metadata
library_name: transformers
license: mit
base_model: neuralmind/bert-base-portuguese-cased
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: finetuned-bertimbau-portuguese-metaphors-4fold
results: []
finetuned-bertimbau-portuguese-metaphors-4fold
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: 0.2013
- Precision: 0.5294
- Recall: 0.4091
- F1: 0.4615
- Accuracy: 0.9536
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.4658 | 1.0 | 23 | 0.2788 | 0.0 | 0.0 | 0.0 | 0.9174 |
| 0.2197 | 2.0 | 46 | 0.2264 | 0.2 | 0.0909 | 0.1250 | 0.9391 |
| 0.1186 | 3.0 | 69 | 0.2154 | 0.7273 | 0.3636 | 0.4848 | 0.9536 |
| 0.0637 | 4.0 | 92 | 0.2114 | 0.5625 | 0.4091 | 0.4737 | 0.9478 |
| 0.0375 | 5.0 | 115 | 0.2013 | 0.5294 | 0.4091 | 0.4615 | 0.9536 |
Framework versions
- Transformers 4.57.2
- Pytorch 2.9.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1