MemeDetector_SG / trainer_state.json
friedrice231's picture
End of training
385e05a verified
{
"best_metric": 0.9586165533786486,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-51",
"epoch": 3.0,
"eval_steps": 500,
"global_step": 153,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.2,
"learning_rate": 3.125e-05,
"loss": 0.1412,
"step": 10
},
{
"epoch": 0.39,
"learning_rate": 4.854014598540147e-05,
"loss": 0.1664,
"step": 20
},
{
"epoch": 0.59,
"learning_rate": 4.489051094890511e-05,
"loss": 0.1379,
"step": 30
},
{
"epoch": 0.78,
"learning_rate": 4.124087591240876e-05,
"loss": 0.143,
"step": 40
},
{
"epoch": 0.98,
"learning_rate": 3.759124087591241e-05,
"loss": 0.1516,
"step": 50
},
{
"epoch": 1.0,
"eval_accuracy": 0.9586165533786486,
"eval_loss": 0.10476645827293396,
"eval_runtime": 175.3791,
"eval_samples_per_second": 28.521,
"eval_steps_per_second": 0.45,
"step": 51
},
{
"epoch": 1.18,
"learning_rate": 3.3941605839416055e-05,
"loss": 0.1234,
"step": 60
},
{
"epoch": 1.37,
"learning_rate": 3.029197080291971e-05,
"loss": 0.123,
"step": 70
},
{
"epoch": 1.57,
"learning_rate": 2.664233576642336e-05,
"loss": 0.1174,
"step": 80
},
{
"epoch": 1.76,
"learning_rate": 2.2992700729927008e-05,
"loss": 0.112,
"step": 90
},
{
"epoch": 1.96,
"learning_rate": 1.934306569343066e-05,
"loss": 0.1099,
"step": 100
},
{
"epoch": 2.0,
"eval_accuracy": 0.9446221511395442,
"eval_loss": 0.14113304018974304,
"eval_runtime": 157.5076,
"eval_samples_per_second": 31.757,
"eval_steps_per_second": 0.502,
"step": 102
},
{
"epoch": 2.16,
"learning_rate": 1.569343065693431e-05,
"loss": 0.1061,
"step": 110
},
{
"epoch": 2.35,
"learning_rate": 1.2043795620437957e-05,
"loss": 0.0913,
"step": 120
},
{
"epoch": 2.55,
"learning_rate": 8.394160583941606e-06,
"loss": 0.0929,
"step": 130
},
{
"epoch": 2.75,
"learning_rate": 4.744525547445255e-06,
"loss": 0.1019,
"step": 140
},
{
"epoch": 2.94,
"learning_rate": 1.0948905109489052e-06,
"loss": 0.0999,
"step": 150
},
{
"epoch": 3.0,
"eval_accuracy": 0.955217912834866,
"eval_loss": 0.11470602452754974,
"eval_runtime": 437.3167,
"eval_samples_per_second": 11.438,
"eval_steps_per_second": 0.181,
"step": 153
},
{
"epoch": 3.0,
"step": 153,
"total_flos": 9.695327063822991e+17,
"train_loss": 0.12046643585161446,
"train_runtime": 4806.9451,
"train_samples_per_second": 8.115,
"train_steps_per_second": 0.032
}
],
"logging_steps": 10,
"max_steps": 153,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 9.695327063822991e+17,
"train_batch_size": 64,
"trial_name": null,
"trial_params": null
}