|
{ |
|
"best_metric": 0.8327526132404182, |
|
"best_model_checkpoint": "swin-tiny-patch4-window7-224-brain-tumour-classification/checkpoint-60", |
|
"epoch": 2.962962962962963, |
|
"eval_steps": 500, |
|
"global_step": 60, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.49, |
|
"grad_norm": 7.377062797546387, |
|
"learning_rate": 4.62962962962963e-05, |
|
"loss": 1.2993, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"grad_norm": 8.843204498291016, |
|
"learning_rate": 3.7037037037037037e-05, |
|
"loss": 0.8673, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"eval_accuracy": 0.7456445993031359, |
|
"eval_loss": 0.6314167380332947, |
|
"eval_runtime": 68.7483, |
|
"eval_samples_per_second": 4.175, |
|
"eval_steps_per_second": 0.131, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 1.48, |
|
"grad_norm": 13.86172866821289, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.5745, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 1.98, |
|
"grad_norm": 90.3827133178711, |
|
"learning_rate": 1.8518518518518518e-05, |
|
"loss": 0.499, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 1.98, |
|
"eval_accuracy": 0.8153310104529616, |
|
"eval_loss": 0.4273669421672821, |
|
"eval_runtime": 63.007, |
|
"eval_samples_per_second": 4.555, |
|
"eval_steps_per_second": 0.143, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"grad_norm": 13.6727933883667, |
|
"learning_rate": 9.259259259259259e-06, |
|
"loss": 0.4234, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"grad_norm": 10.45642375946045, |
|
"learning_rate": 0.0, |
|
"loss": 0.3988, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"eval_accuracy": 0.8327526132404182, |
|
"eval_loss": 0.3846690058708191, |
|
"eval_runtime": 63.2618, |
|
"eval_samples_per_second": 4.537, |
|
"eval_steps_per_second": 0.142, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"step": 60, |
|
"total_flos": 1.9045723068223488e+17, |
|
"train_loss": 0.6770469148953756, |
|
"train_runtime": 5445.1813, |
|
"train_samples_per_second": 1.423, |
|
"train_steps_per_second": 0.011 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 60, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"total_flos": 1.9045723068223488e+17, |
|
"train_batch_size": 32, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|