---
license: mit
base_model: neuralmind/bert-base-portuguese-cased
tags:
- generated_from_trainer
datasets:
- harem
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: harem-ner
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: harem
      type: harem
      config: default
      split: test
      args: default
    metrics:
    - name: Precision
      type: precision
      value: 0.6869415807560137
    - name: Recall
      type: recall
      value: 0.7467314157639149
    - name: F1
      type: f1
      value: 0.7155897619473779
    - name: Accuracy
      type: accuracy
      value: 0.9527588964414234
---

<!-- 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. -->

# harem-ner

This model is a fine-tuned version of [neuralmind/bert-base-portuguese-cased](https://huggingface.co/neuralmind/bert-base-portuguese-cased) on the harem dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2411
- Precision: 0.6869
- Recall: 0.7467
- F1: 0.7156
- Accuracy: 0.9528

## 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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 300

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| No log        | 1.0   | 16   | 0.7683          | 0.0       | 0.0    | 0.0    | 0.8358   |
| No log        | 2.0   | 32   | 0.4727          | 0.3375    | 0.2955 | 0.3151 | 0.8803   |
| No log        | 3.0   | 48   | 0.3498          | 0.4859    | 0.4838 | 0.4848 | 0.9090   |
| No log        | 4.0   | 64   | 0.2771          | 0.5651    | 0.6223 | 0.5924 | 0.9354   |
| No log        | 5.0   | 80   | 0.2309          | 0.5901    | 0.6743 | 0.6294 | 0.9424   |
| No log        | 6.0   | 96   | 0.2195          | 0.6229    | 0.6997 | 0.6590 | 0.9469   |
| No log        | 7.0   | 112  | 0.2151          | 0.6239    | 0.6903 | 0.6554 | 0.9480   |
| No log        | 8.0   | 128  | 0.2178          | 0.6682    | 0.7236 | 0.6948 | 0.9504   |
| No log        | 9.0   | 144  | 0.2210          | 0.6808    | 0.7426 | 0.7104 | 0.9514   |
| No log        | 10.0  | 160  | 0.2292          | 0.6863    | 0.7348 | 0.7097 | 0.9512   |
| No log        | 11.0  | 176  | 0.2312          | 0.6932    | 0.7452 | 0.7183 | 0.9522   |
| No log        | 12.0  | 192  | 0.2258          | 0.6966    | 0.7523 | 0.7234 | 0.9535   |
| No log        | 13.0  | 208  | 0.2337          | 0.7076    | 0.7557 | 0.7309 | 0.9537   |
| No log        | 14.0  | 224  | 0.2299          | 0.6907    | 0.7549 | 0.7214 | 0.9533   |
| No log        | 15.0  | 240  | 0.2381          | 0.6980    | 0.7553 | 0.7255 | 0.9524   |
| No log        | 16.0  | 256  | 0.2411          | 0.6869    | 0.7467 | 0.7156 | 0.9528   |


### Framework versions

- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2