chinese-bert-wwm-finetuned
This model is a fine-tuned version of hfl/chinese-bert-wwm on a very small subset of TNEWS dataset. It achieves the following results on the evaluation set:
- Loss: 4.2495
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: 3
- eval_batch_size: 3
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.4411 | 1.0 | 11 | 3.7846 |
| 3.3828 | 2.0 | 22 | 3.6168 |
| 3.5233 | 3.0 | 33 | 3.9919 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2