metadata
language:
- pt
license: apache-2.0
tags:
- toxicity
- portuguese
- hate speech
- offensive language
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: dougtrajano/toxic-comment-classification
results: []
dougtrajano/toxic-comment-classification
This model is a fine-tuned version of neuralmind/bert-base-portuguese-cased on the OLID-BR dataset. It achieves the following results on the evaluation set:
- Loss: 0.5590
- Accuracy: 0.8578
- F1: 0.8580
- Precision: 0.8594
- Recall: 0.8578
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: 3.255788747459486e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 1993
- optimizer: Adam with betas=(0.8445637934160373,0.8338816842140165) and epsilon=2.527092625455385e-08
- lr_scheduler_type: linear
- num_epochs: 30
- label_smoothing_factor: 0.07158711257743958
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.4422 | 1.0 | 1408 | 0.4197 | 0.8466 | 0.8470 | 0.8505 | 0.8466 |
| 0.3566 | 2.0 | 2816 | 0.4724 | 0.8413 | 0.8394 | 0.8453 | 0.8413 |
| 0.3135 | 3.0 | 4224 | 0.4801 | 0.8447 | 0.8434 | 0.8470 | 0.8447 |
| 0.2638 | 4.0 | 5632 | 0.5590 | 0.8578 | 0.8580 | 0.8594 | 0.8578 |
| 0.2314 | 5.0 | 7040 | 0.5605 | 0.8491 | 0.8487 | 0.8489 | 0.8491 |
| 0.2221 | 6.0 | 8448 | 0.6369 | 0.8416 | 0.8414 | 0.8414 | 0.8416 |
| 0.1939 | 7.0 | 9856 | 0.6518 | 0.8400 | 0.8402 | 0.8405 | 0.8400 |
| 0.2015 | 8.0 | 11264 | 0.6042 | 0.8462 | 0.8457 | 0.8465 | 0.8462 |
| 0.1989 | 9.0 | 12672 | 0.6236 | 0.8500 | 0.8496 | 0.8499 | 0.8500 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.10.2+cu113
- Datasets 2.9.0
- Tokenizers 0.13.2