|
{ |
|
"best_metric": 0.02395050786435604, |
|
"best_model_checkpoint": "vit-large-patch16-224-new-dungeon-geo-morphs-22Nov24-007/checkpoint-30", |
|
"epoch": 20.0, |
|
"eval_steps": 10, |
|
"global_step": 30, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 6.666666666666667, |
|
"grad_norm": 15.621172904968262, |
|
"learning_rate": 1.6296296296296297e-05, |
|
"loss": 1.0851, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 6.666666666666667, |
|
"eval_accuracy": 0.9259259259259259, |
|
"eval_loss": 0.42449721693992615, |
|
"eval_runtime": 0.4582, |
|
"eval_samples_per_second": 58.93, |
|
"eval_steps_per_second": 8.73, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 13.333333333333334, |
|
"grad_norm": 1.7913310527801514, |
|
"learning_rate": 8.888888888888888e-06, |
|
"loss": 0.121, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 13.333333333333334, |
|
"eval_accuracy": 1.0, |
|
"eval_loss": 0.07044178247451782, |
|
"eval_runtime": 0.4549, |
|
"eval_samples_per_second": 59.356, |
|
"eval_steps_per_second": 8.793, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"grad_norm": 0.31953659653663635, |
|
"learning_rate": 1.4814814814814815e-06, |
|
"loss": 0.0094, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 1.0, |
|
"eval_loss": 0.02395050786435604, |
|
"eval_runtime": 0.4569, |
|
"eval_samples_per_second": 59.097, |
|
"eval_steps_per_second": 8.755, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"step": 30, |
|
"total_flos": 2.6297859010461696e+17, |
|
"train_loss": 0.4051565408706665, |
|
"train_runtime": 240.5669, |
|
"train_samples_per_second": 5.986, |
|
"train_steps_per_second": 0.125 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 30, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 30, |
|
"save_steps": 10, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 2.6297859010461696e+17, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|