metadata
license: mit
base_model: neuralmind/bert-base-portuguese-cased
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: hate_BERTimbau_v1
results: []
hate_BERTimbau_v1
This model is a fine-tuned version of neuralmind/bert-base-portuguese-cased on the Fortuna et al (2019) dataset. It achieves the following results on the evaluation set:
- Loss: 1.0655
- Precision: 0.7690
- Recall: 0.7690
- F1: 0.7690
- Accuracy: 0.7690
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: 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 |
|---|---|---|---|---|---|---|---|
| 0.5264 | 1.0 | 284 | 0.5050 | 0.7619 | 0.7619 | 0.7619 | 0.7619 |
| 0.4133 | 2.0 | 568 | 0.4516 | 0.7937 | 0.7937 | 0.7937 | 0.7937 |
| 0.289 | 3.0 | 852 | 0.6485 | 0.7531 | 0.7531 | 0.7531 | 0.7531 |
| 0.1804 | 4.0 | 1136 | 0.7828 | 0.7813 | 0.7813 | 0.7813 | 0.7813 |
| 0.1147 | 5.0 | 1420 | 1.0655 | 0.7690 | 0.7690 | 0.7690 | 0.7690 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1