punctuation-nilc-bert-large
This model is a fine-tuned version of neuralmind/bert-large-portuguese-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1585
- Precision: 0.9053
- Recall: 0.8923
- F1: 0.8988
- Accuracy: 0.9755
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: 5e-05
- 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: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0742 | 1.0 | 1172 | 0.0653 | 0.9194 | 0.8702 | 0.8941 | 0.9742 |
| 0.0396 | 2.0 | 2344 | 0.0773 | 0.9088 | 0.8834 | 0.8959 | 0.9748 |
| 0.0153 | 3.0 | 3516 | 0.1171 | 0.8996 | 0.8817 | 0.8906 | 0.9739 |
| 0.0059 | 4.0 | 4688 | 0.1390 | 0.9174 | 0.8719 | 0.8941 | 0.9747 |
| 0.0024 | 5.0 | 5860 | 0.1585 | 0.9053 | 0.8923 | 0.8988 | 0.9755 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2