nli_sts_roberta-large_lr1e-05_wd1e-03_ep3_lr1e-05_wd1e-03_ep3_ckpt

This model is a fine-tuned version of ys7yoo/sts_roberta-large_lr1e-05_wd1e-03_ep3 on the klue dataset. It achieves the following results on the evaluation set:

• Loss: 0.3529
• Accuracy: 0.894
• F1: 0.8939

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.5292	1.0	391	0.3853	0.857	0.8565
0.2468	2.0	782	0.3176	0.89	0.8899
0.1412	3.0	1173	0.3529	0.894	0.8939

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu117
• Datasets 2.13.0
• Tokenizers 0.13.3