focalnet-small-patacon / trainer_state.json
frncscp's picture
End of training
b51bf7e
{
"best_metric": 0.9659090909090909,
"best_model_checkpoint": "focal-patacotron/checkpoint-48",
"epoch": 1.9393939393939394,
"global_step": 48,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.08,
"learning_rate": 2e-05,
"loss": 0.6836,
"step": 2
},
{
"epoch": 0.16,
"learning_rate": 4e-05,
"loss": 0.5915,
"step": 4
},
{
"epoch": 0.24,
"learning_rate": 4.883720930232558e-05,
"loss": 0.4642,
"step": 6
},
{
"epoch": 0.32,
"learning_rate": 4.651162790697675e-05,
"loss": 0.3705,
"step": 8
},
{
"epoch": 0.4,
"learning_rate": 4.418604651162791e-05,
"loss": 0.185,
"step": 10
},
{
"epoch": 0.48,
"learning_rate": 4.186046511627907e-05,
"loss": 0.1973,
"step": 12
},
{
"epoch": 0.57,
"learning_rate": 3.953488372093023e-05,
"loss": 0.1505,
"step": 14
},
{
"epoch": 0.65,
"learning_rate": 3.7209302325581394e-05,
"loss": 0.1137,
"step": 16
},
{
"epoch": 0.73,
"learning_rate": 3.488372093023256e-05,
"loss": 0.094,
"step": 18
},
{
"epoch": 0.81,
"learning_rate": 3.2558139534883724e-05,
"loss": 0.0736,
"step": 20
},
{
"epoch": 0.89,
"learning_rate": 3.0232558139534883e-05,
"loss": 0.0946,
"step": 22
},
{
"epoch": 0.97,
"learning_rate": 2.7906976744186048e-05,
"loss": 0.0575,
"step": 24
},
{
"epoch": 0.97,
"eval_accuracy": 0.9545454545454546,
"eval_loss": 0.12081997096538544,
"eval_runtime": 1.7666,
"eval_samples_per_second": 49.813,
"eval_steps_per_second": 6.227,
"step": 24
},
{
"epoch": 1.05,
"learning_rate": 2.5581395348837212e-05,
"loss": 0.0139,
"step": 26
},
{
"epoch": 1.13,
"learning_rate": 2.3255813953488374e-05,
"loss": 0.0574,
"step": 28
},
{
"epoch": 1.21,
"learning_rate": 2.0930232558139536e-05,
"loss": 0.1247,
"step": 30
},
{
"epoch": 1.29,
"learning_rate": 1.8604651162790697e-05,
"loss": 0.0401,
"step": 32
},
{
"epoch": 1.37,
"learning_rate": 1.6279069767441862e-05,
"loss": 0.2808,
"step": 34
},
{
"epoch": 1.45,
"learning_rate": 1.3953488372093024e-05,
"loss": 0.0545,
"step": 36
},
{
"epoch": 1.54,
"learning_rate": 1.1627906976744187e-05,
"loss": 0.0295,
"step": 38
},
{
"epoch": 1.62,
"learning_rate": 9.302325581395349e-06,
"loss": 0.0947,
"step": 40
},
{
"epoch": 1.7,
"learning_rate": 6.976744186046512e-06,
"loss": 0.0312,
"step": 42
},
{
"epoch": 1.78,
"learning_rate": 4.651162790697674e-06,
"loss": 0.1001,
"step": 44
},
{
"epoch": 1.86,
"learning_rate": 2.325581395348837e-06,
"loss": 0.014,
"step": 46
},
{
"epoch": 1.94,
"learning_rate": 0.0,
"loss": 0.0275,
"step": 48
},
{
"epoch": 1.94,
"eval_accuracy": 0.9659090909090909,
"eval_loss": 0.08971412479877472,
"eval_runtime": 1.8839,
"eval_samples_per_second": 46.711,
"eval_steps_per_second": 5.839,
"step": 48
},
{
"epoch": 1.94,
"step": 48,
"total_flos": 8.866156938264576e+16,
"train_loss": 0.16435068647842854,
"train_runtime": 116.6536,
"train_samples_per_second": 13.476,
"train_steps_per_second": 0.411
}
],
"max_steps": 48,
"num_train_epochs": 2,
"total_flos": 8.866156938264576e+16,
"trial_name": null,
"trial_params": null
}