| { | |
| "best_metric": 0.8285714285714286, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148", | |
| "epoch": 3.22972972972973, | |
| "eval_steps": 500, | |
| "global_step": 148, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06756756756756757, | |
| "grad_norm": 6.462096214294434, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.3605, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 7.138033866882324, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 2.14, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 7.901597023010254, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 2.1483, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25675675675675674, | |
| "eval_accuracy": 0.38571428571428573, | |
| "eval_loss": 1.8423467874526978, | |
| "eval_runtime": 7.0633, | |
| "eval_samples_per_second": 9.91, | |
| "eval_steps_per_second": 1.274, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 12.548489570617676, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 1.9367, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 13.077637672424316, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 1.5488, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 12.642136573791504, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 1.2444, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 6.2350873947143555, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 0.8601, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.972441554069519, | |
| "eval_runtime": 6.8861, | |
| "eval_samples_per_second": 10.165, | |
| "eval_steps_per_second": 1.307, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 5.450301170349121, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 0.8434, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 12.654264450073242, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 0.6005, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 11.712494850158691, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 0.5745, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 6.82019567489624, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.5181, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.2567567567567566, | |
| "eval_accuracy": 0.7571428571428571, | |
| "eval_loss": 0.6128838658332825, | |
| "eval_runtime": 6.845, | |
| "eval_samples_per_second": 10.226, | |
| "eval_steps_per_second": 1.315, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0405405405405403, | |
| "grad_norm": 3.097586154937744, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.3446, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.108108108108108, | |
| "grad_norm": 3.016726016998291, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.3227, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.175675675675676, | |
| "grad_norm": 5.630105495452881, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.3207, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8285714285714286, | |
| "eval_loss": 0.5353689789772034, | |
| "eval_runtime": 6.5995, | |
| "eval_samples_per_second": 10.607, | |
| "eval_steps_per_second": 1.364, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "step": 148, | |
| "total_flos": 1.460491890402263e+18, | |
| "train_loss": 1.0781523762522518, | |
| "train_runtime": 231.4897, | |
| "train_samples_per_second": 5.115, | |
| "train_steps_per_second": 0.639 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.864516129032258, | |
| "eval_loss": 0.5425320863723755, | |
| "eval_runtime": 14.3433, | |
| "eval_samples_per_second": 10.806, | |
| "eval_steps_per_second": 1.394, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.864516129032258, | |
| "eval_loss": 0.5425320267677307, | |
| "eval_runtime": 14.1728, | |
| "eval_samples_per_second": 10.936, | |
| "eval_steps_per_second": 1.411, | |
| "step": 148 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 148, | |
| "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": 1.460491890402263e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |