aaraki's picture
End of training
63acc43
raw
history blame
5.57 kB
{
"best_metric": 0.9788,
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-cifar10/checkpoint-390",
"epoch": 0.9980806142034548,
"global_step": 390,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 1.282051282051282e-05,
"loss": 2.3249,
"step": 10
},
{
"epoch": 0.05,
"learning_rate": 2.564102564102564e-05,
"loss": 2.2694,
"step": 20
},
{
"epoch": 0.08,
"learning_rate": 3.846153846153846e-05,
"loss": 2.1649,
"step": 30
},
{
"epoch": 0.1,
"learning_rate": 4.985754985754986e-05,
"loss": 1.9757,
"step": 40
},
{
"epoch": 0.13,
"learning_rate": 4.8433048433048433e-05,
"loss": 1.6938,
"step": 50
},
{
"epoch": 0.15,
"learning_rate": 4.700854700854701e-05,
"loss": 1.4069,
"step": 60
},
{
"epoch": 0.18,
"learning_rate": 4.558404558404559e-05,
"loss": 1.1938,
"step": 70
},
{
"epoch": 0.2,
"learning_rate": 4.415954415954416e-05,
"loss": 1.045,
"step": 80
},
{
"epoch": 0.23,
"learning_rate": 4.2735042735042735e-05,
"loss": 0.9435,
"step": 90
},
{
"epoch": 0.26,
"learning_rate": 4.131054131054131e-05,
"loss": 0.8542,
"step": 100
},
{
"epoch": 0.28,
"learning_rate": 3.988603988603989e-05,
"loss": 0.7705,
"step": 110
},
{
"epoch": 0.31,
"learning_rate": 3.846153846153846e-05,
"loss": 0.7295,
"step": 120
},
{
"epoch": 0.33,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.6826,
"step": 130
},
{
"epoch": 0.36,
"learning_rate": 3.561253561253561e-05,
"loss": 0.6886,
"step": 140
},
{
"epoch": 0.38,
"learning_rate": 3.418803418803419e-05,
"loss": 0.604,
"step": 150
},
{
"epoch": 0.41,
"learning_rate": 3.2763532763532764e-05,
"loss": 0.5829,
"step": 160
},
{
"epoch": 0.44,
"learning_rate": 3.133903133903134e-05,
"loss": 0.5991,
"step": 170
},
{
"epoch": 0.46,
"learning_rate": 2.9914529914529915e-05,
"loss": 0.5614,
"step": 180
},
{
"epoch": 0.49,
"learning_rate": 2.8490028490028492e-05,
"loss": 0.5863,
"step": 190
},
{
"epoch": 0.51,
"learning_rate": 2.706552706552707e-05,
"loss": 0.5146,
"step": 200
},
{
"epoch": 0.54,
"learning_rate": 2.564102564102564e-05,
"loss": 0.5278,
"step": 210
},
{
"epoch": 0.56,
"learning_rate": 2.4216524216524217e-05,
"loss": 0.4996,
"step": 220
},
{
"epoch": 0.59,
"learning_rate": 2.2792022792022794e-05,
"loss": 0.5387,
"step": 230
},
{
"epoch": 0.61,
"learning_rate": 2.1367521367521368e-05,
"loss": 0.4954,
"step": 240
},
{
"epoch": 0.64,
"learning_rate": 1.9943019943019945e-05,
"loss": 0.5052,
"step": 250
},
{
"epoch": 0.67,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.4617,
"step": 260
},
{
"epoch": 0.69,
"learning_rate": 1.7094017094017095e-05,
"loss": 0.4812,
"step": 270
},
{
"epoch": 0.72,
"learning_rate": 1.566951566951567e-05,
"loss": 0.4475,
"step": 280
},
{
"epoch": 0.74,
"learning_rate": 1.4245014245014246e-05,
"loss": 0.4786,
"step": 290
},
{
"epoch": 0.77,
"learning_rate": 1.282051282051282e-05,
"loss": 0.4722,
"step": 300
},
{
"epoch": 0.79,
"learning_rate": 1.1396011396011397e-05,
"loss": 0.4365,
"step": 310
},
{
"epoch": 0.82,
"learning_rate": 9.971509971509972e-06,
"loss": 0.4803,
"step": 320
},
{
"epoch": 0.84,
"learning_rate": 8.547008547008548e-06,
"loss": 0.4504,
"step": 330
},
{
"epoch": 0.87,
"learning_rate": 7.122507122507123e-06,
"loss": 0.4467,
"step": 340
},
{
"epoch": 0.9,
"learning_rate": 5.6980056980056985e-06,
"loss": 0.4477,
"step": 350
},
{
"epoch": 0.92,
"learning_rate": 4.273504273504274e-06,
"loss": 0.4257,
"step": 360
},
{
"epoch": 0.95,
"learning_rate": 2.8490028490028492e-06,
"loss": 0.4344,
"step": 370
},
{
"epoch": 0.97,
"learning_rate": 1.4245014245014246e-06,
"loss": 0.4493,
"step": 380
},
{
"epoch": 1.0,
"learning_rate": 0.0,
"loss": 0.4291,
"step": 390
},
{
"epoch": 1.0,
"eval_accuracy": 0.9788,
"eval_loss": 0.25641629099845886,
"eval_runtime": 262.0666,
"eval_samples_per_second": 38.158,
"eval_steps_per_second": 1.194,
"step": 390
},
{
"epoch": 1.0,
"step": 390,
"total_flos": 3.86867749153407e+18,
"train_loss": 0.7974265074118589,
"train_runtime": 3555.934,
"train_samples_per_second": 14.061,
"train_steps_per_second": 0.11
}
],
"max_steps": 390,
"num_train_epochs": 1,
"total_flos": 3.86867749153407e+18,
"trial_name": null,
"trial_params": null
}