my_doccls_korean_model
This model is a fine-tuned version of beomi/kcbert-base on the nsmc dataset. It achieves the following results on the evaluation set:
- Loss: 0.2942
- Accuracy: 0.9037
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.267 | 1.0 | 2344 | 0.2482 | 0.8987 |
| 0.1751 | 2.0 | 4688 | 0.2523 | 0.9024 |
| 0.1108 | 3.0 | 7032 | 0.2942 | 0.9037 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3