|
{ |
|
"best_metric": 0.921760391198044, |
|
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-sealv1/checkpoint-118", |
|
"epoch": 9.491525423728813, |
|
"eval_steps": 500, |
|
"global_step": 140, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.68, |
|
"grad_norm": 8.243694305419922, |
|
"learning_rate": 3.571428571428572e-05, |
|
"loss": 1.1068, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"eval_accuracy": 0.706601466992665, |
|
"eval_loss": 0.6517727375030518, |
|
"eval_runtime": 2.518, |
|
"eval_samples_per_second": 162.428, |
|
"eval_steps_per_second": 5.163, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 1.36, |
|
"grad_norm": 11.443496704101562, |
|
"learning_rate": 4.761904761904762e-05, |
|
"loss": 0.4912, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 1.97, |
|
"eval_accuracy": 0.843520782396088, |
|
"eval_loss": 0.46679040789604187, |
|
"eval_runtime": 2.5032, |
|
"eval_samples_per_second": 163.39, |
|
"eval_steps_per_second": 5.193, |
|
"step": 29 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"grad_norm": 9.79934024810791, |
|
"learning_rate": 4.3650793650793655e-05, |
|
"loss": 0.3366, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 2.71, |
|
"grad_norm": 8.352068901062012, |
|
"learning_rate": 3.968253968253968e-05, |
|
"loss": 0.2749, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 2.98, |
|
"eval_accuracy": 0.8704156479217604, |
|
"eval_loss": 0.41268301010131836, |
|
"eval_runtime": 2.4728, |
|
"eval_samples_per_second": 165.399, |
|
"eval_steps_per_second": 5.257, |
|
"step": 44 |
|
}, |
|
{ |
|
"epoch": 3.39, |
|
"grad_norm": 7.004426002502441, |
|
"learning_rate": 3.571428571428572e-05, |
|
"loss": 0.3189, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.8875305623471883, |
|
"eval_loss": 0.36258456110954285, |
|
"eval_runtime": 2.4887, |
|
"eval_samples_per_second": 164.344, |
|
"eval_steps_per_second": 5.224, |
|
"step": 59 |
|
}, |
|
{ |
|
"epoch": 4.07, |
|
"grad_norm": 4.621389865875244, |
|
"learning_rate": 3.1746031746031745e-05, |
|
"loss": 0.2412, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 4.75, |
|
"grad_norm": 6.152388572692871, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.2226, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"eval_accuracy": 0.9046454767726161, |
|
"eval_loss": 0.26376911997795105, |
|
"eval_runtime": 2.4808, |
|
"eval_samples_per_second": 164.864, |
|
"eval_steps_per_second": 5.24, |
|
"step": 73 |
|
}, |
|
{ |
|
"epoch": 5.42, |
|
"grad_norm": 6.929192066192627, |
|
"learning_rate": 2.380952380952381e-05, |
|
"loss": 0.2394, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 5.97, |
|
"eval_accuracy": 0.8801955990220048, |
|
"eval_loss": 0.358397513628006, |
|
"eval_runtime": 2.4802, |
|
"eval_samples_per_second": 164.908, |
|
"eval_steps_per_second": 5.242, |
|
"step": 88 |
|
}, |
|
{ |
|
"epoch": 6.1, |
|
"grad_norm": 6.410078525543213, |
|
"learning_rate": 1.984126984126984e-05, |
|
"loss": 0.2156, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 6.78, |
|
"grad_norm": 6.870254039764404, |
|
"learning_rate": 1.5873015873015872e-05, |
|
"loss": 0.2241, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 6.98, |
|
"eval_accuracy": 0.9046454767726161, |
|
"eval_loss": 0.2820737659931183, |
|
"eval_runtime": 2.4944, |
|
"eval_samples_per_second": 163.967, |
|
"eval_steps_per_second": 5.212, |
|
"step": 103 |
|
}, |
|
{ |
|
"epoch": 7.46, |
|
"grad_norm": 20.99500274658203, |
|
"learning_rate": 1.1904761904761905e-05, |
|
"loss": 0.1815, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.921760391198044, |
|
"eval_loss": 0.21383894979953766, |
|
"eval_runtime": 2.4781, |
|
"eval_samples_per_second": 165.045, |
|
"eval_steps_per_second": 5.246, |
|
"step": 118 |
|
}, |
|
{ |
|
"epoch": 8.14, |
|
"grad_norm": 10.073310852050781, |
|
"learning_rate": 7.936507936507936e-06, |
|
"loss": 0.1715, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 8.81, |
|
"grad_norm": 5.79032564163208, |
|
"learning_rate": 3.968253968253968e-06, |
|
"loss": 0.1862, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 8.95, |
|
"eval_accuracy": 0.9046454767726161, |
|
"eval_loss": 0.27377307415008545, |
|
"eval_runtime": 2.4752, |
|
"eval_samples_per_second": 165.24, |
|
"eval_steps_per_second": 5.252, |
|
"step": 132 |
|
}, |
|
{ |
|
"epoch": 9.49, |
|
"grad_norm": 9.280794143676758, |
|
"learning_rate": 0.0, |
|
"loss": 0.1942, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 9.49, |
|
"eval_accuracy": 0.9119804400977995, |
|
"eval_loss": 0.25534436106681824, |
|
"eval_runtime": 2.4962, |
|
"eval_samples_per_second": 163.851, |
|
"eval_steps_per_second": 5.208, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 9.49, |
|
"step": 140, |
|
"total_flos": 4.447730655960883e+17, |
|
"train_loss": 0.31461275049618315, |
|
"train_runtime": 310.8231, |
|
"train_samples_per_second": 60.645, |
|
"train_steps_per_second": 0.45 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 140, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"total_flos": 4.447730655960883e+17, |
|
"train_batch_size": 32, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|