| { | |
| "best_metric": 0.8571428571428571, | |
| "best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-ucf101-subset/checkpoint-19", | |
| "epoch": 3.2083333333333335, | |
| "eval_steps": 500, | |
| "global_step": 72, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1388888888888889, | |
| "grad_norm": 11.173284530639648, | |
| "learning_rate": 4.8437500000000005e-05, | |
| "loss": 0.8729, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2638888888888889, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.41932326555252075, | |
| "eval_runtime": 147.7861, | |
| "eval_samples_per_second": 2.558, | |
| "eval_steps_per_second": 0.325, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.0138888888888888, | |
| "grad_norm": 4.447381019592285, | |
| "learning_rate": 4.0625000000000005e-05, | |
| "loss": 0.4879, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.1527777777777777, | |
| "grad_norm": 3.373873233795166, | |
| "learning_rate": 3.2812500000000005e-05, | |
| "loss": 0.5106, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2638888888888888, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.45790210366249084, | |
| "eval_runtime": 146.1696, | |
| "eval_samples_per_second": 2.586, | |
| "eval_steps_per_second": 0.328, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.0277777777777777, | |
| "grad_norm": 0.6665381193161011, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2482, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.1666666666666665, | |
| "grad_norm": 8.423311233520508, | |
| "learning_rate": 1.71875e-05, | |
| "loss": 0.4699, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.263888888888889, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.4250260889530182, | |
| "eval_runtime": 146.1132, | |
| "eval_samples_per_second": 2.587, | |
| "eval_steps_per_second": 0.329, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 3.0416666666666665, | |
| "grad_norm": 0.42366838455200195, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 0.0982, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.1805555555555554, | |
| "grad_norm": 4.436389446258545, | |
| "learning_rate": 1.5625e-06, | |
| "loss": 0.109, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.2083333333333335, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.5306517481803894, | |
| "eval_runtime": 145.944, | |
| "eval_samples_per_second": 2.59, | |
| "eval_steps_per_second": 0.329, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.2083333333333335, | |
| "step": 72, | |
| "total_flos": 4.967755240837939e+17, | |
| "train_loss": 0.3919994781414668, | |
| "train_runtime": 877.0447, | |
| "train_samples_per_second": 0.657, | |
| "train_steps_per_second": 0.082 | |
| }, | |
| { | |
| "epoch": 3.2083333333333335, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.4621736705303192, | |
| "eval_runtime": 114.6382, | |
| "eval_samples_per_second": 2.512, | |
| "eval_steps_per_second": 0.314, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.2083333333333335, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.462173730134964, | |
| "eval_runtime": 113.0732, | |
| "eval_samples_per_second": 2.547, | |
| "eval_steps_per_second": 0.318, | |
| "step": 72 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 72, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 4.967755240837939e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |