|
{ |
|
"best_metric": 0.5285714285714286, |
|
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-75", |
|
"epoch": 1.0, |
|
"eval_steps": 500, |
|
"global_step": 75, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.13, |
|
"grad_norm": 9.625768661499023, |
|
"learning_rate": 4.850746268656717e-05, |
|
"loss": 2.3821, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"grad_norm": 10.342418670654297, |
|
"learning_rate": 4.104477611940299e-05, |
|
"loss": 2.2626, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 9.640764236450195, |
|
"learning_rate": 3.358208955223881e-05, |
|
"loss": 2.1037, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"grad_norm": 9.587983131408691, |
|
"learning_rate": 2.6119402985074626e-05, |
|
"loss": 1.9337, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"grad_norm": 18.011554718017578, |
|
"learning_rate": 1.865671641791045e-05, |
|
"loss": 1.9766, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 10.898039817810059, |
|
"learning_rate": 1.119402985074627e-05, |
|
"loss": 1.8025, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"grad_norm": 17.220788955688477, |
|
"learning_rate": 3.7313432835820893e-06, |
|
"loss": 1.8562, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.5285714285714286, |
|
"eval_loss": 1.6040252447128296, |
|
"eval_runtime": 13.1446, |
|
"eval_samples_per_second": 5.325, |
|
"eval_steps_per_second": 1.369, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 75, |
|
"total_flos": 3.73846047031296e+17, |
|
"train_loss": 2.01439790725708, |
|
"train_runtime": 183.0609, |
|
"train_samples_per_second": 1.639, |
|
"train_steps_per_second": 0.41 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.33548387096774196, |
|
"eval_loss": 1.9053514003753662, |
|
"eval_runtime": 29.2689, |
|
"eval_samples_per_second": 5.296, |
|
"eval_steps_per_second": 1.332, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.33548387096774196, |
|
"eval_loss": 1.9053514003753662, |
|
"eval_runtime": 27.0656, |
|
"eval_samples_per_second": 5.727, |
|
"eval_steps_per_second": 1.441, |
|
"step": 75 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 75, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 9223372036854775807, |
|
"save_steps": 500, |
|
"total_flos": 3.73846047031296e+17, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|