maixbach's picture
End of training
51b5359
raw
history blame
5.72 kB
{
"best_metric": 0.882689556509299,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-trash_classification/checkpoint-198",
"epoch": 10.0,
"global_step": 220,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.45,
"learning_rate": 2.272727272727273e-05,
"loss": 0.531,
"step": 10
},
{
"epoch": 0.91,
"learning_rate": 4.545454545454546e-05,
"loss": 0.4991,
"step": 20
},
{
"epoch": 1.0,
"eval_accuracy": 0.7911301859799714,
"eval_loss": 0.5482170581817627,
"eval_runtime": 7.0774,
"eval_samples_per_second": 98.766,
"eval_steps_per_second": 3.109,
"step": 22
},
{
"epoch": 1.36,
"learning_rate": 4.797979797979798e-05,
"loss": 0.4159,
"step": 30
},
{
"epoch": 1.82,
"learning_rate": 4.545454545454546e-05,
"loss": 0.4008,
"step": 40
},
{
"epoch": 2.0,
"eval_accuracy": 0.7954220314735336,
"eval_loss": 0.5192908644676208,
"eval_runtime": 7.6454,
"eval_samples_per_second": 91.427,
"eval_steps_per_second": 2.878,
"step": 44
},
{
"epoch": 2.27,
"learning_rate": 4.292929292929293e-05,
"loss": 0.3623,
"step": 50
},
{
"epoch": 2.73,
"learning_rate": 4.0404040404040405e-05,
"loss": 0.3659,
"step": 60
},
{
"epoch": 3.0,
"eval_accuracy": 0.8397711015736766,
"eval_loss": 0.44643908739089966,
"eval_runtime": 7.1094,
"eval_samples_per_second": 98.321,
"eval_steps_per_second": 3.095,
"step": 66
},
{
"epoch": 3.18,
"learning_rate": 3.787878787878788e-05,
"loss": 0.3551,
"step": 70
},
{
"epoch": 3.64,
"learning_rate": 3.535353535353535e-05,
"loss": 0.372,
"step": 80
},
{
"epoch": 4.0,
"eval_accuracy": 0.8397711015736766,
"eval_loss": 0.4384276270866394,
"eval_runtime": 7.6735,
"eval_samples_per_second": 91.093,
"eval_steps_per_second": 2.867,
"step": 88
},
{
"epoch": 4.09,
"learning_rate": 3.282828282828283e-05,
"loss": 0.3277,
"step": 90
},
{
"epoch": 4.55,
"learning_rate": 3.0303030303030306e-05,
"loss": 0.3252,
"step": 100
},
{
"epoch": 5.0,
"learning_rate": 2.777777777777778e-05,
"loss": 0.3388,
"step": 110
},
{
"epoch": 5.0,
"eval_accuracy": 0.8454935622317596,
"eval_loss": 0.42808040976524353,
"eval_runtime": 7.1513,
"eval_samples_per_second": 97.745,
"eval_steps_per_second": 3.076,
"step": 110
},
{
"epoch": 5.45,
"learning_rate": 2.5252525252525256e-05,
"loss": 0.2767,
"step": 120
},
{
"epoch": 5.91,
"learning_rate": 2.272727272727273e-05,
"loss": 0.2654,
"step": 130
},
{
"epoch": 6.0,
"eval_accuracy": 0.871244635193133,
"eval_loss": 0.3618273138999939,
"eval_runtime": 7.6567,
"eval_samples_per_second": 91.293,
"eval_steps_per_second": 2.873,
"step": 132
},
{
"epoch": 6.36,
"learning_rate": 2.0202020202020203e-05,
"loss": 0.235,
"step": 140
},
{
"epoch": 6.82,
"learning_rate": 1.7676767676767676e-05,
"loss": 0.2326,
"step": 150
},
{
"epoch": 7.0,
"eval_accuracy": 0.8755364806866953,
"eval_loss": 0.3550046682357788,
"eval_runtime": 7.1895,
"eval_samples_per_second": 97.226,
"eval_steps_per_second": 3.06,
"step": 154
},
{
"epoch": 7.27,
"learning_rate": 1.5151515151515153e-05,
"loss": 0.219,
"step": 160
},
{
"epoch": 7.73,
"learning_rate": 1.2626262626262628e-05,
"loss": 0.2354,
"step": 170
},
{
"epoch": 8.0,
"eval_accuracy": 0.8798283261802575,
"eval_loss": 0.3400857448577881,
"eval_runtime": 7.5656,
"eval_samples_per_second": 92.392,
"eval_steps_per_second": 2.908,
"step": 176
},
{
"epoch": 8.18,
"learning_rate": 1.0101010101010101e-05,
"loss": 0.2455,
"step": 180
},
{
"epoch": 8.64,
"learning_rate": 7.5757575757575764e-06,
"loss": 0.1774,
"step": 190
},
{
"epoch": 9.0,
"eval_accuracy": 0.882689556509299,
"eval_loss": 0.337208092212677,
"eval_runtime": 7.0903,
"eval_samples_per_second": 98.586,
"eval_steps_per_second": 3.103,
"step": 198
},
{
"epoch": 9.09,
"learning_rate": 5.050505050505051e-06,
"loss": 0.2007,
"step": 200
},
{
"epoch": 9.55,
"learning_rate": 2.5252525252525253e-06,
"loss": 0.226,
"step": 210
},
{
"epoch": 10.0,
"learning_rate": 0.0,
"loss": 0.1849,
"step": 220
},
{
"epoch": 10.0,
"eval_accuracy": 0.882689556509299,
"eval_loss": 0.33802154660224915,
"eval_runtime": 7.6016,
"eval_samples_per_second": 91.954,
"eval_steps_per_second": 2.894,
"step": 220
},
{
"epoch": 10.0,
"step": 220,
"total_flos": 6.950317097403187e+17,
"train_loss": 0.30874538096514614,
"train_runtime": 658.9789,
"train_samples_per_second": 42.429,
"train_steps_per_second": 0.334
}
],
"max_steps": 220,
"num_train_epochs": 10,
"total_flos": 6.950317097403187e+17,
"trial_name": null,
"trial_params": null
}