|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: google-bert/bert-base-multilingual-cased |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- told-br |
|
metrics: |
|
- precision |
|
- recall |
|
- accuracy |
|
- f1 |
|
model-index: |
|
- name: bert-base-multilingual-cased-finetuned-hate-speech-ptbr |
|
results: |
|
- task: |
|
name: Text Classification |
|
type: text-classification |
|
dataset: |
|
name: told-br |
|
type: told-br |
|
config: binary |
|
split: validation |
|
args: binary |
|
metrics: |
|
- name: Precision |
|
type: precision |
|
value: 0.702020202020202 |
|
- name: Recall |
|
type: recall |
|
value: 0.7654185022026432 |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.758095238095238 |
|
- name: F1 |
|
type: f1 |
|
value: 0.7590123199739615 |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# bert-base-multilingual-cased-finetuned-hate-speech-ptbr |
|
|
|
This model is a fine-tuned version of [google-bert/bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) on the told-br dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.6224 |
|
- Precision: 0.7020 |
|
- Recall: 0.7654 |
|
- Accuracy: 0.7581 |
|
- F1: 0.7590 |
|
|
|
## 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 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: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Accuracy | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:--------:|:------:| |
|
| 0.5127 | 1.0 | 1050 | 0.4978 | 0.6500 | 0.8756 | 0.7424 | 0.7418 | |
|
| 0.4415 | 2.0 | 2100 | 0.5206 | 0.7143 | 0.7104 | 0.7519 | 0.7518 | |
|
| 0.3623 | 3.0 | 3150 | 0.6204 | 0.6747 | 0.8293 | 0.7533 | 0.7542 | |
|
| 0.283 | 4.0 | 4200 | 0.6224 | 0.7020 | 0.7654 | 0.7581 | 0.7590 | |
|
| 0.2196 | 5.0 | 5250 | 0.7572 | 0.6954 | 0.7742 | 0.7557 | 0.7568 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.46.3 |
|
- Pytorch 2.5.1+cu121 |
|
- Datasets 3.1.0 |
|
- Tokenizers 0.20.3 |
|
|