| { | |
| "best_metric": 0.9432314410480349, | |
| "best_model_checkpoint": "videomae-base-finetuned-good-gesturePhaseV12/checkpoint-630", | |
| "epoch": 4.193650793650794, | |
| "eval_steps": 500, | |
| "global_step": 630, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.015873015873015872, | |
| "grad_norm": 9.435199737548828, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 1.3901, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.031746031746031744, | |
| "grad_norm": 4.234468936920166, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 1.0891, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.047619047619047616, | |
| "grad_norm": 4.78691291809082, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.933, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06349206349206349, | |
| "grad_norm": 5.06723690032959, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.902, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07936507936507936, | |
| "grad_norm": 3.5777130126953125, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.9257, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 5.7673845291137695, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 1.0482, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 2.839975118637085, | |
| "learning_rate": 4.938271604938271e-05, | |
| "loss": 0.8308, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12698412698412698, | |
| "grad_norm": 6.74424934387207, | |
| "learning_rate": 4.850088183421517e-05, | |
| "loss": 0.9529, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 3.8246517181396484, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.9361, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.15873015873015872, | |
| "grad_norm": 3.8471875190734863, | |
| "learning_rate": 4.673721340388007e-05, | |
| "loss": 0.8691, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1746031746031746, | |
| "grad_norm": 6.3369975090026855, | |
| "learning_rate": 4.585537918871252e-05, | |
| "loss": 0.9276, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 3.3979837894439697, | |
| "learning_rate": 4.4973544973544974e-05, | |
| "loss": 1.1508, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2015873015873016, | |
| "eval_accuracy": 0.6899563318777293, | |
| "eval_accuracy_hold": 0.0, | |
| "eval_accuracy_preparation": 1.0, | |
| "eval_accuracy_recovery": 0.0, | |
| "eval_accuracy_stroke": 0.0, | |
| "eval_accuracy_unknown": 0.0, | |
| "eval_loss": 1.0099235773086548, | |
| "eval_runtime": 40.2895, | |
| "eval_samples_per_second": 5.684, | |
| "eval_steps_per_second": 0.72, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 1.0047619047619047, | |
| "grad_norm": 1.5407054424285889, | |
| "learning_rate": 4.409171075837743e-05, | |
| "loss": 0.9457, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0206349206349206, | |
| "grad_norm": 6.706632614135742, | |
| "learning_rate": 4.3209876543209875e-05, | |
| "loss": 0.9163, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0365079365079366, | |
| "grad_norm": 2.2654550075531006, | |
| "learning_rate": 4.232804232804233e-05, | |
| "loss": 1.007, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0523809523809524, | |
| "grad_norm": 2.78928279876709, | |
| "learning_rate": 4.144620811287478e-05, | |
| "loss": 0.8156, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0682539682539682, | |
| "grad_norm": 8.601720809936523, | |
| "learning_rate": 4.056437389770723e-05, | |
| "loss": 0.9077, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.084126984126984, | |
| "grad_norm": 5.783897876739502, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.7815, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.842205286026001, | |
| "learning_rate": 3.8800705467372136e-05, | |
| "loss": 0.9623, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.115873015873016, | |
| "grad_norm": 4.1139302253723145, | |
| "learning_rate": 3.791887125220458e-05, | |
| "loss": 0.7145, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1317460317460317, | |
| "grad_norm": 6.981729507446289, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.8596, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1476190476190475, | |
| "grad_norm": 6.828718185424805, | |
| "learning_rate": 3.615520282186949e-05, | |
| "loss": 0.7746, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.1634920634920636, | |
| "grad_norm": 4.089468955993652, | |
| "learning_rate": 3.527336860670194e-05, | |
| "loss": 0.6024, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.1793650793650794, | |
| "grad_norm": 15.972381591796875, | |
| "learning_rate": 3.439153439153439e-05, | |
| "loss": 0.6134, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.1952380952380952, | |
| "grad_norm": 4.078704833984375, | |
| "learning_rate": 3.3509700176366844e-05, | |
| "loss": 0.7497, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2015873015873015, | |
| "eval_accuracy": 0.7248908296943232, | |
| "eval_accuracy_hold": 0.3076923076923077, | |
| "eval_accuracy_preparation": 1.0, | |
| "eval_accuracy_recovery": 0.0, | |
| "eval_accuracy_stroke": 0.0, | |
| "eval_accuracy_unknown": 0.0, | |
| "eval_loss": 0.7447668313980103, | |
| "eval_runtime": 21.0507, | |
| "eval_samples_per_second": 10.879, | |
| "eval_steps_per_second": 1.378, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 2.0095238095238095, | |
| "grad_norm": 2.710218667984009, | |
| "learning_rate": 3.262786596119929e-05, | |
| "loss": 0.6647, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0253968253968253, | |
| "grad_norm": 17.5618839263916, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.7224, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.041269841269841, | |
| "grad_norm": 20.599666595458984, | |
| "learning_rate": 3.08641975308642e-05, | |
| "loss": 0.6155, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.057142857142857, | |
| "grad_norm": 2.6241402626037598, | |
| "learning_rate": 2.998236331569665e-05, | |
| "loss": 0.5442, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.073015873015873, | |
| "grad_norm": 7.473824501037598, | |
| "learning_rate": 2.91005291005291e-05, | |
| "loss": 0.4203, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.088888888888889, | |
| "grad_norm": 2.051452398300171, | |
| "learning_rate": 2.8218694885361552e-05, | |
| "loss": 0.6926, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.104761904761905, | |
| "grad_norm": 3.222371816635132, | |
| "learning_rate": 2.7336860670194003e-05, | |
| "loss": 0.5005, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.1206349206349207, | |
| "grad_norm": 3.2319962978363037, | |
| "learning_rate": 2.6455026455026456e-05, | |
| "loss": 0.3705, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.1365079365079365, | |
| "grad_norm": 19.56564712524414, | |
| "learning_rate": 2.5573192239858906e-05, | |
| "loss": 0.469, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.1523809523809523, | |
| "grad_norm": 27.630725860595703, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 0.3801, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.168253968253968, | |
| "grad_norm": 3.2996511459350586, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.6634, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.1841269841269844, | |
| "grad_norm": 3.424922227859497, | |
| "learning_rate": 2.292768959435626e-05, | |
| "loss": 0.4151, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 1.88832688331604, | |
| "learning_rate": 2.2045855379188714e-05, | |
| "loss": 0.3044, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.2015873015873018, | |
| "eval_accuracy": 0.8602620087336245, | |
| "eval_accuracy_hold": 0.7692307692307693, | |
| "eval_accuracy_preparation": 0.9620253164556962, | |
| "eval_accuracy_recovery": 0.5882352941176471, | |
| "eval_accuracy_stroke": 0.0, | |
| "eval_accuracy_unknown": 0.6818181818181818, | |
| "eval_loss": 0.4170438051223755, | |
| "eval_runtime": 20.4849, | |
| "eval_samples_per_second": 11.179, | |
| "eval_steps_per_second": 1.416, | |
| "step": 381 | |
| }, | |
| { | |
| "epoch": 3.0142857142857142, | |
| "grad_norm": 1.2836856842041016, | |
| "learning_rate": 2.1164021164021164e-05, | |
| "loss": 0.2214, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.03015873015873, | |
| "grad_norm": 3.735811233520508, | |
| "learning_rate": 2.0282186948853614e-05, | |
| "loss": 0.5429, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.046031746031746, | |
| "grad_norm": 1.2549328804016113, | |
| "learning_rate": 1.9400352733686068e-05, | |
| "loss": 0.4081, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.0619047619047617, | |
| "grad_norm": 0.5135648846626282, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.211, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.077777777777778, | |
| "grad_norm": 12.976859092712402, | |
| "learning_rate": 1.763668430335097e-05, | |
| "loss": 0.2547, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.0936507936507938, | |
| "grad_norm": 8.389816284179688, | |
| "learning_rate": 1.6754850088183422e-05, | |
| "loss": 0.3457, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.1095238095238096, | |
| "grad_norm": 3.8119888305664062, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.3689, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.1253968253968254, | |
| "grad_norm": 10.705780029296875, | |
| "learning_rate": 1.4991181657848324e-05, | |
| "loss": 0.1528, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.141269841269841, | |
| "grad_norm": 0.21885545551776886, | |
| "learning_rate": 1.4109347442680776e-05, | |
| "loss": 0.1199, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.157142857142857, | |
| "grad_norm": 8.286587715148926, | |
| "learning_rate": 1.3227513227513228e-05, | |
| "loss": 0.2536, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.173015873015873, | |
| "grad_norm": 55.86492156982422, | |
| "learning_rate": 1.2345679012345678e-05, | |
| "loss": 0.2634, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.188888888888889, | |
| "grad_norm": 3.395249366760254, | |
| "learning_rate": 1.146384479717813e-05, | |
| "loss": 0.1617, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.2015873015873018, | |
| "eval_accuracy": 0.9126637554585153, | |
| "eval_accuracy_hold": 0.7307692307692307, | |
| "eval_accuracy_preparation": 0.9810126582278481, | |
| "eval_accuracy_recovery": 0.8823529411764706, | |
| "eval_accuracy_stroke": 0.16666666666666666, | |
| "eval_accuracy_unknown": 0.8636363636363636, | |
| "eval_loss": 0.3626626133918762, | |
| "eval_runtime": 17.8375, | |
| "eval_samples_per_second": 12.838, | |
| "eval_steps_per_second": 1.626, | |
| "step": 508 | |
| }, | |
| { | |
| "epoch": 4.003174603174603, | |
| "grad_norm": 1.1130430698394775, | |
| "learning_rate": 1.0582010582010582e-05, | |
| "loss": 0.0515, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.019047619047619, | |
| "grad_norm": 2.0611793994903564, | |
| "learning_rate": 9.700176366843034e-06, | |
| "loss": 0.1072, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.034920634920635, | |
| "grad_norm": 6.3466410636901855, | |
| "learning_rate": 8.818342151675484e-06, | |
| "loss": 0.3303, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.050793650793651, | |
| "grad_norm": 2.479144811630249, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.1395, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.066666666666666, | |
| "grad_norm": 9.980076789855957, | |
| "learning_rate": 7.054673721340388e-06, | |
| "loss": 0.2201, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.082539682539682, | |
| "grad_norm": 0.09916899353265762, | |
| "learning_rate": 6.172839506172839e-06, | |
| "loss": 0.1523, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.098412698412698, | |
| "grad_norm": 0.09749830514192581, | |
| "learning_rate": 5.291005291005291e-06, | |
| "loss": 0.1487, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.114285714285714, | |
| "grad_norm": 21.505535125732422, | |
| "learning_rate": 4.409171075837742e-06, | |
| "loss": 0.1248, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.130158730158731, | |
| "grad_norm": 0.08216376602649689, | |
| "learning_rate": 3.527336860670194e-06, | |
| "loss": 0.1414, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.146031746031746, | |
| "grad_norm": 0.6915600895881653, | |
| "learning_rate": 2.6455026455026455e-06, | |
| "loss": 0.0609, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.161904761904762, | |
| "grad_norm": 0.09014752507209778, | |
| "learning_rate": 1.763668430335097e-06, | |
| "loss": 0.1159, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.177777777777778, | |
| "grad_norm": 0.4283609688282013, | |
| "learning_rate": 8.818342151675485e-07, | |
| "loss": 0.0382, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.193650793650794, | |
| "grad_norm": 0.2592800259590149, | |
| "learning_rate": 0.0, | |
| "loss": 0.0765, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.193650793650794, | |
| "eval_accuracy": 0.9432314410480349, | |
| "eval_accuracy_hold": 0.8461538461538461, | |
| "eval_accuracy_preparation": 0.9746835443037974, | |
| "eval_accuracy_recovery": 0.8823529411764706, | |
| "eval_accuracy_stroke": 0.6666666666666666, | |
| "eval_accuracy_unknown": 0.9545454545454546, | |
| "eval_loss": 0.21748152375221252, | |
| "eval_runtime": 17.4057, | |
| "eval_samples_per_second": 13.157, | |
| "eval_steps_per_second": 1.666, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.193650793650794, | |
| "step": 630, | |
| "total_flos": 6.275349161424323e+18, | |
| "train_loss": 0.5457079762504214, | |
| "train_runtime": 1399.0814, | |
| "train_samples_per_second": 3.602, | |
| "train_steps_per_second": 0.45 | |
| }, | |
| { | |
| "epoch": 4.193650793650794, | |
| "eval_accuracy": 0.9543568464730291, | |
| "eval_accuracy_hold": 1.0, | |
| "eval_accuracy_preparation": 0.9811320754716981, | |
| "eval_accuracy_recovery": 0.8947368421052632, | |
| "eval_accuracy_stroke": 0.42857142857142855, | |
| "eval_accuracy_unknown": 0.9285714285714286, | |
| "eval_loss": 0.24868811666965485, | |
| "eval_runtime": 21.852, | |
| "eval_samples_per_second": 11.029, | |
| "eval_steps_per_second": 1.419, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.193650793650794, | |
| "eval_accuracy": 0.9543568464730291, | |
| "eval_accuracy_hold": 1.0, | |
| "eval_accuracy_preparation": 0.9811320754716981, | |
| "eval_accuracy_recovery": 0.8947368421052632, | |
| "eval_accuracy_stroke": 0.42857142857142855, | |
| "eval_accuracy_unknown": 0.9285714285714286, | |
| "eval_loss": 0.24868811666965485, | |
| "eval_runtime": 18.5601, | |
| "eval_samples_per_second": 12.985, | |
| "eval_steps_per_second": 1.67, | |
| "step": 630 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 630, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.275349161424323e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |