julienmercier's picture
End of training
d352e4e
raw
history blame
3.95 kB
{
"best_metric": 0.9910828025477707,
"best_model_checkpoint": "vit-base-patch16-224-in21k-mobile-eye-tracking-dataset-v1/checkpoint-219",
"epoch": 2.969491525423729,
"global_step": 219,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.14,
"learning_rate": 2.272727272727273e-05,
"loss": 1.0467,
"step": 10
},
{
"epoch": 0.27,
"learning_rate": 4.545454545454546e-05,
"loss": 0.8237,
"step": 20
},
{
"epoch": 0.41,
"learning_rate": 4.7969543147208126e-05,
"loss": 0.5956,
"step": 30
},
{
"epoch": 0.54,
"learning_rate": 4.543147208121827e-05,
"loss": 0.3786,
"step": 40
},
{
"epoch": 0.68,
"learning_rate": 4.289340101522843e-05,
"loss": 0.2238,
"step": 50
},
{
"epoch": 0.81,
"learning_rate": 4.035532994923858e-05,
"loss": 0.1553,
"step": 60
},
{
"epoch": 0.95,
"learning_rate": 3.7817258883248735e-05,
"loss": 0.1179,
"step": 70
},
{
"epoch": 0.99,
"eval_accuracy": 0.9885350318471338,
"eval_loss": 0.09772440046072006,
"eval_runtime": 10.8587,
"eval_samples_per_second": 72.293,
"eval_steps_per_second": 3.039,
"step": 73
},
{
"epoch": 1.08,
"learning_rate": 3.527918781725888e-05,
"loss": 0.0909,
"step": 80
},
{
"epoch": 1.22,
"learning_rate": 3.2741116751269036e-05,
"loss": 0.0679,
"step": 90
},
{
"epoch": 1.36,
"learning_rate": 3.020304568527919e-05,
"loss": 0.0613,
"step": 100
},
{
"epoch": 1.49,
"learning_rate": 2.766497461928934e-05,
"loss": 0.0593,
"step": 110
},
{
"epoch": 1.63,
"learning_rate": 2.5126903553299492e-05,
"loss": 0.057,
"step": 120
},
{
"epoch": 1.76,
"learning_rate": 2.2588832487309646e-05,
"loss": 0.0612,
"step": 130
},
{
"epoch": 1.9,
"learning_rate": 2.0050761421319797e-05,
"loss": 0.06,
"step": 140
},
{
"epoch": 1.99,
"eval_accuracy": 0.9898089171974522,
"eval_loss": 0.0692891925573349,
"eval_runtime": 10.8744,
"eval_samples_per_second": 72.188,
"eval_steps_per_second": 3.035,
"step": 147
},
{
"epoch": 2.03,
"learning_rate": 1.751269035532995e-05,
"loss": 0.0481,
"step": 150
},
{
"epoch": 2.17,
"learning_rate": 1.4974619289340103e-05,
"loss": 0.0398,
"step": 160
},
{
"epoch": 2.31,
"learning_rate": 1.2436548223350254e-05,
"loss": 0.0461,
"step": 170
},
{
"epoch": 2.44,
"learning_rate": 9.898477157360408e-06,
"loss": 0.0356,
"step": 180
},
{
"epoch": 2.58,
"learning_rate": 7.360406091370558e-06,
"loss": 0.0482,
"step": 190
},
{
"epoch": 2.71,
"learning_rate": 4.822335025380711e-06,
"loss": 0.0397,
"step": 200
},
{
"epoch": 2.85,
"learning_rate": 2.284263959390863e-06,
"loss": 0.0376,
"step": 210
},
{
"epoch": 2.97,
"eval_accuracy": 0.9910828025477707,
"eval_loss": 0.0603952594101429,
"eval_runtime": 10.8431,
"eval_samples_per_second": 72.396,
"eval_steps_per_second": 3.043,
"step": 219
},
{
"epoch": 2.97,
"step": 219,
"total_flos": 1.6261064842200515e+18,
"train_loss": 0.18832773187933446,
"train_runtime": 875.8058,
"train_samples_per_second": 24.183,
"train_steps_per_second": 0.25
}
],
"max_steps": 219,
"num_train_epochs": 3,
"total_flos": 1.6261064842200515e+18,
"trial_name": null,
"trial_params": null
}