hBERTv1_mrpc / trainer_state.json
gokuls's picture
End of training
f6b49a6
raw
history blame
5.87 kB
{
"best_metric": 0.6050820350646973,
"best_model_checkpoint": "hBERTv1_mrpc/checkpoint-105",
"epoch": 12.0,
"global_step": 180,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"learning_rate": 4.9e-05,
"loss": 0.6536,
"step": 15
},
{
"epoch": 1.0,
"eval_accuracy": 0.6838235294117647,
"eval_combined_score": 0.7480253018237863,
"eval_f1": 0.8122270742358079,
"eval_loss": 0.6242610216140747,
"eval_runtime": 0.5132,
"eval_samples_per_second": 794.982,
"eval_steps_per_second": 3.897,
"step": 15
},
{
"epoch": 2.0,
"learning_rate": 4.8e-05,
"loss": 0.6275,
"step": 30
},
{
"epoch": 2.0,
"eval_accuracy": 0.7009803921568627,
"eval_combined_score": 0.7563543936092956,
"eval_f1": 0.8117283950617284,
"eval_loss": 0.6174396872520447,
"eval_runtime": 0.511,
"eval_samples_per_second": 798.461,
"eval_steps_per_second": 3.914,
"step": 30
},
{
"epoch": 3.0,
"learning_rate": 4.7e-05,
"loss": 0.6129,
"step": 45
},
{
"epoch": 3.0,
"eval_accuracy": 0.696078431372549,
"eval_combined_score": 0.7571301247771836,
"eval_f1": 0.8181818181818181,
"eval_loss": 0.6089469194412231,
"eval_runtime": 0.5125,
"eval_samples_per_second": 796.115,
"eval_steps_per_second": 3.903,
"step": 45
},
{
"epoch": 4.0,
"learning_rate": 4.600000000000001e-05,
"loss": 0.6087,
"step": 60
},
{
"epoch": 4.0,
"eval_accuracy": 0.6887254901960784,
"eval_combined_score": 0.7508428629183632,
"eval_f1": 0.8129602356406479,
"eval_loss": 0.606163740158081,
"eval_runtime": 0.5136,
"eval_samples_per_second": 794.368,
"eval_steps_per_second": 3.894,
"step": 60
},
{
"epoch": 5.0,
"learning_rate": 4.5e-05,
"loss": 0.5939,
"step": 75
},
{
"epoch": 5.0,
"eval_accuracy": 0.6862745098039216,
"eval_combined_score": 0.739911448450348,
"eval_f1": 0.7935483870967743,
"eval_loss": 0.6104128956794739,
"eval_runtime": 0.5141,
"eval_samples_per_second": 793.575,
"eval_steps_per_second": 3.89,
"step": 75
},
{
"epoch": 6.0,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.5707,
"step": 90
},
{
"epoch": 6.0,
"eval_accuracy": 0.7083333333333334,
"eval_combined_score": 0.7633269720101782,
"eval_f1": 0.818320610687023,
"eval_loss": 0.6184039115905762,
"eval_runtime": 0.5141,
"eval_samples_per_second": 793.682,
"eval_steps_per_second": 3.891,
"step": 90
},
{
"epoch": 7.0,
"learning_rate": 4.3e-05,
"loss": 0.5426,
"step": 105
},
{
"epoch": 7.0,
"eval_accuracy": 0.6862745098039216,
"eval_combined_score": 0.7431372549019608,
"eval_f1": 0.7999999999999999,
"eval_loss": 0.6050820350646973,
"eval_runtime": 0.5141,
"eval_samples_per_second": 793.592,
"eval_steps_per_second": 3.89,
"step": 105
},
{
"epoch": 8.0,
"learning_rate": 4.2e-05,
"loss": 0.4819,
"step": 120
},
{
"epoch": 8.0,
"eval_accuracy": 0.6936274509803921,
"eval_combined_score": 0.7477645971225257,
"eval_f1": 0.8019017432646594,
"eval_loss": 0.6560007929801941,
"eval_runtime": 0.515,
"eval_samples_per_second": 792.192,
"eval_steps_per_second": 3.883,
"step": 120
},
{
"epoch": 9.0,
"learning_rate": 4.1e-05,
"loss": 0.4279,
"step": 135
},
{
"epoch": 9.0,
"eval_accuracy": 0.6887254901960784,
"eval_combined_score": 0.7282749937269242,
"eval_f1": 0.7678244972577698,
"eval_loss": 0.6672889590263367,
"eval_runtime": 0.5142,
"eval_samples_per_second": 793.515,
"eval_steps_per_second": 3.89,
"step": 135
},
{
"epoch": 10.0,
"learning_rate": 4e-05,
"loss": 0.3374,
"step": 150
},
{
"epoch": 10.0,
"eval_accuracy": 0.6862745098039216,
"eval_combined_score": 0.7382192221150756,
"eval_f1": 0.7901639344262296,
"eval_loss": 0.8091686367988586,
"eval_runtime": 0.5155,
"eval_samples_per_second": 791.402,
"eval_steps_per_second": 3.879,
"step": 150
},
{
"epoch": 11.0,
"learning_rate": 3.9000000000000006e-05,
"loss": 0.2789,
"step": 165
},
{
"epoch": 11.0,
"eval_accuracy": 0.6887254901960784,
"eval_combined_score": 0.741110712577714,
"eval_f1": 0.7934959349593496,
"eval_loss": 0.9342193603515625,
"eval_runtime": 0.5139,
"eval_samples_per_second": 793.922,
"eval_steps_per_second": 3.892,
"step": 165
},
{
"epoch": 12.0,
"learning_rate": 3.8e-05,
"loss": 0.2216,
"step": 180
},
{
"epoch": 12.0,
"eval_accuracy": 0.6838235294117647,
"eval_combined_score": 0.7324041246379707,
"eval_f1": 0.7809847198641766,
"eval_loss": 0.9707943201065063,
"eval_runtime": 0.5142,
"eval_samples_per_second": 793.529,
"eval_steps_per_second": 3.89,
"step": 180
},
{
"epoch": 12.0,
"step": 180,
"total_flos": 5559706315653120.0,
"train_loss": 0.4964691546228197,
"train_runtime": 203.7797,
"train_samples_per_second": 899.991,
"train_steps_per_second": 3.68
}
],
"max_steps": 750,
"num_train_epochs": 50,
"total_flos": 5559706315653120.0,
"trial_name": null,
"trial_params": null
}