| { | |
| "best_metric": 0.6067961165048543, | |
| "best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-right-hand-conflab-baseline-2/checkpoint-590", | |
| "epoch": 12.05042735042735, | |
| "eval_steps": 500, | |
| "global_step": 767, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008547008547008548, | |
| "grad_norm": 5.778216361999512, | |
| "learning_rate": 4.273504273504274e-06, | |
| "loss": 1.8458, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.017094017094017096, | |
| "grad_norm": 7.9287519454956055, | |
| "learning_rate": 8.547008547008548e-06, | |
| "loss": 1.8477, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02564102564102564, | |
| "grad_norm": 8.069726943969727, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 1.819, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03418803418803419, | |
| "grad_norm": 8.647719383239746, | |
| "learning_rate": 1.7094017094017095e-05, | |
| "loss": 1.7469, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.042735042735042736, | |
| "grad_norm": 9.172955513000488, | |
| "learning_rate": 2.1367521367521368e-05, | |
| "loss": 1.7757, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05042735042735043, | |
| "eval_accuracy": 0.33980582524271846, | |
| "eval_loss": 1.7543749809265137, | |
| "eval_runtime": 39.6507, | |
| "eval_samples_per_second": 5.195, | |
| "eval_steps_per_second": 0.328, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.0008547008547009, | |
| "grad_norm": 7.786515235900879, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 1.7826, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0094017094017094, | |
| "grad_norm": 8.755845069885254, | |
| "learning_rate": 2.9914529914529915e-05, | |
| "loss": 1.6611, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0179487179487179, | |
| "grad_norm": 8.502706527709961, | |
| "learning_rate": 3.418803418803419e-05, | |
| "loss": 1.7393, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0264957264957264, | |
| "grad_norm": 7.265880107879639, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.5471, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0350427350427351, | |
| "grad_norm": 9.460399627685547, | |
| "learning_rate": 4.2735042735042735e-05, | |
| "loss": 1.6016, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0435897435897437, | |
| "grad_norm": 9.125144958496094, | |
| "learning_rate": 4.700854700854701e-05, | |
| "loss": 1.4807, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0504273504273505, | |
| "eval_accuracy": 0.36893203883495146, | |
| "eval_loss": 1.706697940826416, | |
| "eval_runtime": 14.918, | |
| "eval_samples_per_second": 13.809, | |
| "eval_steps_per_second": 0.871, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 2.0017094017094017, | |
| "grad_norm": 7.854036331176758, | |
| "learning_rate": 4.985754985754986e-05, | |
| "loss": 1.692, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.01025641025641, | |
| "grad_norm": 7.147447109222412, | |
| "learning_rate": 4.938271604938271e-05, | |
| "loss": 1.3991, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0188034188034187, | |
| "grad_norm": 11.069567680358887, | |
| "learning_rate": 4.890788224121557e-05, | |
| "loss": 1.4116, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.0273504273504273, | |
| "grad_norm": 6.93538236618042, | |
| "learning_rate": 4.8433048433048433e-05, | |
| "loss": 1.485, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0358974358974358, | |
| "grad_norm": 7.68374490737915, | |
| "learning_rate": 4.7958214624881294e-05, | |
| "loss": 1.3291, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0444444444444443, | |
| "grad_norm": 7.971122741699219, | |
| "learning_rate": 4.7483380816714154e-05, | |
| "loss": 1.4444, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0504273504273502, | |
| "eval_accuracy": 0.4320388349514563, | |
| "eval_loss": 1.4684412479400635, | |
| "eval_runtime": 13.0884, | |
| "eval_samples_per_second": 15.739, | |
| "eval_steps_per_second": 0.993, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 3.0025641025641026, | |
| "grad_norm": 7.5584797859191895, | |
| "learning_rate": 4.700854700854701e-05, | |
| "loss": 1.3167, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.011111111111111, | |
| "grad_norm": 9.996468544006348, | |
| "learning_rate": 4.653371320037987e-05, | |
| "loss": 1.2033, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.0196581196581196, | |
| "grad_norm": 8.742140769958496, | |
| "learning_rate": 4.605887939221273e-05, | |
| "loss": 1.0363, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.028205128205128, | |
| "grad_norm": 10.190289497375488, | |
| "learning_rate": 4.558404558404559e-05, | |
| "loss": 1.118, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.0367521367521366, | |
| "grad_norm": 10.13549518585205, | |
| "learning_rate": 4.510921177587845e-05, | |
| "loss": 1.1718, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.045299145299145, | |
| "grad_norm": 11.223580360412598, | |
| "learning_rate": 4.463437796771131e-05, | |
| "loss": 1.022, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.0504273504273502, | |
| "eval_accuracy": 0.5388349514563107, | |
| "eval_loss": 1.432210087776184, | |
| "eval_runtime": 13.2983, | |
| "eval_samples_per_second": 15.491, | |
| "eval_steps_per_second": 0.978, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 4.003418803418803, | |
| "grad_norm": 9.429991722106934, | |
| "learning_rate": 4.415954415954416e-05, | |
| "loss": 1.0957, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.011965811965812, | |
| "grad_norm": 10.795949935913086, | |
| "learning_rate": 4.368471035137702e-05, | |
| "loss": 0.8767, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.02051282051282, | |
| "grad_norm": 8.301912307739258, | |
| "learning_rate": 4.3209876543209875e-05, | |
| "loss": 0.7676, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.029059829059829, | |
| "grad_norm": 11.052982330322266, | |
| "learning_rate": 4.2735042735042735e-05, | |
| "loss": 0.7545, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.0376068376068375, | |
| "grad_norm": 9.92408275604248, | |
| "learning_rate": 4.2260208926875595e-05, | |
| "loss": 0.9624, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.046153846153846, | |
| "grad_norm": 5.574217319488525, | |
| "learning_rate": 4.1785375118708455e-05, | |
| "loss": 0.7796, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.05042735042735, | |
| "eval_accuracy": 0.558252427184466, | |
| "eval_loss": 1.243977427482605, | |
| "eval_runtime": 14.2794, | |
| "eval_samples_per_second": 14.426, | |
| "eval_steps_per_second": 0.91, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 5.004273504273504, | |
| "grad_norm": 9.098281860351562, | |
| "learning_rate": 4.131054131054131e-05, | |
| "loss": 0.7113, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.012820512820513, | |
| "grad_norm": 8.787925720214844, | |
| "learning_rate": 4.083570750237417e-05, | |
| "loss": 0.6767, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.021367521367521, | |
| "grad_norm": 12.598801612854004, | |
| "learning_rate": 4.036087369420703e-05, | |
| "loss": 0.5864, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.02991452991453, | |
| "grad_norm": 11.910761833190918, | |
| "learning_rate": 3.988603988603989e-05, | |
| "loss": 0.5178, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.038461538461538, | |
| "grad_norm": 6.0374369621276855, | |
| "learning_rate": 3.941120607787275e-05, | |
| "loss": 0.5098, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.047008547008547, | |
| "grad_norm": 11.50935173034668, | |
| "learning_rate": 3.893637226970561e-05, | |
| "loss": 0.5612, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.05042735042735, | |
| "eval_accuracy": 0.5679611650485437, | |
| "eval_loss": 1.3034228086471558, | |
| "eval_runtime": 14.2513, | |
| "eval_samples_per_second": 14.455, | |
| "eval_steps_per_second": 0.912, | |
| "step": 354 | |
| }, | |
| { | |
| "epoch": 6.005128205128205, | |
| "grad_norm": 8.892311096191406, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.4646, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.013675213675214, | |
| "grad_norm": 7.864949703216553, | |
| "learning_rate": 3.798670465337132e-05, | |
| "loss": 0.419, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 6.022222222222222, | |
| "grad_norm": 7.739443302154541, | |
| "learning_rate": 3.7511870845204176e-05, | |
| "loss": 0.3893, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.030769230769231, | |
| "grad_norm": 5.160546779632568, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.4071, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.039316239316239, | |
| "grad_norm": 11.32081413269043, | |
| "learning_rate": 3.65622032288699e-05, | |
| "loss": 0.3748, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.047863247863248, | |
| "grad_norm": 8.12364673614502, | |
| "learning_rate": 3.608736942070276e-05, | |
| "loss": 0.4722, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.05042735042735, | |
| "eval_accuracy": 0.5728155339805825, | |
| "eval_loss": 1.4142879247665405, | |
| "eval_runtime": 14.4857, | |
| "eval_samples_per_second": 14.221, | |
| "eval_steps_per_second": 0.897, | |
| "step": 413 | |
| }, | |
| { | |
| "epoch": 7.005982905982906, | |
| "grad_norm": 9.602105140686035, | |
| "learning_rate": 3.561253561253561e-05, | |
| "loss": 0.335, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 7.0145299145299145, | |
| "grad_norm": 6.589150905609131, | |
| "learning_rate": 3.513770180436847e-05, | |
| "loss": 0.3204, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 7.023076923076923, | |
| "grad_norm": 11.828797340393066, | |
| "learning_rate": 3.466286799620133e-05, | |
| "loss": 0.2932, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.0316239316239315, | |
| "grad_norm": 6.758438587188721, | |
| "learning_rate": 3.418803418803419e-05, | |
| "loss": 0.296, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.04017094017094, | |
| "grad_norm": 6.294919013977051, | |
| "learning_rate": 3.371320037986705e-05, | |
| "loss": 0.41, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.0487179487179485, | |
| "grad_norm": 12.272725105285645, | |
| "learning_rate": 3.323836657169991e-05, | |
| "loss": 0.3215, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.05042735042735, | |
| "eval_accuracy": 0.5825242718446602, | |
| "eval_loss": 1.3449476957321167, | |
| "eval_runtime": 14.8147, | |
| "eval_samples_per_second": 13.905, | |
| "eval_steps_per_second": 0.878, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 8.006837606837607, | |
| "grad_norm": 4.883133888244629, | |
| "learning_rate": 3.2763532763532764e-05, | |
| "loss": 0.1878, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 8.015384615384615, | |
| "grad_norm": 4.400292873382568, | |
| "learning_rate": 3.2288698955365625e-05, | |
| "loss": 0.1816, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 8.023931623931624, | |
| "grad_norm": 7.798640251159668, | |
| "learning_rate": 3.181386514719848e-05, | |
| "loss": 0.2256, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.032478632478632, | |
| "grad_norm": 9.348875999450684, | |
| "learning_rate": 3.133903133903134e-05, | |
| "loss": 0.1961, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.04102564102564, | |
| "grad_norm": 9.914623260498047, | |
| "learning_rate": 3.08641975308642e-05, | |
| "loss": 0.2161, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.04957264957265, | |
| "grad_norm": 4.433971881866455, | |
| "learning_rate": 3.0389363722697055e-05, | |
| "loss": 0.2223, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.05042735042735, | |
| "eval_accuracy": 0.6019417475728155, | |
| "eval_loss": 1.3281605243682861, | |
| "eval_runtime": 13.2927, | |
| "eval_samples_per_second": 15.497, | |
| "eval_steps_per_second": 0.978, | |
| "step": 531 | |
| }, | |
| { | |
| "epoch": 9.007692307692308, | |
| "grad_norm": 2.7526516914367676, | |
| "learning_rate": 2.9914529914529915e-05, | |
| "loss": 0.2092, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 9.016239316239316, | |
| "grad_norm": 6.128594875335693, | |
| "learning_rate": 2.9439696106362775e-05, | |
| "loss": 0.1576, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 9.024786324786325, | |
| "grad_norm": 5.773240089416504, | |
| "learning_rate": 2.8964862298195632e-05, | |
| "loss": 0.1468, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 9.033333333333333, | |
| "grad_norm": 8.840475082397461, | |
| "learning_rate": 2.8490028490028492e-05, | |
| "loss": 0.1607, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 9.041880341880342, | |
| "grad_norm": 7.966043472290039, | |
| "learning_rate": 2.8015194681861352e-05, | |
| "loss": 0.1912, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.05042735042735, | |
| "grad_norm": 8.73734188079834, | |
| "learning_rate": 2.754036087369421e-05, | |
| "loss": 0.1452, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.05042735042735, | |
| "eval_accuracy": 0.6067961165048543, | |
| "eval_loss": 1.4134429693222046, | |
| "eval_runtime": 13.2803, | |
| "eval_samples_per_second": 15.512, | |
| "eval_steps_per_second": 0.979, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 10.008547008547009, | |
| "grad_norm": 15.796429634094238, | |
| "learning_rate": 2.706552706552707e-05, | |
| "loss": 0.1742, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 10.017094017094017, | |
| "grad_norm": 4.9247260093688965, | |
| "learning_rate": 2.6590693257359926e-05, | |
| "loss": 0.0984, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 10.025641025641026, | |
| "grad_norm": 9.770017623901367, | |
| "learning_rate": 2.611585944919278e-05, | |
| "loss": 0.082, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 10.034188034188034, | |
| "grad_norm": 0.7237369418144226, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 0.125, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 10.042735042735043, | |
| "grad_norm": 2.9003889560699463, | |
| "learning_rate": 2.51661918328585e-05, | |
| "loss": 0.1771, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 10.05042735042735, | |
| "eval_accuracy": 0.6019417475728155, | |
| "eval_loss": 1.448883056640625, | |
| "eval_runtime": 13.1165, | |
| "eval_samples_per_second": 15.705, | |
| "eval_steps_per_second": 0.991, | |
| "step": 649 | |
| }, | |
| { | |
| "epoch": 11.0008547008547, | |
| "grad_norm": 6.064362525939941, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 0.1175, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 11.00940170940171, | |
| "grad_norm": 8.064191818237305, | |
| "learning_rate": 2.4216524216524217e-05, | |
| "loss": 0.0595, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 11.017948717948718, | |
| "grad_norm": 5.565966606140137, | |
| "learning_rate": 2.3741690408357077e-05, | |
| "loss": 0.0772, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 11.026495726495726, | |
| "grad_norm": 1.7832452058792114, | |
| "learning_rate": 2.3266856600189934e-05, | |
| "loss": 0.0472, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 11.035042735042735, | |
| "grad_norm": 8.308904647827148, | |
| "learning_rate": 2.2792022792022794e-05, | |
| "loss": 0.0673, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 11.043589743589743, | |
| "grad_norm": 0.9400803446769714, | |
| "learning_rate": 2.2317188983855654e-05, | |
| "loss": 0.0472, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 11.05042735042735, | |
| "eval_accuracy": 0.5922330097087378, | |
| "eval_loss": 1.5543662309646606, | |
| "eval_runtime": 14.0642, | |
| "eval_samples_per_second": 14.647, | |
| "eval_steps_per_second": 0.924, | |
| "step": 708 | |
| }, | |
| { | |
| "epoch": 12.001709401709402, | |
| "grad_norm": 2.194042921066284, | |
| "learning_rate": 2.184235517568851e-05, | |
| "loss": 0.1381, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 12.01025641025641, | |
| "grad_norm": 5.982657432556152, | |
| "learning_rate": 2.1367521367521368e-05, | |
| "loss": 0.0442, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 12.018803418803419, | |
| "grad_norm": 1.7736109495162964, | |
| "learning_rate": 2.0892687559354228e-05, | |
| "loss": 0.0246, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 12.027350427350427, | |
| "grad_norm": 0.1784723848104477, | |
| "learning_rate": 2.0417853751187084e-05, | |
| "loss": 0.0388, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 12.035897435897436, | |
| "grad_norm": 1.0902857780456543, | |
| "learning_rate": 1.9943019943019945e-05, | |
| "loss": 0.0207, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 12.044444444444444, | |
| "grad_norm": 3.468409299850464, | |
| "learning_rate": 1.9468186134852805e-05, | |
| "loss": 0.0384, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 12.05042735042735, | |
| "eval_accuracy": 0.587378640776699, | |
| "eval_loss": 1.6466865539550781, | |
| "eval_runtime": 13.2724, | |
| "eval_samples_per_second": 15.521, | |
| "eval_steps_per_second": 0.979, | |
| "step": 767 | |
| }, | |
| { | |
| "epoch": 12.05042735042735, | |
| "step": 767, | |
| "total_flos": 1.5260123470530871e+19, | |
| "train_loss": 0.6703926856390345, | |
| "train_runtime": 1815.9944, | |
| "train_samples_per_second": 10.308, | |
| "train_steps_per_second": 0.644 | |
| }, | |
| { | |
| "epoch": 12.05042735042735, | |
| "eval_accuracy": 0.5951219512195122, | |
| "eval_loss": 1.4525096416473389, | |
| "eval_runtime": 19.5792, | |
| "eval_samples_per_second": 10.47, | |
| "eval_steps_per_second": 0.664, | |
| "step": 767 | |
| }, | |
| { | |
| "epoch": 12.05042735042735, | |
| "eval_accuracy": 0.5951219512195122, | |
| "eval_loss": 1.4525096416473389, | |
| "eval_runtime": 22.7791, | |
| "eval_samples_per_second": 8.999, | |
| "eval_steps_per_second": 0.571, | |
| "step": 767 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1170, | |
| "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": 1.5260123470530871e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |