punctuation-nilc-bert-base
This model is a fine-tuned version of neuralmind/bert-base-portuguese-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1649
- Precision: 0.8928
- Recall: 0.8830
- F1: 0.8878
- Accuracy: 0.9724
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 |
|---|---|---|---|---|---|---|---|
| 0.0826 | 1.0 | 1172 | 0.0759 | 0.9289 | 0.8257 | 0.8743 | 0.9702 |
| 0.0457 | 2.0 | 2344 | 0.0908 | 0.9088 | 0.8664 | 0.8871 | 0.9726 |
| 0.0198 | 3.0 | 3516 | 0.1102 | 0.8912 | 0.8825 | 0.8869 | 0.9724 |
| 0.0092 | 4.0 | 4688 | 0.1314 | 0.8984 | 0.8774 | 0.8878 | 0.9726 |
| 0.0038 | 5.0 | 5860 | 0.1649 | 0.8928 | 0.8830 | 0.8878 | 0.9724 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2