swin-tiny-patch4-window7-224 / trainer_state.json
t1msan's picture
End of training
5b91158 verified
{
"best_metric": 0.1355326622724533,
"best_model_checkpoint": "microsoft/swin-tiny-patch4-window7-224/checkpoint-9",
"epoch": 4.2105263157894735,
"eval_steps": 500,
"global_step": 20,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.84,
"eval_loss": 0.6557008028030396,
"eval_runtime": 2.3009,
"eval_samples_per_second": 57.37,
"eval_steps_per_second": 1.304,
"step": 4
},
{
"epoch": 1.89,
"eval_loss": 0.1355326622724533,
"eval_runtime": 2.1272,
"eval_samples_per_second": 62.052,
"eval_steps_per_second": 1.41,
"step": 9
},
{
"epoch": 2.11,
"grad_norm": 11.091564178466797,
"learning_rate": 2.777777777777778e-05,
"loss": 0.0289,
"step": 10
},
{
"epoch": 2.95,
"eval_loss": 0.21629740297794342,
"eval_runtime": 2.0797,
"eval_samples_per_second": 63.472,
"eval_steps_per_second": 1.443,
"step": 14
},
{
"epoch": 4.0,
"eval_loss": 0.1559712290763855,
"eval_runtime": 2.1584,
"eval_samples_per_second": 61.156,
"eval_steps_per_second": 1.39,
"step": 19
},
{
"epoch": 4.21,
"grad_norm": 14.637483596801758,
"learning_rate": 0.0,
"loss": 0.0616,
"step": 20
},
{
"epoch": 4.21,
"eval_loss": 0.15557000041007996,
"eval_runtime": 2.0834,
"eval_samples_per_second": 63.359,
"eval_steps_per_second": 1.44,
"step": 20
},
{
"epoch": 4.21,
"step": 20,
"total_flos": 1.2713250126082867e+17,
"train_loss": 0.04524529278278351,
"train_runtime": 114.2042,
"train_samples_per_second": 51.662,
"train_steps_per_second": 0.175
}
],
"logging_steps": 10,
"max_steps": 20,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 1.2713250126082867e+17,
"train_batch_size": 64,
"trial_name": null,
"trial_params": null
}