| { | |
| "best_metric": 0.8475336322869955, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-390", | |
| "epoch": 2.3282051282051284, | |
| "eval_steps": 500, | |
| "global_step": 390, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02564102564102564, | |
| "grad_norm": 9.303430557250977, | |
| "learning_rate": 6.41025641025641e-07, | |
| "loss": 0.191, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05128205128205128, | |
| "grad_norm": 17.580381393432617, | |
| "learning_rate": 1.282051282051282e-06, | |
| "loss": 0.1535, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 3.0345261096954346, | |
| "learning_rate": 1.9230769230769234e-06, | |
| "loss": 0.1446, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10256410256410256, | |
| "grad_norm": 5.7094550132751465, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 0.1564, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1282051282051282, | |
| "grad_norm": 1.9542657136917114, | |
| "learning_rate": 3.205128205128206e-06, | |
| "loss": 0.1486, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 8.122542381286621, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.1577, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1794871794871795, | |
| "grad_norm": 9.375791549682617, | |
| "learning_rate": 4.487179487179488e-06, | |
| "loss": 0.1942, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.20512820512820512, | |
| "grad_norm": 11.179394721984863, | |
| "learning_rate": 4.967948717948718e-06, | |
| "loss": 0.2429, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 11.36959457397461, | |
| "learning_rate": 4.807692307692308e-06, | |
| "loss": 0.1481, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2564102564102564, | |
| "grad_norm": 1.9242831468582153, | |
| "learning_rate": 4.647435897435898e-06, | |
| "loss": 0.1894, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28205128205128205, | |
| "grad_norm": 3.764237642288208, | |
| "learning_rate": 4.487179487179488e-06, | |
| "loss": 0.1557, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.1338346004486084, | |
| "learning_rate": 4.326923076923077e-06, | |
| "loss": 0.1373, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 1.327778935432434, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.1658, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.33589743589743587, | |
| "eval_accuracy": 0.8251121076233184, | |
| "eval_loss": 0.5804738998413086, | |
| "eval_runtime": 56.2864, | |
| "eval_samples_per_second": 3.962, | |
| "eval_steps_per_second": 0.497, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 1.023076923076923, | |
| "grad_norm": 0.7002243399620056, | |
| "learning_rate": 4.006410256410257e-06, | |
| "loss": 0.1178, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0487179487179488, | |
| "grad_norm": 1.249237298965454, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.1735, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0743589743589743, | |
| "grad_norm": 8.169507026672363, | |
| "learning_rate": 3.6858974358974363e-06, | |
| "loss": 0.1527, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 4.84148645401001, | |
| "learning_rate": 3.5256410256410263e-06, | |
| "loss": 0.1313, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1256410256410256, | |
| "grad_norm": 2.1306402683258057, | |
| "learning_rate": 3.365384615384616e-06, | |
| "loss": 0.0957, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1512820512820512, | |
| "grad_norm": 1.54770827293396, | |
| "learning_rate": 3.205128205128206e-06, | |
| "loss": 0.1137, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.176923076923077, | |
| "grad_norm": 13.978232383728027, | |
| "learning_rate": 3.044871794871795e-06, | |
| "loss": 0.1326, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.2025641025641025, | |
| "grad_norm": 4.250465393066406, | |
| "learning_rate": 2.8846153846153845e-06, | |
| "loss": 0.0616, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.2282051282051283, | |
| "grad_norm": 1.337228775024414, | |
| "learning_rate": 2.7243589743589744e-06, | |
| "loss": 0.0948, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.2538461538461538, | |
| "grad_norm": 11.519142150878906, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 0.0906, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2794871794871794, | |
| "grad_norm": 9.989028930664062, | |
| "learning_rate": 2.403846153846154e-06, | |
| "loss": 0.1384, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.3051282051282052, | |
| "grad_norm": 0.850464940071106, | |
| "learning_rate": 2.243589743589744e-06, | |
| "loss": 0.1329, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3307692307692307, | |
| "grad_norm": 2.935410976409912, | |
| "learning_rate": 2.0833333333333334e-06, | |
| "loss": 0.1039, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.3358974358974358, | |
| "eval_accuracy": 0.8251121076233184, | |
| "eval_loss": 0.5684685707092285, | |
| "eval_runtime": 60.4317, | |
| "eval_samples_per_second": 3.69, | |
| "eval_steps_per_second": 0.463, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 2.0205128205128204, | |
| "grad_norm": 2.507692575454712, | |
| "learning_rate": 1.9230769230769234e-06, | |
| "loss": 0.1267, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.046153846153846, | |
| "grad_norm": 9.046551704406738, | |
| "learning_rate": 1.7628205128205131e-06, | |
| "loss": 0.0909, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.071794871794872, | |
| "grad_norm": 1.059535264968872, | |
| "learning_rate": 1.602564102564103e-06, | |
| "loss": 0.1108, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0974358974358975, | |
| "grad_norm": 7.851165294647217, | |
| "learning_rate": 1.4423076923076922e-06, | |
| "loss": 0.0695, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.123076923076923, | |
| "grad_norm": 1.073171615600586, | |
| "learning_rate": 1.282051282051282e-06, | |
| "loss": 0.1031, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1487179487179486, | |
| "grad_norm": 2.39035964012146, | |
| "learning_rate": 1.121794871794872e-06, | |
| "loss": 0.115, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.174358974358974, | |
| "grad_norm": 1.7656793594360352, | |
| "learning_rate": 9.615384615384617e-07, | |
| "loss": 0.0919, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 19.47994613647461, | |
| "learning_rate": 8.012820512820515e-07, | |
| "loss": 0.1017, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.2256410256410257, | |
| "grad_norm": 14.330218315124512, | |
| "learning_rate": 6.41025641025641e-07, | |
| "loss": 0.0826, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.2512820512820513, | |
| "grad_norm": 3.4437031745910645, | |
| "learning_rate": 4.807692307692308e-07, | |
| "loss": 0.1271, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.276923076923077, | |
| "grad_norm": 3.954479932785034, | |
| "learning_rate": 3.205128205128205e-07, | |
| "loss": 0.1132, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.3025641025641024, | |
| "grad_norm": 0.6898751854896545, | |
| "learning_rate": 1.6025641025641025e-07, | |
| "loss": 0.0906, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.3282051282051284, | |
| "grad_norm": 2.044902801513672, | |
| "learning_rate": 0.0, | |
| "loss": 0.0832, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.3282051282051284, | |
| "eval_accuracy": 0.8475336322869955, | |
| "eval_loss": 0.5565493702888489, | |
| "eval_runtime": 62.3378, | |
| "eval_samples_per_second": 3.577, | |
| "eval_steps_per_second": 0.449, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.3282051282051284, | |
| "step": 390, | |
| "total_flos": 3.8861995625761997e+18, | |
| "train_loss": 0.12900116260235125, | |
| "train_runtime": 2004.3611, | |
| "train_samples_per_second": 1.557, | |
| "train_steps_per_second": 0.195 | |
| }, | |
| { | |
| "epoch": 2.3282051282051284, | |
| "eval_accuracy": 0.6438848920863309, | |
| "eval_loss": 2.050309658050537, | |
| "eval_runtime": 149.2566, | |
| "eval_samples_per_second": 3.725, | |
| "eval_steps_per_second": 0.469, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.3282051282051284, | |
| "eval_accuracy": 0.639344262295082, | |
| "eval_loss": 2.076159954071045, | |
| "eval_runtime": 145.7908, | |
| "eval_samples_per_second": 3.766, | |
| "eval_steps_per_second": 0.473, | |
| "step": 390 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 390, | |
| "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": 3.8861995625761997e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |