t1msan's picture
End of training
6f9af57 verified
{
"best_metric": 0.006183081306517124,
"best_model_checkpoint": "swin-base-patch4-window7-224-in22k-Kontur-competition-1.4K/checkpoint-19",
"epoch": 8.421052631578947,
"eval_steps": 500,
"global_step": 40,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.84,
"eval_loss": 0.5152684450149536,
"eval_runtime": 5.7612,
"eval_samples_per_second": 23.086,
"eval_steps_per_second": 0.521,
"step": 4
},
{
"epoch": 1.89,
"eval_loss": 0.14250300824642181,
"eval_runtime": 3.5004,
"eval_samples_per_second": 37.995,
"eval_steps_per_second": 0.857,
"step": 9
},
{
"epoch": 2.11,
"grad_norm": 2.0780093669891357,
"learning_rate": 4.166666666666667e-05,
"loss": 0.4963,
"step": 10
},
{
"epoch": 2.95,
"eval_loss": 0.03258229419589043,
"eval_runtime": 2.856,
"eval_samples_per_second": 46.569,
"eval_steps_per_second": 1.05,
"step": 14
},
{
"epoch": 4.0,
"eval_loss": 0.006183081306517124,
"eval_runtime": 2.8656,
"eval_samples_per_second": 46.412,
"eval_steps_per_second": 1.047,
"step": 19
},
{
"epoch": 4.21,
"grad_norm": 4.404883861541748,
"learning_rate": 2.777777777777778e-05,
"loss": 0.0546,
"step": 20
},
{
"epoch": 4.84,
"eval_loss": 0.11129578948020935,
"eval_runtime": 2.942,
"eval_samples_per_second": 45.207,
"eval_steps_per_second": 1.02,
"step": 23
},
{
"epoch": 5.89,
"eval_loss": 0.019418848678469658,
"eval_runtime": 3.0717,
"eval_samples_per_second": 43.299,
"eval_steps_per_second": 0.977,
"step": 28
},
{
"epoch": 6.32,
"grad_norm": 4.566842079162598,
"learning_rate": 1.388888888888889e-05,
"loss": 0.0216,
"step": 30
},
{
"epoch": 6.95,
"eval_loss": 0.006876213010400534,
"eval_runtime": 2.9022,
"eval_samples_per_second": 45.828,
"eval_steps_per_second": 1.034,
"step": 33
},
{
"epoch": 8.0,
"eval_loss": 0.009907796047627926,
"eval_runtime": 2.917,
"eval_samples_per_second": 45.594,
"eval_steps_per_second": 1.028,
"step": 38
},
{
"epoch": 8.42,
"grad_norm": 2.6469647884368896,
"learning_rate": 0.0,
"loss": 0.0203,
"step": 40
},
{
"epoch": 8.42,
"eval_loss": 0.012587403878569603,
"eval_runtime": 2.9395,
"eval_samples_per_second": 45.246,
"eval_steps_per_second": 1.021,
"step": 40
},
{
"epoch": 8.42,
"step": 40,
"total_flos": 7.88469671840514e+17,
"train_loss": 0.14818734787404536,
"train_runtime": 550.7447,
"train_samples_per_second": 21.68,
"train_steps_per_second": 0.073
}
],
"logging_steps": 10,
"max_steps": 40,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 7.88469671840514e+17,
"train_batch_size": 64,
"trial_name": null,
"trial_params": null
}