krich97's picture
Training in progress, epoch 0
ad979c5
raw
history blame
2.9 kB
{
"best_metric": 0.6930860033726813,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-123",
"epoch": 2.946107784431138,
"eval_steps": 500,
"global_step": 123,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.24,
"learning_rate": 3.846153846153846e-05,
"loss": 0.689,
"step": 10
},
{
"epoch": 0.48,
"learning_rate": 4.681818181818182e-05,
"loss": 0.683,
"step": 20
},
{
"epoch": 0.72,
"learning_rate": 4.2272727272727275e-05,
"loss": 0.6604,
"step": 30
},
{
"epoch": 0.96,
"learning_rate": 3.7727272727272725e-05,
"loss": 0.6277,
"step": 40
},
{
"epoch": 0.98,
"eval_accuracy": 0.642495784148398,
"eval_loss": 0.6069725751876831,
"eval_runtime": 3.7565,
"eval_samples_per_second": 157.86,
"eval_steps_per_second": 5.058,
"step": 41
},
{
"epoch": 1.2,
"learning_rate": 3.318181818181819e-05,
"loss": 0.5928,
"step": 50
},
{
"epoch": 1.44,
"learning_rate": 2.863636363636364e-05,
"loss": 0.6013,
"step": 60
},
{
"epoch": 1.68,
"learning_rate": 2.4090909090909093e-05,
"loss": 0.5974,
"step": 70
},
{
"epoch": 1.92,
"learning_rate": 1.9545454545454546e-05,
"loss": 0.557,
"step": 80
},
{
"epoch": 1.99,
"eval_accuracy": 0.6644182124789207,
"eval_loss": 0.6569493412971497,
"eval_runtime": 3.6917,
"eval_samples_per_second": 160.63,
"eval_steps_per_second": 5.147,
"step": 83
},
{
"epoch": 2.16,
"learning_rate": 1.5e-05,
"loss": 0.546,
"step": 90
},
{
"epoch": 2.4,
"learning_rate": 1.0454545454545455e-05,
"loss": 0.5579,
"step": 100
},
{
"epoch": 2.63,
"learning_rate": 5.909090909090909e-06,
"loss": 0.5135,
"step": 110
},
{
"epoch": 2.87,
"learning_rate": 1.3636363636363636e-06,
"loss": 0.5445,
"step": 120
},
{
"epoch": 2.95,
"eval_accuracy": 0.6930860033726813,
"eval_loss": 0.6096150279045105,
"eval_runtime": 3.9493,
"eval_samples_per_second": 150.152,
"eval_steps_per_second": 4.811,
"step": 123
},
{
"epoch": 2.95,
"step": 123,
"total_flos": 3.9088528279157146e+17,
"train_loss": 0.5964590165673233,
"train_runtime": 256.7411,
"train_samples_per_second": 62.339,
"train_steps_per_second": 0.479
}
],
"logging_steps": 10,
"max_steps": 123,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 3.9088528279157146e+17,
"trial_name": null,
"trial_params": null
}