|
{ |
|
"best_metric": 0.33599379658699036, |
|
"best_model_checkpoint": "VIT-MUSH/checkpoint-19350", |
|
"epoch": 10.0, |
|
"eval_steps": 500, |
|
"global_step": 19350, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 9.766839378238344e-06, |
|
"loss": 2.9937, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 9.50777202072539e-06, |
|
"loss": 1.4942, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 9.248704663212435e-06, |
|
"loss": 1.0432, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.8197325062996704, |
|
"eval_loss": 0.8224276304244995, |
|
"eval_runtime": 202.4817, |
|
"eval_samples_per_second": 76.437, |
|
"eval_steps_per_second": 9.556, |
|
"step": 1935 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 8.989637305699482e-06, |
|
"loss": 0.8007, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"learning_rate": 8.730569948186529e-06, |
|
"loss": 0.6374, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 1.55, |
|
"learning_rate": 8.471502590673576e-06, |
|
"loss": 0.5648, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"learning_rate": 8.212435233160623e-06, |
|
"loss": 0.5066, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8676746139432706, |
|
"eval_loss": 0.5276621580123901, |
|
"eval_runtime": 206.3635, |
|
"eval_samples_per_second": 74.999, |
|
"eval_steps_per_second": 9.377, |
|
"step": 3870 |
|
}, |
|
{ |
|
"epoch": 2.07, |
|
"learning_rate": 7.953367875647669e-06, |
|
"loss": 0.4381, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"learning_rate": 7.694300518134716e-06, |
|
"loss": 0.3837, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"learning_rate": 7.435233160621762e-06, |
|
"loss": 0.3633, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"learning_rate": 7.176165803108809e-06, |
|
"loss": 0.3591, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8844091232150934, |
|
"eval_loss": 0.4341337978839874, |
|
"eval_runtime": 214.6924, |
|
"eval_samples_per_second": 72.089, |
|
"eval_steps_per_second": 9.013, |
|
"step": 5805 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"learning_rate": 6.917098445595856e-06, |
|
"loss": 0.3152, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 3.36, |
|
"learning_rate": 6.658031088082902e-06, |
|
"loss": 0.2801, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 3.62, |
|
"learning_rate": 6.398963730569949e-06, |
|
"loss": 0.2786, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 3.88, |
|
"learning_rate": 6.139896373056995e-06, |
|
"loss": 0.2697, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.8936486399172966, |
|
"eval_loss": 0.38577526807785034, |
|
"eval_runtime": 212.7498, |
|
"eval_samples_per_second": 72.747, |
|
"eval_steps_per_second": 9.095, |
|
"step": 7740 |
|
}, |
|
{ |
|
"epoch": 4.13, |
|
"learning_rate": 5.880829015544042e-06, |
|
"loss": 0.2565, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 4.39, |
|
"learning_rate": 5.621761658031088e-06, |
|
"loss": 0.2292, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 4.65, |
|
"learning_rate": 5.3626943005181356e-06, |
|
"loss": 0.2257, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 4.91, |
|
"learning_rate": 5.103626943005182e-06, |
|
"loss": 0.2243, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.8954577760547909, |
|
"eval_loss": 0.3654685914516449, |
|
"eval_runtime": 212.081, |
|
"eval_samples_per_second": 72.977, |
|
"eval_steps_per_second": 9.124, |
|
"step": 9675 |
|
}, |
|
{ |
|
"epoch": 5.17, |
|
"learning_rate": 4.844559585492228e-06, |
|
"loss": 0.1937, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"learning_rate": 4.585492227979275e-06, |
|
"loss": 0.1847, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 5.68, |
|
"learning_rate": 4.326424870466322e-06, |
|
"loss": 0.1846, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 5.94, |
|
"learning_rate": 4.067357512953368e-06, |
|
"loss": 0.1879, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.898365316275764, |
|
"eval_loss": 0.35233184695243835, |
|
"eval_runtime": 207.9214, |
|
"eval_samples_per_second": 74.437, |
|
"eval_steps_per_second": 9.306, |
|
"step": 11610 |
|
}, |
|
{ |
|
"epoch": 6.2, |
|
"learning_rate": 3.808290155440415e-06, |
|
"loss": 0.1643, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 6.46, |
|
"learning_rate": 3.549222797927461e-06, |
|
"loss": 0.1613, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 6.72, |
|
"learning_rate": 3.2901554404145083e-06, |
|
"loss": 0.1562, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 6.98, |
|
"learning_rate": 3.031088082901555e-06, |
|
"loss": 0.1581, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.9004329004329005, |
|
"eval_loss": 0.34385889768600464, |
|
"eval_runtime": 210.6923, |
|
"eval_samples_per_second": 73.458, |
|
"eval_steps_per_second": 9.184, |
|
"step": 13545 |
|
}, |
|
{ |
|
"epoch": 7.24, |
|
"learning_rate": 2.772020725388601e-06, |
|
"loss": 0.1425, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 7.49, |
|
"learning_rate": 2.512953367875648e-06, |
|
"loss": 0.1324, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 7.75, |
|
"learning_rate": 2.2538860103626944e-06, |
|
"loss": 0.1375, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.9014020805065581, |
|
"eval_loss": 0.3380953073501587, |
|
"eval_runtime": 212.303, |
|
"eval_samples_per_second": 72.901, |
|
"eval_steps_per_second": 9.114, |
|
"step": 15480 |
|
}, |
|
{ |
|
"epoch": 8.01, |
|
"learning_rate": 1.994818652849741e-06, |
|
"loss": 0.1314, |
|
"step": 15500 |
|
}, |
|
{ |
|
"epoch": 8.27, |
|
"learning_rate": 1.7357512953367877e-06, |
|
"loss": 0.1175, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 8.53, |
|
"learning_rate": 1.4766839378238342e-06, |
|
"loss": 0.1231, |
|
"step": 16500 |
|
}, |
|
{ |
|
"epoch": 8.79, |
|
"learning_rate": 1.217616580310881e-06, |
|
"loss": 0.1294, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.9023066485753053, |
|
"eval_loss": 0.33753469586372375, |
|
"eval_runtime": 213.07, |
|
"eval_samples_per_second": 72.638, |
|
"eval_steps_per_second": 9.082, |
|
"step": 17415 |
|
}, |
|
{ |
|
"epoch": 9.04, |
|
"learning_rate": 9.585492227979275e-07, |
|
"loss": 0.1153, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 9.3, |
|
"learning_rate": 6.994818652849742e-07, |
|
"loss": 0.1098, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 9.56, |
|
"learning_rate": 4.404145077720207e-07, |
|
"loss": 0.1135, |
|
"step": 18500 |
|
}, |
|
{ |
|
"epoch": 9.82, |
|
"learning_rate": 1.8134715025906736e-07, |
|
"loss": 0.1162, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.903146604639142, |
|
"eval_loss": 0.33599379658699036, |
|
"eval_runtime": 212.5068, |
|
"eval_samples_per_second": 72.831, |
|
"eval_steps_per_second": 9.106, |
|
"step": 19350 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 19350, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"total_flos": 4.801355152861962e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|