| { | |
| "best_metric": 0.6019417475728155, | |
| "best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-conflab-traj-direction-rh-v10/checkpoint-819", | |
| "epoch": 6.135531135531136, | |
| "eval_steps": 500, | |
| "global_step": 819, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01221001221001221, | |
| "grad_norm": 8.122791290283203, | |
| "learning_rate": 6.0975609756097564e-06, | |
| "loss": 2.1095, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02442002442002442, | |
| "grad_norm": 6.661695957183838, | |
| "learning_rate": 1.2195121951219513e-05, | |
| "loss": 2.0298, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03663003663003663, | |
| "grad_norm": 7.086618900299072, | |
| "learning_rate": 1.8292682926829268e-05, | |
| "loss": 2.0189, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04884004884004884, | |
| "grad_norm": 8.178045272827148, | |
| "learning_rate": 2.4390243902439026e-05, | |
| "loss": 2.0099, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06105006105006105, | |
| "grad_norm": 7.585304260253906, | |
| "learning_rate": 3.048780487804878e-05, | |
| "loss": 1.9196, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07326007326007326, | |
| "grad_norm": 7.4838409423828125, | |
| "learning_rate": 3.6585365853658535e-05, | |
| "loss": 1.9165, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08547008547008547, | |
| "grad_norm": 6.174314498901367, | |
| "learning_rate": 4.26829268292683e-05, | |
| "loss": 1.8358, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09768009768009768, | |
| "grad_norm": 5.699541091918945, | |
| "learning_rate": 4.878048780487805e-05, | |
| "loss": 1.8794, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.10989010989010989, | |
| "grad_norm": 6.571558475494385, | |
| "learning_rate": 4.94572591587517e-05, | |
| "loss": 1.8931, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1221001221001221, | |
| "grad_norm": 7.583756446838379, | |
| "learning_rate": 4.877883310719132e-05, | |
| "loss": 1.8477, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1343101343101343, | |
| "grad_norm": 6.113456726074219, | |
| "learning_rate": 4.810040705563094e-05, | |
| "loss": 1.989, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.14407814407814407, | |
| "eval_accuracy": 0.2669902912621359, | |
| "eval_loss": 1.8786594867706299, | |
| "eval_runtime": 28.8858, | |
| "eval_samples_per_second": 7.132, | |
| "eval_steps_per_second": 0.9, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 1.0024420024420024, | |
| "grad_norm": 6.775885105133057, | |
| "learning_rate": 4.742198100407056e-05, | |
| "loss": 1.7872, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0146520146520146, | |
| "grad_norm": 5.531322956085205, | |
| "learning_rate": 4.674355495251018e-05, | |
| "loss": 1.8331, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0268620268620268, | |
| "grad_norm": 7.1956353187561035, | |
| "learning_rate": 4.60651289009498e-05, | |
| "loss": 1.7703, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0390720390720392, | |
| "grad_norm": 8.779479026794434, | |
| "learning_rate": 4.5386702849389416e-05, | |
| "loss": 1.6198, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0512820512820513, | |
| "grad_norm": 10.621439933776855, | |
| "learning_rate": 4.470827679782904e-05, | |
| "loss": 1.6431, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0634920634920635, | |
| "grad_norm": 7.896602630615234, | |
| "learning_rate": 4.402985074626866e-05, | |
| "loss": 1.6686, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0757020757020757, | |
| "grad_norm": 6.063083171844482, | |
| "learning_rate": 4.335142469470828e-05, | |
| "loss": 1.4762, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0879120879120878, | |
| "grad_norm": 10.578692436218262, | |
| "learning_rate": 4.26729986431479e-05, | |
| "loss": 1.6948, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.1001221001221002, | |
| "grad_norm": 8.612629890441895, | |
| "learning_rate": 4.199457259158752e-05, | |
| "loss": 1.5687, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1123321123321124, | |
| "grad_norm": 8.634360313415527, | |
| "learning_rate": 4.131614654002714e-05, | |
| "loss": 1.4138, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1245421245421245, | |
| "grad_norm": 8.842486381530762, | |
| "learning_rate": 4.063772048846676e-05, | |
| "loss": 1.4202, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.1367521367521367, | |
| "grad_norm": 8.472060203552246, | |
| "learning_rate": 3.995929443690638e-05, | |
| "loss": 1.3092, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.144078144078144, | |
| "eval_accuracy": 0.42718446601941745, | |
| "eval_loss": 1.6427617073059082, | |
| "eval_runtime": 21.3576, | |
| "eval_samples_per_second": 9.645, | |
| "eval_steps_per_second": 1.217, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 2.004884004884005, | |
| "grad_norm": 9.463373184204102, | |
| "learning_rate": 3.9280868385345995e-05, | |
| "loss": 1.5065, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.017094017094017, | |
| "grad_norm": 8.447351455688477, | |
| "learning_rate": 3.860244233378562e-05, | |
| "loss": 1.3116, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.029304029304029, | |
| "grad_norm": 9.61158561706543, | |
| "learning_rate": 3.792401628222524e-05, | |
| "loss": 1.0903, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0415140415140414, | |
| "grad_norm": 10.502912521362305, | |
| "learning_rate": 3.724559023066486e-05, | |
| "loss": 1.1713, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.0537240537240535, | |
| "grad_norm": 11.798907279968262, | |
| "learning_rate": 3.656716417910448e-05, | |
| "loss": 1.1561, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.065934065934066, | |
| "grad_norm": 16.395263671875, | |
| "learning_rate": 3.58887381275441e-05, | |
| "loss": 1.2809, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0781440781440783, | |
| "grad_norm": 11.183971405029297, | |
| "learning_rate": 3.521031207598372e-05, | |
| "loss": 1.1276, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0903540903540905, | |
| "grad_norm": 10.743521690368652, | |
| "learning_rate": 3.453188602442334e-05, | |
| "loss": 1.3756, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1025641025641026, | |
| "grad_norm": 8.513188362121582, | |
| "learning_rate": 3.385345997286296e-05, | |
| "loss": 0.9218, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.114774114774115, | |
| "grad_norm": 12.773633003234863, | |
| "learning_rate": 3.3175033921302575e-05, | |
| "loss": 1.2335, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.126984126984127, | |
| "grad_norm": 11.963506698608398, | |
| "learning_rate": 3.24966078697422e-05, | |
| "loss": 1.1391, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.139194139194139, | |
| "grad_norm": 10.355384826660156, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 1.0096, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.144078144078144, | |
| "eval_accuracy": 0.47572815533980584, | |
| "eval_loss": 1.4351158142089844, | |
| "eval_runtime": 27.857, | |
| "eval_samples_per_second": 7.395, | |
| "eval_steps_per_second": 0.933, | |
| "step": 354 | |
| }, | |
| { | |
| "epoch": 3.0073260073260073, | |
| "grad_norm": 10.925836563110352, | |
| "learning_rate": 3.113975576662144e-05, | |
| "loss": 0.9783, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.0195360195360195, | |
| "grad_norm": 6.559803009033203, | |
| "learning_rate": 3.046132971506106e-05, | |
| "loss": 0.8608, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.0317460317460316, | |
| "grad_norm": 10.63405990600586, | |
| "learning_rate": 2.9782903663500678e-05, | |
| "loss": 0.7684, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.043956043956044, | |
| "grad_norm": 11.322102546691895, | |
| "learning_rate": 2.91044776119403e-05, | |
| "loss": 0.8308, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.056166056166056, | |
| "grad_norm": 5.719593524932861, | |
| "learning_rate": 2.842605156037992e-05, | |
| "loss": 0.7868, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.0683760683760686, | |
| "grad_norm": 13.967921257019043, | |
| "learning_rate": 2.7747625508819542e-05, | |
| "loss": 0.7707, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.0805860805860807, | |
| "grad_norm": 9.134116172790527, | |
| "learning_rate": 2.7069199457259158e-05, | |
| "loss": 0.5804, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.092796092796093, | |
| "grad_norm": 14.580177307128906, | |
| "learning_rate": 2.639077340569878e-05, | |
| "loss": 0.7846, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.105006105006105, | |
| "grad_norm": 12.265727043151855, | |
| "learning_rate": 2.57123473541384e-05, | |
| "loss": 0.6807, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.1172161172161172, | |
| "grad_norm": 15.816527366638184, | |
| "learning_rate": 2.5033921302578023e-05, | |
| "loss": 0.6878, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.1294261294261294, | |
| "grad_norm": 9.39810848236084, | |
| "learning_rate": 2.4355495251017642e-05, | |
| "loss": 0.5625, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.1416361416361416, | |
| "grad_norm": 7.950680732727051, | |
| "learning_rate": 2.367706919945726e-05, | |
| "loss": 0.604, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.144078144078144, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.3919281959533691, | |
| "eval_runtime": 21.6391, | |
| "eval_samples_per_second": 9.52, | |
| "eval_steps_per_second": 1.202, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 4.00976800976801, | |
| "grad_norm": 3.859422206878662, | |
| "learning_rate": 2.299864314789688e-05, | |
| "loss": 0.3818, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.021978021978022, | |
| "grad_norm": 4.586574077606201, | |
| "learning_rate": 2.2320217096336503e-05, | |
| "loss": 0.3743, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.034188034188034, | |
| "grad_norm": 11.923030853271484, | |
| "learning_rate": 2.164179104477612e-05, | |
| "loss": 0.4857, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.046398046398046, | |
| "grad_norm": 8.866025924682617, | |
| "learning_rate": 2.0963364993215738e-05, | |
| "loss": 0.5601, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.058608058608058, | |
| "grad_norm": 8.028688430786133, | |
| "learning_rate": 2.028493894165536e-05, | |
| "loss": 0.4649, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.070818070818071, | |
| "grad_norm": 8.852441787719727, | |
| "learning_rate": 1.960651289009498e-05, | |
| "loss": 0.3592, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.083028083028083, | |
| "grad_norm": 22.12917137145996, | |
| "learning_rate": 1.89280868385346e-05, | |
| "loss": 0.4787, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 9.262681007385254, | |
| "learning_rate": 1.824966078697422e-05, | |
| "loss": 0.4364, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.107448107448107, | |
| "grad_norm": 5.102321624755859, | |
| "learning_rate": 1.757123473541384e-05, | |
| "loss": 0.3868, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.119658119658119, | |
| "grad_norm": 13.144558906555176, | |
| "learning_rate": 1.689280868385346e-05, | |
| "loss": 0.4023, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.131868131868132, | |
| "grad_norm": 16.35342788696289, | |
| "learning_rate": 1.6214382632293083e-05, | |
| "loss": 0.4355, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.1440781440781445, | |
| "grad_norm": 12.760746955871582, | |
| "learning_rate": 1.55359565807327e-05, | |
| "loss": 0.2381, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.1440781440781445, | |
| "eval_accuracy": 0.5436893203883495, | |
| "eval_loss": 1.355545163154602, | |
| "eval_runtime": 21.2157, | |
| "eval_samples_per_second": 9.71, | |
| "eval_steps_per_second": 1.226, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.012210012210012, | |
| "grad_norm": 7.593497276306152, | |
| "learning_rate": 1.485753052917232e-05, | |
| "loss": 0.2075, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.024420024420024, | |
| "grad_norm": 4.697848320007324, | |
| "learning_rate": 1.417910447761194e-05, | |
| "loss": 0.2499, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0366300366300365, | |
| "grad_norm": 5.646294116973877, | |
| "learning_rate": 1.3500678426051561e-05, | |
| "loss": 0.1955, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.048840048840049, | |
| "grad_norm": 1.646572232246399, | |
| "learning_rate": 1.282225237449118e-05, | |
| "loss": 0.2132, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.061050061050061, | |
| "grad_norm": 13.153250694274902, | |
| "learning_rate": 1.2143826322930801e-05, | |
| "loss": 0.2291, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.073260073260073, | |
| "grad_norm": 3.749263286590576, | |
| "learning_rate": 1.1465400271370422e-05, | |
| "loss": 0.2316, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.085470085470085, | |
| "grad_norm": 1.2367647886276245, | |
| "learning_rate": 1.0786974219810041e-05, | |
| "loss": 0.1593, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.097680097680097, | |
| "grad_norm": 14.28999137878418, | |
| "learning_rate": 1.010854816824966e-05, | |
| "loss": 0.1864, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.1098901098901095, | |
| "grad_norm": 4.065025329589844, | |
| "learning_rate": 9.430122116689281e-06, | |
| "loss": 0.261, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.122100122100122, | |
| "grad_norm": 5.471700668334961, | |
| "learning_rate": 8.751696065128902e-06, | |
| "loss": 0.1759, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.134310134310135, | |
| "grad_norm": 1.2977887392044067, | |
| "learning_rate": 8.073270013568522e-06, | |
| "loss": 0.2201, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.1440781440781445, | |
| "eval_accuracy": 0.5776699029126213, | |
| "eval_loss": 1.3875343799591064, | |
| "eval_runtime": 33.2439, | |
| "eval_samples_per_second": 6.197, | |
| "eval_steps_per_second": 0.782, | |
| "step": 708 | |
| }, | |
| { | |
| "epoch": 6.002442002442002, | |
| "grad_norm": 5.552870750427246, | |
| "learning_rate": 7.394843962008141e-06, | |
| "loss": 0.2692, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 6.014652014652015, | |
| "grad_norm": 20.285839080810547, | |
| "learning_rate": 6.716417910447762e-06, | |
| "loss": 0.1318, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 6.026862026862027, | |
| "grad_norm": 1.199399471282959, | |
| "learning_rate": 6.037991858887382e-06, | |
| "loss": 0.0786, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 6.039072039072039, | |
| "grad_norm": 0.7117233872413635, | |
| "learning_rate": 5.359565807327002e-06, | |
| "loss": 0.0873, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.051282051282051, | |
| "grad_norm": 1.9136446714401245, | |
| "learning_rate": 4.681139755766622e-06, | |
| "loss": 0.1112, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.063492063492063, | |
| "grad_norm": 3.076906204223633, | |
| "learning_rate": 4.002713704206242e-06, | |
| "loss": 0.1026, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.075702075702075, | |
| "grad_norm": 26.244754791259766, | |
| "learning_rate": 3.324287652645862e-06, | |
| "loss": 0.1529, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.087912087912088, | |
| "grad_norm": 0.7138678431510925, | |
| "learning_rate": 2.645861601085482e-06, | |
| "loss": 0.1352, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.1001221001221, | |
| "grad_norm": 10.830814361572266, | |
| "learning_rate": 1.967435549525102e-06, | |
| "loss": 0.0969, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 6.112332112332112, | |
| "grad_norm": 5.247445106506348, | |
| "learning_rate": 1.289009497964722e-06, | |
| "loss": 0.1702, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.124542124542124, | |
| "grad_norm": 1.4733346700668335, | |
| "learning_rate": 6.10583446404342e-07, | |
| "loss": 0.1171, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.135531135531136, | |
| "eval_accuracy": 0.6019417475728155, | |
| "eval_loss": 1.3527742624282837, | |
| "eval_runtime": 23.7516, | |
| "eval_samples_per_second": 8.673, | |
| "eval_steps_per_second": 1.095, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 6.135531135531136, | |
| "step": 819, | |
| "total_flos": 8.149698472747991e+18, | |
| "train_loss": 0.8887154050216861, | |
| "train_runtime": 1543.5432, | |
| "train_samples_per_second": 4.245, | |
| "train_steps_per_second": 0.531 | |
| }, | |
| { | |
| "epoch": 6.135531135531136, | |
| "eval_accuracy": 0.5756097560975609, | |
| "eval_loss": 1.497787594795227, | |
| "eval_runtime": 33.2888, | |
| "eval_samples_per_second": 6.158, | |
| "eval_steps_per_second": 0.781, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 6.135531135531136, | |
| "eval_accuracy": 0.5756097560975609, | |
| "eval_loss": 1.497787356376648, | |
| "eval_runtime": 33.1995, | |
| "eval_samples_per_second": 6.175, | |
| "eval_steps_per_second": 0.783, | |
| "step": 819 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 819, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.149698472747991e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |