|
{ |
|
"best_metric": 0.9785185185185186, |
|
"best_model_checkpoint": "vit-mae-base-finetuned-eurosat/checkpoint-141", |
|
"epoch": 2.968421052631579, |
|
"eval_steps": 500, |
|
"global_step": 141, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.21052631578947367, |
|
"grad_norm": 3.079359292984009, |
|
"learning_rate": 3.3333333333333335e-05, |
|
"loss": 2.1965, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.42105263157894735, |
|
"grad_norm": 1.6466779708862305, |
|
"learning_rate": 4.801587301587302e-05, |
|
"loss": 1.1697, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.631578947368421, |
|
"grad_norm": 0.9692021012306213, |
|
"learning_rate": 4.404761904761905e-05, |
|
"loss": 0.4723, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.8421052631578947, |
|
"grad_norm": 1.0668781995773315, |
|
"learning_rate": 4.007936507936508e-05, |
|
"loss": 0.274, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.9894736842105263, |
|
"eval_accuracy": 0.9618518518518518, |
|
"eval_loss": 0.13042595982551575, |
|
"eval_runtime": 6.1004, |
|
"eval_samples_per_second": 442.594, |
|
"eval_steps_per_second": 3.606, |
|
"step": 47 |
|
}, |
|
{ |
|
"epoch": 1.0526315789473684, |
|
"grad_norm": 0.8004966974258423, |
|
"learning_rate": 3.611111111111111e-05, |
|
"loss": 0.2051, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 1.263157894736842, |
|
"grad_norm": 1.0032081604003906, |
|
"learning_rate": 3.2142857142857144e-05, |
|
"loss": 0.1876, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.4736842105263157, |
|
"grad_norm": 1.218029260635376, |
|
"learning_rate": 2.8174603174603175e-05, |
|
"loss": 0.161, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.6842105263157894, |
|
"grad_norm": 1.2025845050811768, |
|
"learning_rate": 2.4206349206349206e-05, |
|
"loss": 0.1434, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.8947368421052633, |
|
"grad_norm": 1.054893136024475, |
|
"learning_rate": 2.023809523809524e-05, |
|
"loss": 0.137, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.9762962962962963, |
|
"eval_loss": 0.08248429000377655, |
|
"eval_runtime": 6.1084, |
|
"eval_samples_per_second": 442.017, |
|
"eval_steps_per_second": 3.602, |
|
"step": 95 |
|
}, |
|
{ |
|
"epoch": 2.1052631578947367, |
|
"grad_norm": 1.0172799825668335, |
|
"learning_rate": 1.626984126984127e-05, |
|
"loss": 0.1453, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 2.3157894736842106, |
|
"grad_norm": 0.862003743648529, |
|
"learning_rate": 1.2301587301587301e-05, |
|
"loss": 0.1379, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 2.526315789473684, |
|
"grad_norm": 0.868192732334137, |
|
"learning_rate": 8.333333333333334e-06, |
|
"loss": 0.1256, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.736842105263158, |
|
"grad_norm": 0.8207984566688538, |
|
"learning_rate": 4.365079365079365e-06, |
|
"loss": 0.1152, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.9473684210526314, |
|
"grad_norm": 0.761037290096283, |
|
"learning_rate": 3.9682539682539683e-07, |
|
"loss": 0.1401, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.968421052631579, |
|
"eval_accuracy": 0.9785185185185186, |
|
"eval_loss": 0.06672514975070953, |
|
"eval_runtime": 6.7122, |
|
"eval_samples_per_second": 402.254, |
|
"eval_steps_per_second": 3.278, |
|
"step": 141 |
|
}, |
|
{ |
|
"epoch": 2.968421052631579, |
|
"step": 141, |
|
"total_flos": 5.591602932074643e+18, |
|
"train_loss": 0.39885918373334495, |
|
"train_runtime": 292.9944, |
|
"train_samples_per_second": 248.81, |
|
"train_steps_per_second": 0.481 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 141, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"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": 5.591602932074643e+18, |
|
"train_batch_size": 128, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|