1.0112664874267578, "train_acc": 0.69498, "val_loss": 0.8159060410499572, "val_acc": 0.7196, "lyapunov": 0.5688003923581995, "grad_norm": 1.7800055414547078, "time_sec": 177.2447235584259 }, { "epoch": 41, "train_loss": 1.0045425395965577, "train_acc": 0.69726, "val_loss": 0.8107423727035522, "val_acc": 0.7282, "lyapunov": 0.5694415163048698, "grad_norm": 1.784610660218704, "time_sec": 177.2320418357849 }, { "epoch": 42, "train_loss": 1.0022872575378419, "train_acc": 0.69796, "val_loss": 0.8473073066711426, "val_acc": 0.7136, "lyapunov": 0.5713582132630945, "grad_norm": 1.7783584614860928, "time_sec": 177.26206636428833 }, { "epoch": 43, "train_loss": 1.0038085291290284, "train_acc": 0.69738, "val_loss": 0.8214761892318726, "val_acc": 0.7185, "lyapunov": 0.568825355728569, "grad_norm": 1.7870992162881894, "time_sec": 177.22639846801758 }, { "epoch": 44, "train_loss": 1.001431153793335, "train_acc": 0.70028, "val_loss": 0.8304349187850952, "val_acc": 0.7189, "lyapunov": 0.570763453803099, "grad_norm": 1.776886679025421, "time_sec": 177.23433327674866 }, { "epoch": 45, "train_loss": 0.9971029626846314, "train_acc": 0.69918, "val_loss": 0.8164439687728882, "val_acc": 0.7211, "lyapunov": 0.5681786440370028, "grad_norm": 1.791195789851525, "time_sec": 177.22069835662842 }, { "epoch": 46, "train_loss": 0.993794185295105, "train_acc": 0.70168, "val_loss": 0.8193776789665222, "val_acc": 0.7223, "lyapunov": 0.5720337757369136, "grad_norm": 1.7734976718581525, "time_sec": 177.1769037246704 }, { "epoch": 47, "train_loss": 0.9959239391326904, "train_acc": 0.70196, "val_loss": 0.8060067218780518, "val_acc": 0.7272, "lyapunov": 0.5692485049557503, "grad_norm": 1.7695865743994337, "time_sec": 177.21714162826538 }, { "epoch": 48, "train_loss": 0.997929087677002, "train_acc": 0.70236, "val_loss": 0.8053930685043335, "val_acc": 0.726, "lyapunov": 0.5706733504829504, "grad_norm": 1.7872155562389298, "time_sec": 177.22382926940918 }, { "epoch": 49, "train_loss": 0.9953638632011413, "train_acc": 0.70096, "val_loss": 0.8031024269104003, "val_acc": 0.7275, "lyapunov": 0.572128653221423, "grad_norm": 1.7958740827868154, "time_sec": 177.23827815055847 }, { "epoch": 50, "train_loss": 0.996152099685669, "train_acc": 0.70046, "val_loss": 0.8484622112274169, "val_acc": 0.7138, "lyapunov": 0.5700928757867545, "grad_norm": 1.7750658879845262, "time_sec": 177.20626497268677 } ], "3": [ { "epoch": 1, "train_loss": 9.886637321777343, "train_acc": 0.26848, "val_loss": 5.565004052734375, "val_acc": 0.3456, "lyapunov": 1.8496360443437192, "grad_norm": 9.999999967402346, "time_sec": 219.1390881538391 }, { "epoch": 2, "train_loss": 5.418411326141357, "train_acc": 0.3209, "val_loss": 3.7224776931762698, "val_acc": 0.4075, "lyapunov": 1.4648683302847625, "grad_norm": 10.000000101663693, "time_sec": 219.07946133613586 }, { "epoch": 3, "train_loss": 3.650782730026245, "train_acc": 0.38092, "val_loss": 2.0602984258651733, "val_acc": 0.4174, "lyapunov": 1.2480086103424697, "grad_norm": 9.999999946225584, "time_sec": 219.0771780014038 }, { "epoch": 4, "train_loss": 2.9090908791351318, "train_acc": 0.41886, "val_loss": 2.1794303848266603, "val_acc": 0.4612, "lyapunov": 1.1326882084617225, "grad_norm": 9.999999541203039, "time_sec": 219.11379528045654 }, { "epoch": 5, "train_loss": 2.395601005325317, "train_acc": 0.44532, "val_loss": 1.4789248741149903, "val_acc": 0.5109, "lyapunov": 1.0623350192213912, "grad_norm": 9.728179880050458, "time_sec": 219.06865787506104 }, { "epoch": 6, "train_loss": 1.8056759689331054, "train_acc": 0.48266, "val_loss": 1.238930571937561, "val_acc": 0.5635, "lyapunov": 0.9768910978151404, "grad_norm": 3.4335858047830485, "time_sec": 219.0736849308014 }, { "epoch": 7, "train_loss": 1.687066988182068, "train_acc": 0.51668, "val_loss": 1.2204061416625978, "val_acc": 0.5729, "lyapunov": 0.9417259869977946, "grad_norm": 2.1114961149077396, "time_sec": 219.04715156555176 }, { "epoch": 8, "train_loss": 1.6375336548614503, "train_acc": 0.53186, "val_loss": 1.1962882843017577, "val_acc": 0.5805, "lyapunov": 0.9210993046955684, "grad_norm": 1.9303766089141603, "time_sec": 219.0722496509552 }, { "epoch": 9, "train_loss": 1.6033491290283204, "train_acc": 0.53924, "val_loss": 1.1552103568077088, "val_acc": 0.5961, "lyapunov": 0.9094567945241319, "grad_norm": 1.8416646813127695, "time_sec": 219.06279873847961 }, { "epoch": 10, "train_loss": 1.5746748036956788, "train_acc": 0.5494, "val_loss": 1.132537739944458, "val_acc": 0.6006, "lyapunov": 0.9074819725187843, "grad_norm": 1.7646141676782297, "time_sec": 219.07225108146667 }, { "epoch": 11, "train_loss": 1.5544266968536378, "train_acc": 0.55674, "val_loss": 1.1234738679885865, "val_acc": 0.6095, "lyapunov": 0.9017732987928269, "grad_norm": 1.7022438864081173, "time_sec": 219.04692029953003 }, { "epoch": 12, "train_loss": 1.5270355632019044, "train_acc": 0.56824, "val_loss": 1.1039700313568115, "val_acc": 0.6113, "lyapunov": 0.8970051421533765, "grad_norm": 1.686004618284347, "time_sec": 219.09399819374084 }, { "epoch": 13, "train_loss": 1.5147102879333496, "train_acc": 0.57448, "val_loss": 1.1072441032409668, "val_acc": 0.6124, "lyapunov": 0.9001958841253119, "grad_norm": 1.662854581910154, "time_sec": 219.0895917415619 }, { "epoch": 14, "train_loss": 1.4993468006134034, "train_acc": 0.57916, "val_loss": 1.0862106854438782, "val_acc": 0.6218, "lyapunov": 0.8944531209633478, "grad_norm": 1.635496861194711, "time_sec": 219.12134218215942 }, { "epoch": 15, "train_loss": 1.4819088359451293, "train_acc": 0.57984, "val_loss": 1.06874298992157, "val_acc": 0.6275, "lyapunov": 0.8945445736960682, "grad_norm": 1.6235776440268546, "time_sec": 219.09631848335266 }, { "epoch": 16, "train_loss": 1.4684392934036254, "train_acc": 0.58558, "val_loss": 1.0290869824409485, "val_acc": 0.6368, "lyapunov": 0.889524121571075, "grad_norm": 1.618824365813371, "time_sec": 219.08521628379822 }, { "epoch": 17, "train_loss": 1.4566676177215576, "train_acc": 0.59174, "val_loss": 1.0311160945892335, "val_acc": 0.645, "lyapunov": 0.8914786020813086, "grad_norm": 1.6193885880914556, "time_sec": 219.1012179851532 }, { "epoch": 18, "train_loss": 1.447174312324524, "train_acc": 0.59298, "val_loss": 1.0276520294189453, "val_acc": 0.6382, "lyapunov": 0.8889464463114434, "grad_norm": 1.6230962631323445, "time_sec": 219.0996561050415 }, { "epoch": 19, "train_loss": 1.4324742112350464, "train_acc": 0.59858, "val_loss": 1.0219873019218444, "val_acc": 0.6407, "lyapunov": 0.8865713836896755, "grad_norm": 1.5870203143380517, "time_sec": 219.03141736984253 }, { "epoch": 20, "train_loss": 1.4191896475982666, "train_acc": 0.60328, "val_loss": 1.023682218170166, "val_acc": 0.6438, "lyapunov": 0.8828371200720062, "grad_norm": 1.622239828865246, "time_sec": 219.0444712638855 }, { "epoch": 21, "train_loss": 1.4107023125457763, "train_acc": 0.6056, "val_loss": 1.0241008337020874, "val_acc": 0.645, "lyapunov": 0.879414847134934, "grad_norm": 1.5932543454385197, "time_sec": 219.05590295791626 }, { "epoch": 22, "train_loss": 1.4019806679534912, "train_acc": 0.60984, "val_loss": 0.9887507389068604, "val_acc": 0.6555, "lyapunov": 0.8817187197068158, "grad_norm": 1.6025708656723956, "time_sec": 219.0642237663269 }, { "epoch": 23, "train_loss": 1.397972855873108, "train_acc": 0.61208, "val_loss": 1.0166104084014893, "val_acc": 0.6494, "lyapunov": 0.8882192618706647, "grad_norm": 1.6263535429671192, "time_sec": 219.102778673172 }, { "epoch": 24, "train_loss": 1.3810771157455444, "train_acc": 0.61566, "val_loss": 0.9922153280258179, "val_acc": 0.6493, "lyapunov": 0.8784188744052291, "grad_norm": 1.6058554101552192, "time_sec": 219.08169221878052 }, { "epoch": 25, "train_loss": 1.3791738171386718, "train_acc": 0.61852, "val_loss": 0.977457004737854, "val_acc": 0.6613, "lyapunov": 0.8845000114587261, "grad_norm": 1.5997112892609862, "time_sec": 219.1283619403839 }, { "epoch": 26, "train_loss": 1.3683585110092162, "train_acc": 0.62242, "val_loss": 0.9783126133918763, "val_acc": 0.6574, "lyapunov": 0.8817774313490104, "grad_norm": 1.599428458028868, "time_sec": 219.06762385368347 }, { "epoch": 27, "train_loss": 1.3609592821121215, "train_acc": 0.62566, "val_loss": 0.9719936261177063, "val_acc": 0.6624, "lyapunov": 0.880928546114041, "grad_norm": 1.6153498758376685, "time_sec": 219.089661359787 }, { "epoch": 28, "train_loss": 1.3513427503585815, "train_acc": 0.62728, "val_loss": 0.9601885829925537, "val_acc": 0.6619, "lyapunov": 0.8753232303482797, "grad_norm": 1.6308248667486007, "time_sec": 219.1258945465088 }, { "epoch": 29, "train_loss": 1.349013708114624, "train_acc": 0.6271, "val_loss": 1.021242923259735, "val_acc": 0.6596, "lyapunov": 0.8816047269669945, "grad_norm": 1.6322710335257884, "time_sec": 219.11040592193604 }, { "epoch": 30, "train_loss": 1.3382837686920166, "train_acc": 0.63418, "val_loss": 0.949721337890625, "val_acc": 0.6721, "lyapunov": 0.8785617967395831, "grad_norm": 1.6267116459919577, "time_sec": 219.0830078125 }, { "epoch": 31, "train_loss": 1.334988670425415, "train_acc": 0.63302, "val_loss": 0.9434608960151672, "val_acc": 0.6733, "lyapunov": 0.879274760060908, "grad_norm": 1.6375901529516637, "time_sec": 219.16829204559326 }, { "epoch": 32, "train_loss": 1.3272022075653076, "train_acc": 0.63748, "val_loss": 0.9399231541633606, "val_acc": 0.6742, "lyapunov": 0.8773337799264952, "grad_norm": 1.6584354131565497, "time_sec": 219.16436123847961 }, { "epoch": 33, "train_loss": 1.3261497624969483, "train_acc": 0.63866, "val_loss": 0.9691579351425171, "val_acc": 0.6642, "lyapunov": 0.880438138304464, "grad_norm": 1.6311048225485472, "time_sec": 219.2267725467682 }, { "epoch": 34, "train_loss": 1.3194347483062745, "train_acc": 0.63966, "val_loss": 0.9293135822296142, "val_acc": 0.6774, "lyapunov": 0.882823745612903, "grad_norm": 1.6424652064703287, "time_sec": 219.09363412857056 }, { "epoch": 35, "train_loss": 1.3166248292541505, "train_acc": 0.63946, "val_loss": 0.934987894821167, "val_acc": 0.6737, "lyapunov": 0.8773819311805393, "grad_norm": 1.6712941684477247, "time_sec": 219.09669399261475 }, { "epoch": 36, "train_loss": 1.3065439778137207, "train_acc": 0.64482, "val_loss": 0.9304282165527343, "val_acc": 0.6777, "lyapunov": 0.8768021194526302, "grad_norm": 1.6930105520996321, "time_sec": 219.0884392261505 }, { "epoch": 37, "train_loss": 1.3023283739089966, "train_acc": 0.64504, "val_loss": 0.9790545976638794, "val_acc": 0.6648, "lyapunov": 0.8778117370727422, "grad_norm": 1.629315772213287, "time_sec": 219.06800413131714 }, { "epoch": 38, "train_loss": 1.2997008090972901, "train_acc": 0.6443, "val_loss": 0.9367351039886475, "val_acc": 0.669, "lyapunov": 0.8753928830251669, "grad_norm": 1.6725969377411714, "time_sec": 219.10576462745667 }, { "epoch": 39, "train_loss": 1.2947760388565064, "train_acc": 0.64918, "val_loss": 0.9696623109817505, "val_acc": 0.6728, "lyapunov": 0.8771997676481067, "grad_norm": 1.6453232439310457, "time_sec": 219.10160493850708 }, { "epoch": 40, "train_loss": 1.2909610508346558, "train_acc": 0.6483, "val_loss": 0.9222309721946716, "val_acc": 0.6798, "lyapunov": 0.8764342953786826, "grad_norm": 1.650036990648839, "time_sec": 219.09872603416443 }, { "epoch": 41, "train_loss": 1.2923656301116944, "train_acc": 0.65116, "val_loss": 0.9597509237289429, "val_acc": 0.6743, "lyapunov": 0.8783976452429886, "grad_norm": 1.6434547264105122, "time_sec": 219.12360668182373 }, { "epoch": 42, "train_loss": 1.2929988737106324, "train_acc": 0.64904, "val_loss": 0.9352736534118652, "val_acc": 0.6793, "lyapunov": 0.8806399698452572, "grad_norm": 1.6455355826726663, "time_sec": 219.1057686805725 }, { "epoch": 43, "train_loss": 1.2854210181427002, "train_acc": 0.65064, "val_loss": 0.9394157413482666, "val_acc": 0.6761, "lyapunov": 0.873351320281358, "grad_norm": 1.6490076033134402, "time_sec": 219.1147177219391 }, { "epoch": 44, "train_loss": 1.2869935691070558, "train_acc": 0.65004, "val_loss": 0.9156728206634521, "val_acc": 0.6796, "lyapunov": 0.877838894534294, "grad_norm": 1.6440486253976125, "time_sec": 219.08031749725342 }, { "epoch": 45, "train_loss": 1.2884554972839355, "train_acc": 0.64852, "val_loss": 0.9147618005752564, "val_acc": 0.684, "lyapunov": 0.8746943867115109, "grad_norm": 1.642373597875226, "time_sec": 219.1208529472351 }, { "epoch": 46, "train_loss": 1.2821145919799806, "train_acc": 0.65198, "val_loss": 0.9232476442337036, "val_acc": 0.6816, "lyapunov": 0.8818693170157235, "grad_norm": 1.6470363025736585, "time_sec": 219.13471055030823 }, { "epoch": 47, "train_loss": 1.2783561626434325, "train_acc": 0.65318, "val_loss": 0.9129555612564086, "val_acc": 0.6814, "lyapunov": 0.8758536895827564, "grad_norm": 1.6422393825142525, "time_sec": 219.11879968643188 }, { "epoch": 48, "train_loss": 1.278603917503357, "train_acc": 0.65008, "val_loss": 0.9061016012191773, "val_acc": 0.6845, "lyapunov": 0.8759899017451059, "grad_norm": 1.6383354053392198, "time_sec": 219.11400628089905 }, { "epoch": 49, "train_loss": 1.2778211384582518, "train_acc": 0.65284, "val_loss": 0.9253755199432373, "val_acc": 0.6792, "lyapunov": 0.8724236613344354, "grad_norm": 1.6314070765195892, "time_sec": 219.09536862373352 }, { "epoch": 50, "train_loss": 1.2822308771133424, "train_acc": 0.653, "val_loss": 0.9114741864204406, "val_acc": 0.6793, "lyapunov": 0.8772178110869034, "grad_norm": 1.626801820842849, "time_sec": 219.1254141330719 } ], "4": [ { "epoch": 1, "train_loss": 10.138670194396973, "train_acc": 0.22564, "val_loss": 2.939440586471558, "val_acc": 0.341, "lyapunov": 2.6423642336560027, "grad_norm": 10.00000005267747, "time_sec": 243.78172659873962 }, { "epoch": 2, "train_loss": 5.361720401916504, "train_acc": 0.23776, "val_loss": 2.5584325256347658, "val_acc": 0.3378, "lyapunov": 1.9307395585662568, "grad_norm": 10.000000025397728, "time_sec": 243.78110790252686 }, { "epoch": 3, "train_loss": 3.8789426130676268, "train_acc": 0.24894, "val_loss": 2.3140427013397216, "val_acc": 0.2919, "lyapunov": 1.6591560602797877, "grad_norm": 10.000000014183167, "time_sec": 243.82367968559265 }, { "epoch": 4, "train_loss": 3.2440808209991454, "train_acc": 0.25728, "val_loss": 1.9694867576599122, "val_acc": 0.3271, "lyapunov": 1.509594388325196, "grad_norm": 10.000000057688851, "time_sec": 243.82085800170898 }, { "epoch": 5, "train_loss": 2.894227194213867, "train_acc": 0.26456, "val_loss": 1.856060424041748, "val_acc": 0.313, "lyapunov": 1.3861996033002653, "grad_norm": 9.999999739821712, "time_sec": 243.8253037929535 }, { "epoch": 6, "train_loss": 2.6495460874176024, "train_acc": 0.28184, "val_loss": 1.8042585342407227, "val_acc": 0.3492, "lyapunov": 1.2907036170935082, "grad_norm": 9.90851203367567, "time_sec": 243.81157994270325 }, { "epoch": 7, "train_loss": 2.3804813381195067, "train_acc": 0.32046, "val_loss": 1.7507558801651002, "val_acc": 0.3673, "lyapunov": 1.211043564407417, "grad_norm": 5.378190649229295, "time_sec": 243.83509850502014 }, { "epoch": 8, "train_loss": 2.2794188507080078, "train_acc": 0.34198, "val_loss": 1.718031328201294, "val_acc": 0.3887, "lyapunov": 1.1711129146768613, "grad_norm": 2.7892842162429767, "time_sec": 243.8507363796234 }, { "epoch": 9, "train_loss": 2.247270139160156, "train_acc": 0.35496, "val_loss": 1.6821900709152222, "val_acc": 0.3918, "lyapunov": 1.1567430586156333, "grad_norm": 2.136753214886518, "time_sec": 243.77942776679993 }, { "epoch": 10, "train_loss": 2.219769735107422, "train_acc": 0.36068, "val_loss": 1.682488171005249, "val_acc": 0.3962, "lyapunov": 1.1400047815059458, "grad_norm": 1.8639655064576182, "time_sec": 243.71889185905457 }, { "epoch": 11, "train_loss": 2.210514054336548, "train_acc": 0.36224, "val_loss": 1.6685603786468506, "val_acc": 0.393, "lyapunov": 1.1393702749705985, "grad_norm": 1.6904279202048615, "time_sec": 243.76552867889404 }, { "epoch": 12, "train_loss": 2.1925585690307616, "train_acc": 0.36712, "val_loss": 1.6659884477615357, "val_acc": 0.398, "lyapunov": 1.1309379650198894, "grad_norm": 1.5492965693388752, "time_sec": 243.83178281784058 }, { "epoch": 13, "train_loss": 2.1770478023529054, "train_acc": 0.36986, "val_loss": 1.6337458795547486, "val_acc": 0.4168, "lyapunov": 1.1225328469825218, "grad_norm": 1.454272563180731, "time_sec": 243.84237337112427 }, { "epoch": 14, "train_loss": 2.1698232682037353, "train_acc": 0.37324, "val_loss": 1.615553874206543, "val_acc": 0.4197, "lyapunov": 1.1240297525435152, "grad_norm": 1.3796952640403297, "time_sec": 243.8380582332611 }, { "epoch": 15, "train_loss": 2.1603004252624514, "train_acc": 0.3775, "val_loss": 1.6288873954772949, "val_acc": 0.4094, "lyapunov": 1.119924979136728, "grad_norm": 1.318967010481674, "time_sec": 243.79564952850342 }, { "epoch": 16, "train_loss": 2.1588982814788817, "train_acc": 0.37904, "val_loss": 1.658633060646057, "val_acc": 0.4079, "lyapunov": 1.1192828095172678, "grad_norm": 1.3042329820779222, "time_sec": 243.82657074928284 }, { "epoch": 17, "train_loss": 2.1462838565063476, "train_acc": 0.38168, "val_loss": 1.607268097305298, "val_acc": 0.4225, "lyapunov": 1.1170157598107673, "grad_norm": 1.2610877399538434, "time_sec": 243.7789568901062 }, { "epoch": 18, "train_loss": 2.145940422592163, "train_acc": 0.38194, "val_loss": 1.6156129001617432, "val_acc": 0.4192, "lyapunov": 1.1201189590232146, "grad_norm": 1.253906206027379, "time_sec": 243.82261872291565 }, { "epoch": 19, "train_loss": 2.1375477339172364, "train_acc": 0.38598, "val_loss": 1.6065894638061524, "val_acc": 0.4208, "lyapunov": 1.1187894129387252, "grad_norm": 1.2335161276444737, "time_sec": 243.80746340751648 }, { "epoch": 20, "train_loss": 2.1291038385009764, "train_acc": 0.38796, "val_loss": 1.615070446395874, "val_acc": 0.4211, "lyapunov": 1.1210469156884781, "grad_norm": 1.2215213823669768, "time_sec": 243.83611369132996 }, { "epoch": 21, "train_loss": 2.1302540149688722, "train_acc": 0.39134, "val_loss": 1.5864741819381714, "val_acc": 0.433, "lyapunov": 1.1274997468494699, "grad_norm": 1.2165337663420797, "time_sec": 243.77375960350037 }, { "epoch": 22, "train_loss": 2.1284910098266603, "train_acc": 0.39524, "val_loss": 1.579381764793396, "val_acc": 0.436, "lyapunov": 1.130478767787709, "grad_norm": 1.2055723454869056, "time_sec": 243.80992245674133 }, { "epoch": 23, "train_loss": 2.1195063771057128, "train_acc": 0.39552, "val_loss": 1.5710469825744628, "val_acc": 0.4409, "lyapunov": 1.1252597959145256, "grad_norm": 1.2283551336587406, "time_sec": 243.9244520664215 }, { "epoch": 24, "train_loss": 2.1138509731292725, "train_acc": 0.39828, "val_loss": 1.564260850906372, "val_acc": 0.4369, "lyapunov": 1.1288096435234676, "grad_norm": 1.2307373296500659, "time_sec": 243.82610821723938 }, { "epoch": 25, "train_loss": 2.1157819132995606, "train_acc": 0.40046, "val_loss": 1.569395032119751, "val_acc": 0.437, "lyapunov": 1.1359551151085387, "grad_norm": 1.2305800755990723, "time_sec": 243.85861563682556 }, { "epoch": 26, "train_loss": 2.1171055084228514, "train_acc": 0.39876, "val_loss": 1.5628763763427735, "val_acc": 0.438, "lyapunov": 1.141420260719631, "grad_norm": 1.2486725958148854, "time_sec": 243.80560755729675 }, { "epoch": 27, "train_loss": 2.1043425176239015, "train_acc": 0.40506, "val_loss": 1.5874193929672242, "val_acc": 0.4285, "lyapunov": 1.137353905021687, "grad_norm": 1.247112749643353, "time_sec": 243.8357720375061 }, { "epoch": 28, "train_loss": 2.103717846984863, "train_acc": 0.40688, "val_loss": 1.5615127603530883, "val_acc": 0.4407, "lyapunov": 1.1431670996844006, "grad_norm": 1.2526816309795223, "time_sec": 243.81778478622437 }, { "epoch": 29, "train_loss": 2.0979717473220827, "train_acc": 0.4047, "val_loss": 1.5667091354370117, "val_acc": 0.4501, "lyapunov": 1.1390000624424965, "grad_norm": 1.3013089305361618, "time_sec": 243.8064706325531 }, { "epoch": 30, "train_loss": 2.0903312964630127, "train_acc": 0.4098, "val_loss": 1.5484002025604249, "val_acc": 0.4434, "lyapunov": 1.138905145628068, "grad_norm": 1.2886270286420318, "time_sec": 243.7907841205597 }, { "epoch": 31, "train_loss": 2.083065502090454, "train_acc": 0.41286, "val_loss": 1.5562454063415527, "val_acc": 0.4461, "lyapunov": 1.139498563068907, "grad_norm": 1.3231590205872965, "time_sec": 243.79992413520813 }, { "epoch": 32, "train_loss": 2.078911770401001, "train_acc": 0.41752, "val_loss": 1.5308426792144776, "val_acc": 0.453, "lyapunov": 1.137989708834597, "grad_norm": 1.3411978634669315, "time_sec": 243.8079583644867 }, { "epoch": 33, "train_loss": 2.078010279083252, "train_acc": 0.4161, "val_loss": 1.5390575723648072, "val_acc": 0.4547, "lyapunov": 1.1383457985680427, "grad_norm": 1.353302395595091, "time_sec": 243.85199880599976 }, { "epoch": 34, "train_loss": 2.0724924634933473, "train_acc": 0.41916, "val_loss": 1.527733098602295, "val_acc": 0.4552, "lyapunov": 1.139322290640048, "grad_norm": 1.3814791249913954, "time_sec": 243.81034541130066 }, { "epoch": 35, "train_loss": 2.0685569482421875, "train_acc": 0.41744, "val_loss": 1.5176776382446289, "val_acc": 0.4585, "lyapunov": 1.1384284941436689, "grad_norm": 1.4038672727984491, "time_sec": 243.82950949668884 }, { "epoch": 36, "train_loss": 2.06793385055542, "train_acc": 0.42, "val_loss": 1.516915723991394, "val_acc": 0.4592, "lyapunov": 1.140784283123358, "grad_norm": 1.4134147957224994, "time_sec": 243.7948546409607 }, { "epoch": 37, "train_loss": 2.0626228358459473, "train_acc": 0.42134, "val_loss": 1.5186110572814941, "val_acc": 0.4584, "lyapunov": 1.1385917383081772, "grad_norm": 1.4188616642430978, "time_sec": 243.77106380462646 }, { "epoch": 38, "train_loss": 2.057733590507507, "train_acc": 0.42528, "val_loss": 1.527320244407654, "val_acc": 0.4517, "lyapunov": 1.1396347296512341, "grad_norm": 1.402281703786186, "time_sec": 243.83526754379272 }, { "epoch": 39, "train_loss": 2.053653130187988, "train_acc": 0.4246, "val_loss": 1.523566467475891, "val_acc": 0.4567, "lyapunov": 1.1354857343237112, "grad_norm": 1.4241389639781625, "time_sec": 243.77314925193787 }, { "epoch": 40, "train_loss": 2.0530851780700683, "train_acc": 0.42374, "val_loss": 1.52158450050354, "val_acc": 0.4582, "lyapunov": 1.1406449866111932, "grad_norm": 1.4370870493938808, "time_sec": 243.86149287223816 }, { "epoch": 41, "train_loss": 2.0475478923797605, "train_acc": 0.4292, "val_loss": 1.5124943691253663, "val_acc": 0.4608, "lyapunov": 1.138276274246938, "grad_norm": 1.4207255079169452, "time_sec": 243.75415897369385 }, { "epoch": 42, "train_loss": 2.0435234954071047, "train_acc": 0.42796, "val_loss": 1.5179149496078492, "val_acc": 0.4613, "lyapunov": 1.1337595958538982, "grad_norm": 1.449551491486115, "time_sec": 243.79569125175476 }, { "epoch": 43, "train_loss": 2.0462114200592043, "train_acc": 0.42704, "val_loss": 1.5187206199645995, "val_acc": 0.4565, "lyapunov": 1.1361882857349523, "grad_norm": 1.471635773548073, "time_sec": 243.9977855682373 }, { "epoch": 44, "train_loss": 2.0437322373962403, "train_acc": 0.4307, "val_loss": 1.5120391357421874, "val_acc": 0.4611, "lyapunov": 1.136338417182493, "grad_norm": 1.4612676327925023, "time_sec": 243.82597041130066 }, { "epoch": 45, "train_loss": 2.038804543457031, "train_acc": 0.43106, "val_loss": 1.5097675659179688, "val_acc": 0.4618, "lyapunov": 1.1376445680628042, "grad_norm": 1.446689082721745, "time_sec": 243.83146262168884 }, { "epoch": 46, "train_loss": 2.0387796462249757, "train_acc": 0.43206, "val_loss": 1.5252808937072755, "val_acc": 0.4604, "lyapunov": 1.1393481210979355, "grad_norm": 1.4264228353429196, "time_sec": 243.79276418685913 }, { "epoch": 47, "train_loss": 2.043452616043091, "train_acc": 0.4308, "val_loss": 1.508095828819275, "val_acc": 0.4667, "lyapunov": 1.14236963923325, "grad_norm": 1.4640736689067098, "time_sec": 243.76182579994202 }, { "epoch": 48, "train_loss": 2.035581900482178, "train_acc": 0.4319, "val_loss": 1.4961794807434081, "val_acc": 0.4705, "lyapunov": 1.1359667210932582, "grad_norm": 1.4548775377604246, "time_sec": 243.76584720611572 }, { "epoch": 49, "train_loss": 2.0361414821624755, "train_acc": 0.42966, "val_loss": 1.5087008377075195, "val_acc": 0.4619, "lyapunov": 1.1356002241754166, "grad_norm": 1.462573337060383, "time_sec": 243.78386449813843 }, { "epoch": 50, "train_loss": 2.042609598236084, "train_acc": 0.42966, "val_loss": 1.5063008777618407, "val_acc": 0.4629, "lyapunov": 1.1417860432963847, "grad_norm": 1.4554161377681805, "time_sec": 243.8414967060089 } ] } }