luciana's picture
End of training
ce60e6f verified
raw
history blame
3.6 kB
{
"best_metric": 0.8910891089108911,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-lungs-disease/checkpoint-86",
"epoch": 4.883720930232558,
"eval_steps": 500,
"global_step": 105,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.47,
"grad_norm": 14.870257377624512,
"learning_rate": 4.545454545454546e-05,
"loss": 1.1236,
"step": 10
},
{
"epoch": 0.93,
"grad_norm": 20.39113998413086,
"learning_rate": 4.5212765957446815e-05,
"loss": 0.7104,
"step": 20
},
{
"epoch": 0.98,
"eval_accuracy": 0.7854785478547854,
"eval_loss": 0.5334098935127258,
"eval_runtime": 84.0412,
"eval_samples_per_second": 3.605,
"eval_steps_per_second": 0.119,
"step": 21
},
{
"epoch": 1.4,
"grad_norm": 11.177618980407715,
"learning_rate": 3.9893617021276594e-05,
"loss": 0.4852,
"step": 30
},
{
"epoch": 1.86,
"grad_norm": 10.540087699890137,
"learning_rate": 3.4574468085106386e-05,
"loss": 0.4067,
"step": 40
},
{
"epoch": 2.0,
"eval_accuracy": 0.834983498349835,
"eval_loss": 0.42113950848579407,
"eval_runtime": 81.1616,
"eval_samples_per_second": 3.733,
"eval_steps_per_second": 0.123,
"step": 43
},
{
"epoch": 2.33,
"grad_norm": 9.996329307556152,
"learning_rate": 2.925531914893617e-05,
"loss": 0.3041,
"step": 50
},
{
"epoch": 2.79,
"grad_norm": 10.684453010559082,
"learning_rate": 2.393617021276596e-05,
"loss": 0.3307,
"step": 60
},
{
"epoch": 2.98,
"eval_accuracy": 0.8778877887788779,
"eval_loss": 0.3033137917518616,
"eval_runtime": 82.0961,
"eval_samples_per_second": 3.691,
"eval_steps_per_second": 0.122,
"step": 64
},
{
"epoch": 3.26,
"grad_norm": 10.821243286132812,
"learning_rate": 1.8617021276595745e-05,
"loss": 0.261,
"step": 70
},
{
"epoch": 3.72,
"grad_norm": 6.7758073806762695,
"learning_rate": 1.3297872340425532e-05,
"loss": 0.287,
"step": 80
},
{
"epoch": 4.0,
"eval_accuracy": 0.8910891089108911,
"eval_loss": 0.2589154839515686,
"eval_runtime": 80.5026,
"eval_samples_per_second": 3.764,
"eval_steps_per_second": 0.124,
"step": 86
},
{
"epoch": 4.19,
"grad_norm": 13.222726821899414,
"learning_rate": 7.97872340425532e-06,
"loss": 0.2682,
"step": 90
},
{
"epoch": 4.65,
"grad_norm": 11.840774536132812,
"learning_rate": 2.6595744680851065e-06,
"loss": 0.2694,
"step": 100
},
{
"epoch": 4.88,
"eval_accuracy": 0.8844884488448845,
"eval_loss": 0.28530430793762207,
"eval_runtime": 79.9576,
"eval_samples_per_second": 3.79,
"eval_steps_per_second": 0.125,
"step": 105
},
{
"epoch": 4.88,
"step": 105,
"total_flos": 3.607403494903972e+17,
"train_loss": 0.43609671706245057,
"train_runtime": 11592.208,
"train_samples_per_second": 1.175,
"train_steps_per_second": 0.009
}
],
"logging_steps": 10,
"max_steps": 105,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 3.607403494903972e+17,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}