sup-VisualCSE-bert-base-uncased / trainer_state.json
gcyzsl's picture
Upload trainer_state.json with huggingface_hub
2c58b5c
{
"best_metric": 0.8642076134243076,
"best_model_checkpoint": "result/my-sup-simcse-bert-base-uncased/SupCon-22-0.5-1e-7-0.05-0.00",
"epoch": 4.996296296296296,
"global_step": 1345,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.46,
"eval_avg_sts": 0.830579264692183,
"eval_sickr_spearman": 0.8086463403909065,
"eval_stsb_spearman": 0.8525121889934594,
"step": 125
},
{
"epoch": 0.93,
"eval_avg_sts": 0.8322678052622576,
"eval_sickr_spearman": 0.8059119778171362,
"eval_stsb_spearman": 0.858623632707379,
"step": 250
},
{
"epoch": 1.39,
"eval_avg_sts": 0.8321564310903964,
"eval_sickr_spearman": 0.8062460341276383,
"eval_stsb_spearman": 0.8580668280531545,
"step": 375
},
{
"epoch": 1.86,
"learning_rate": 3.148148148148148e-05,
"loss": 0.6799,
"step": 500
},
{
"epoch": 1.86,
"eval_avg_sts": 0.8379634197959601,
"eval_sickr_spearman": 0.8117192261676126,
"eval_stsb_spearman": 0.8642076134243076,
"step": 500
},
{
"epoch": 2.32,
"eval_avg_sts": 0.8300517473847898,
"eval_sickr_spearman": 0.800890710410593,
"eval_stsb_spearman": 0.8592127843589866,
"step": 625
},
{
"epoch": 2.79,
"eval_avg_sts": 0.8331867449128545,
"eval_sickr_spearman": 0.8066284097261871,
"eval_stsb_spearman": 0.8597450800995219,
"step": 750
},
{
"epoch": 3.25,
"eval_avg_sts": 0.8318085085379562,
"eval_sickr_spearman": 0.8031442336278662,
"eval_stsb_spearman": 0.8604727834480461,
"step": 875
},
{
"epoch": 3.71,
"learning_rate": 1.2962962962962962e-05,
"loss": 0.4197,
"step": 1000
},
{
"epoch": 3.71,
"eval_avg_sts": 0.8353804358596084,
"eval_sickr_spearman": 0.8108888644859505,
"eval_stsb_spearman": 0.8598720072332664,
"step": 1000
},
{
"epoch": 4.18,
"eval_avg_sts": 0.8325858444728933,
"eval_sickr_spearman": 0.8039816078503385,
"eval_stsb_spearman": 0.8611900810954483,
"step": 1125
},
{
"epoch": 4.64,
"eval_avg_sts": 0.833323493417157,
"eval_sickr_spearman": 0.8065565551984356,
"eval_stsb_spearman": 0.8600904316358784,
"step": 1250
},
{
"epoch": 5.0,
"step": 1345,
"total_flos": 87911398260080640,
"train_runtime": 1044.4281,
"train_samples_per_second": 1.293
}
],
"max_steps": 1350,
"num_train_epochs": 5,
"total_flos": 87911398260080640,
"trial_name": null,
"trial_params": null
}