---
library_name: transformers
license: mit
base_model: neuralmind/bert-large-portuguese-cased
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: BingoGuard-bert-large-custom-only
  results: []
---

<!-- 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. -->

# BingoGuard-bert-large-custom-only

This model is a fine-tuned version of [neuralmind/bert-large-portuguese-cased](https://huggingface.co/neuralmind/bert-large-portuguese-cased) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6941
- Accuracy: 0.8936
- F1: 0.8894

## 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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 8

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 0.3343        | 1.0   | 67   | 0.2588          | 0.9021   | 0.8964 |
| 0.1935        | 2.0   | 134  | 0.2833          | 0.9021   | 0.8955 |
| 0.0663        | 3.0   | 201  | 0.4180          | 0.8936   | 0.8889 |
| 0.0251        | 4.0   | 268  | 0.5927          | 0.8915   | 0.8889 |
| 0.0164        | 5.0   | 335  | 0.5852          | 0.8766   | 0.8771 |
| 0.0096        | 6.0   | 402  | 0.6211          | 0.8979   | 0.8933 |
| 0.0023        | 7.0   | 469  | 0.6735          | 0.8957   | 0.8923 |
| 0.0022        | 8.0   | 536  | 0.6941          | 0.8936   | 0.8894 |


### Framework versions

- Transformers 4.55.4
- Pytorch 2.8.0+cu128
- Datasets 3.6.0
- Tokenizers 0.21.4