RoBERTa-ext-large-lora-chinese-finetuned-ner
This model is a fine-tuned version of hfl/chinese-roberta-wwm-ext-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3762
- Precision: 0.6284
- Recall: 0.7311
- F1: 0.6759
- Accuracy: 0.9107
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: 0.001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.5112 | 1.0 | 252 | 0.3440 | 0.5005 | 0.6105 | 0.5501 | 0.8940 |
| 0.283 | 2.0 | 504 | 0.3198 | 0.5363 | 0.6715 | 0.5963 | 0.9017 |
| 0.2373 | 3.0 | 756 | 0.3104 | 0.5506 | 0.7216 | 0.6246 | 0.9054 |
| 0.1995 | 4.0 | 1008 | 0.3210 | 0.5804 | 0.7236 | 0.6441 | 0.9092 |
| 0.1678 | 5.0 | 1260 | 0.3300 | 0.5828 | 0.7140 | 0.6418 | 0.9077 |
| 0.1435 | 6.0 | 1512 | 0.3274 | 0.5912 | 0.7173 | 0.6482 | 0.9104 |
| 0.1206 | 7.0 | 1764 | 0.3566 | 0.5964 | 0.7351 | 0.6585 | 0.9079 |
| 0.105 | 8.0 | 2016 | 0.3579 | 0.6065 | 0.7281 | 0.6618 | 0.9112 |
| 0.0925 | 9.0 | 2268 | 0.3645 | 0.6148 | 0.7382 | 0.6709 | 0.9103 |
| 0.0835 | 10.0 | 2520 | 0.3762 | 0.6284 | 0.7311 | 0.6759 | 0.9107 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0