roberta-glue / stsb /roberta-base_lr1e-05 /trainer_state.json
lu-vae's picture
Upload folder using huggingface_hub
e52c1e9 verified
{
"best_metric": 0.8823497284035507,
"best_model_checkpoint": "./save_models/stsb/roberta-base_lr1e-05/checkpoint-210",
"epoch": 10.0,
"eval_steps": 500,
"global_step": 210,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_averaged_scores": 0.04498084566189773,
"eval_loss": 5.579086780548096,
"eval_pearson": 0.04953263745637067,
"eval_runtime": 1.0783,
"eval_samples_per_second": 533.238,
"eval_spearmanr": 0.04042905386742479,
"eval_steps_per_second": 2.782,
"step": 21
},
{
"epoch": 2.0,
"eval_averaged_scores": 0.6971092407797712,
"eval_loss": 1.9722641706466675,
"eval_pearson": 0.7207195402215961,
"eval_runtime": 1.0715,
"eval_samples_per_second": 536.624,
"eval_spearmanr": 0.6734989413379462,
"eval_steps_per_second": 2.8,
"step": 42
},
{
"epoch": 3.0,
"eval_averaged_scores": 0.8037596226086652,
"eval_loss": 0.8232775926589966,
"eval_pearson": 0.8147507903817527,
"eval_runtime": 1.0553,
"eval_samples_per_second": 544.863,
"eval_spearmanr": 0.7927684548355777,
"eval_steps_per_second": 2.843,
"step": 63
},
{
"epoch": 4.0,
"eval_averaged_scores": 0.8502990995955461,
"eval_loss": 0.6199266314506531,
"eval_pearson": 0.8589863742460024,
"eval_runtime": 1.2175,
"eval_samples_per_second": 472.263,
"eval_spearmanr": 0.8416118249450897,
"eval_steps_per_second": 2.464,
"step": 84
},
{
"epoch": 5.0,
"eval_averaged_scores": 0.8604081673075137,
"eval_loss": 0.6110144853591919,
"eval_pearson": 0.8686425986396985,
"eval_runtime": 1.2493,
"eval_samples_per_second": 460.273,
"eval_spearmanr": 0.8521737359753287,
"eval_steps_per_second": 2.401,
"step": 105
},
{
"epoch": 6.0,
"eval_averaged_scores": 0.8723167486730705,
"eval_loss": 0.613908052444458,
"eval_pearson": 0.8805257346547132,
"eval_runtime": 1.0534,
"eval_samples_per_second": 545.829,
"eval_spearmanr": 0.8641077626914277,
"eval_steps_per_second": 2.848,
"step": 126
},
{
"epoch": 7.0,
"eval_averaged_scores": 0.8747559757803227,
"eval_loss": 0.5541242957115173,
"eval_pearson": 0.8830275606823972,
"eval_runtime": 1.2127,
"eval_samples_per_second": 474.165,
"eval_spearmanr": 0.8664843908782482,
"eval_steps_per_second": 2.474,
"step": 147
},
{
"epoch": 8.0,
"eval_averaged_scores": 0.878359797369688,
"eval_loss": 0.5588080883026123,
"eval_pearson": 0.886186199129566,
"eval_runtime": 1.0704,
"eval_samples_per_second": 537.158,
"eval_spearmanr": 0.8705333956098098,
"eval_steps_per_second": 2.803,
"step": 168
},
{
"epoch": 9.0,
"eval_averaged_scores": 0.8815656931695632,
"eval_loss": 0.5312618613243103,
"eval_pearson": 0.8892227567089677,
"eval_runtime": 1.0596,
"eval_samples_per_second": 542.641,
"eval_spearmanr": 0.8739086296301587,
"eval_steps_per_second": 2.831,
"step": 189
},
{
"epoch": 10.0,
"eval_averaged_scores": 0.8823497284035507,
"eval_loss": 0.5539296269416809,
"eval_pearson": 0.8899626934510708,
"eval_runtime": 1.1083,
"eval_samples_per_second": 518.798,
"eval_spearmanr": 0.8747367633560307,
"eval_steps_per_second": 2.707,
"step": 210
},
{
"epoch": 10.0,
"step": 210,
"total_flos": 2483955726584208.0,
"train_loss": 1.6283381870814733,
"train_runtime": 129.3055,
"train_samples_per_second": 400.138,
"train_steps_per_second": 1.624
}
],
"logging_steps": 500,
"max_steps": 210,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 2483955726584208.0,
"train_batch_size": 256,
"trial_name": null,
"trial_params": null
}