| { | |
| "best_metric": 0.85, | |
| "best_model_checkpoint": "videomae-base-finetuned-engine-subset-20230310/checkpoint-372", | |
| "epoch": 19.018333333333334, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 2.8307, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.6132, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.5947, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_accuracy": 0.15, | |
| "eval_loss": 2.538285493850708, | |
| "eval_runtime": 64.1944, | |
| "eval_samples_per_second": 1.246, | |
| "eval_steps_per_second": 0.218, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.5062, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 2.665, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 5e-05, | |
| "loss": 2.4195, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_accuracy": 0.15, | |
| "eval_loss": 2.5108141899108887, | |
| "eval_runtime": 62.8569, | |
| "eval_samples_per_second": 1.273, | |
| "eval_steps_per_second": 0.223, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 2.3539, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 2.3374, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 2.2476, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_accuracy": 0.225, | |
| "eval_loss": 2.0532896518707275, | |
| "eval_runtime": 62.7732, | |
| "eval_samples_per_second": 1.274, | |
| "eval_steps_per_second": 0.223, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 2.2757, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 1.9768, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 1.9449, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "eval_accuracy": 0.2375, | |
| "eval_loss": 2.071887493133545, | |
| "eval_runtime": 65.8713, | |
| "eval_samples_per_second": 1.214, | |
| "eval_steps_per_second": 0.213, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 1.7178, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 1.6786, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.5724, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "eval_accuracy": 0.475, | |
| "eval_loss": 1.4755998849868774, | |
| "eval_runtime": 62.826, | |
| "eval_samples_per_second": 1.273, | |
| "eval_steps_per_second": 0.223, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 1.3465, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 1.3641, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 1.395, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.2884117364883423, | |
| "eval_runtime": 63.6279, | |
| "eval_samples_per_second": 1.257, | |
| "eval_steps_per_second": 0.22, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 1.1859, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 1.2224, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 1.0822, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_accuracy": 0.575, | |
| "eval_loss": 1.0678651332855225, | |
| "eval_runtime": 63.3472, | |
| "eval_samples_per_second": 1.263, | |
| "eval_steps_per_second": 0.221, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.9066, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 0.7939, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.0635, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.8040415048599243, | |
| "eval_runtime": 63.8719, | |
| "eval_samples_per_second": 1.253, | |
| "eval_steps_per_second": 0.219, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 1.1727, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.8291, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.8707, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "eval_accuracy": 0.525, | |
| "eval_loss": 0.9334062337875366, | |
| "eval_runtime": 64.0172, | |
| "eval_samples_per_second": 1.25, | |
| "eval_steps_per_second": 0.219, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.8339, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 0.8118, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.8068, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.7042, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.6476640701293945, | |
| "eval_runtime": 63.3755, | |
| "eval_samples_per_second": 1.262, | |
| "eval_steps_per_second": 0.221, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.6771, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7369, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.6543, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "eval_accuracy": 0.7375, | |
| "eval_loss": 0.6962689757347107, | |
| "eval_runtime": 63.3385, | |
| "eval_samples_per_second": 1.263, | |
| "eval_steps_per_second": 0.221, | |
| "step": 341 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.7641, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.5571, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 0.6807, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.49579018354415894, | |
| "eval_runtime": 64.8171, | |
| "eval_samples_per_second": 1.234, | |
| "eval_steps_per_second": 0.216, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.5435, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.5266, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.4924, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "eval_accuracy": 0.775, | |
| "eval_loss": 0.6373826265335083, | |
| "eval_runtime": 67.4155, | |
| "eval_samples_per_second": 1.187, | |
| "eval_steps_per_second": 0.208, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.4775, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.5071, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 0.4822, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.6144760847091675, | |
| "eval_runtime": 64.8207, | |
| "eval_samples_per_second": 1.234, | |
| "eval_steps_per_second": 0.216, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.3259, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.5054, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.4878, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "eval_accuracy": 0.7625, | |
| "eval_loss": 0.6274302005767822, | |
| "eval_runtime": 62.4581, | |
| "eval_samples_per_second": 1.281, | |
| "eval_steps_per_second": 0.224, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 0.3902, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.4728, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 0.4442, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.42305102944374084, | |
| "eval_runtime": 62.8623, | |
| "eval_samples_per_second": 1.273, | |
| "eval_steps_per_second": 0.223, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.443, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.3427, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.2739, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 16.05, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.4999323785305023, | |
| "eval_runtime": 63.9597, | |
| "eval_samples_per_second": 1.251, | |
| "eval_steps_per_second": 0.219, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 0.3676, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.5171, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.3514, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "eval_accuracy": 0.8375, | |
| "eval_loss": 0.4638718068599701, | |
| "eval_runtime": 65.0616, | |
| "eval_samples_per_second": 1.23, | |
| "eval_steps_per_second": 0.215, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.2877, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.3558, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 18.04, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.4158, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 18.05, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.42912358045578003, | |
| "eval_runtime": 64.0669, | |
| "eval_samples_per_second": 1.249, | |
| "eval_steps_per_second": 0.219, | |
| "step": 589 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 0.3158, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "learning_rate": 0.0, | |
| "loss": 0.2689, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.42943257093429565, | |
| "eval_runtime": 64.3343, | |
| "eval_samples_per_second": 1.244, | |
| "eval_steps_per_second": 0.218, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "step": 600, | |
| "total_flos": 4.415279302910214e+18, | |
| "train_loss": 1.063151851495107, | |
| "train_runtime": 5413.8579, | |
| "train_samples_per_second": 0.665, | |
| "train_steps_per_second": 0.111 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.49579018354415894, | |
| "eval_runtime": 66.0758, | |
| "eval_samples_per_second": 1.211, | |
| "eval_steps_per_second": 0.212, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.49579015374183655, | |
| "eval_runtime": 64.7531, | |
| "eval_samples_per_second": 1.235, | |
| "eval_steps_per_second": 0.216, | |
| "step": 600 | |
| } | |
| ], | |
| "max_steps": 600, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 4.415279302910214e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |