roberta-base-finetuned-tc

This model is a fine-tuned version of klue/roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5510
• Accuracy: 0.8442
• F1: 0.8376
• Precision: 0.8466
• Recall: 0.8442

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: 2e-05
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	1.0	15	1.3600	0.5055	0.3395	0.2556	0.5055
No log	2.0	30	0.9601	0.6712	0.5846	0.6041	0.6712
No log	3.0	45	0.7381	0.7865	0.7621	0.7439	0.7865
No log	4.0	60	0.6402	0.8172	0.7964	0.7793	0.8172
No log	5.0	75	0.5886	0.8258	0.8074	0.8163	0.8258
No log	6.0	90	0.5714	0.8344	0.8213	0.8280	0.8344
No log	7.0	105	0.5618	0.8331	0.8233	0.8386	0.8331
No log	8.0	120	0.5559	0.8380	0.8307	0.8428	0.8380
No log	9.0	135	0.5510	0.8442	0.8376	0.8466	0.8442
No log	10.0	150	0.5545	0.8429	0.8355	0.8454	0.8429

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2