Omar
update
88cccb3
raw
history blame
3.21 kB
{
"best_metric": 0.9936221684627228,
"best_model_checkpoint": "final_models/glue_models/structroberta_s2_50ep//finetune/relative_position_control/checkpoint-1400",
"epoch": 10.0,
"global_step": 1420,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.41,
"eval_accuracy": 0.9869555830955505,
"eval_f1": 0.9870911278853518,
"eval_loss": 0.04915473610162735,
"eval_mcc": 0.9741260163219052,
"eval_runtime": 38.5522,
"eval_samples_per_second": 469.286,
"eval_steps_per_second": 58.674,
"step": 200
},
{
"epoch": 2.82,
"eval_accuracy": 0.9912668466567993,
"eval_f1": 0.9913301141352064,
"eval_loss": 0.037929635494947433,
"eval_mcc": 0.9826383382000646,
"eval_runtime": 38.5328,
"eval_samples_per_second": 469.522,
"eval_steps_per_second": 58.703,
"step": 400
},
{
"epoch": 3.52,
"learning_rate": 3.23943661971831e-05,
"loss": 0.0194,
"step": 500
},
{
"epoch": 4.23,
"eval_accuracy": 0.9922617673873901,
"eval_f1": 0.9923119165293793,
"eval_loss": 0.03434313088655472,
"eval_mcc": 0.9846073189946343,
"eval_runtime": 38.5933,
"eval_samples_per_second": 468.786,
"eval_steps_per_second": 58.611,
"step": 600
},
{
"epoch": 5.63,
"eval_accuracy": 0.9929803013801575,
"eval_f1": 0.9930215945931095,
"eval_loss": 0.03289680927991867,
"eval_mcc": 0.9860296265991517,
"eval_runtime": 38.4721,
"eval_samples_per_second": 470.263,
"eval_steps_per_second": 58.796,
"step": 800
},
{
"epoch": 7.04,
"learning_rate": 1.4788732394366198e-05,
"loss": 0.0,
"step": 1000
},
{
"epoch": 7.04,
"eval_accuracy": 0.9932014346122742,
"eval_f1": 0.993239901071723,
"eval_loss": 0.03196977451443672,
"eval_mcc": 0.9864667781069306,
"eval_runtime": 38.303,
"eval_samples_per_second": 472.339,
"eval_steps_per_second": 59.055,
"step": 1000
},
{
"epoch": 8.45,
"eval_accuracy": 0.9935330748558044,
"eval_f1": 0.9935675408213757,
"eval_loss": 0.031383175402879715,
"eval_mcc": 0.9871228589360604,
"eval_runtime": 38.4738,
"eval_samples_per_second": 470.242,
"eval_steps_per_second": 58.793,
"step": 1200
},
{
"epoch": 9.86,
"eval_accuracy": 0.9935883283615112,
"eval_f1": 0.9936221684627228,
"eval_loss": 0.03111557476222515,
"eval_mcc": 0.9872322470407766,
"eval_runtime": 38.3733,
"eval_samples_per_second": 471.474,
"eval_steps_per_second": 58.947,
"step": 1400
},
{
"epoch": 10.0,
"step": 1420,
"total_flos": 8306652693196800.0,
"train_loss": 0.006840947497977127,
"train_runtime": 753.8127,
"train_samples_per_second": 120.295,
"train_steps_per_second": 1.884
}
],
"max_steps": 1420,
"num_train_epochs": 10,
"total_flos": 8306652693196800.0,
"trial_name": null,
"trial_params": null
}