tvtcm's picture
Model save
16d9228
|
raw
history blame
2.31 kB
metadata
license: mit
base_model: neuralmind/bert-base-portuguese-cased
tags:
  - generated_from_trainer
datasets:
  - lener_br
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: bert-base-portuguese-cased-finetuned-ner
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: lener_br
          type: lener_br
          config: lener_br
          split: test
          args: lener_br
        metrics:
          - name: Precision
            type: precision
            value: 0.8717564870259481
          - name: Recall
            type: recall
            value: 0.8995880535530381
          - name: F1
            type: f1
            value: 0.88545362392296
          - name: Accuracy
            type: accuracy
            value: 0.9836487420412604

bert-base-portuguese-cased-finetuned-ner

This model is a fine-tuned version of neuralmind/bert-base-portuguese-cased on the lener_br dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0652
  • Precision: 0.8718
  • Recall: 0.8996
  • F1: 0.8855
  • Accuracy: 0.9836

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: 3

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 490 0.0911 0.8063 0.7703 0.7879 0.9734
0.1901 2.0 980 0.0665 0.8525 0.8929 0.8722 0.9819
0.0419 3.0 1470 0.0652 0.8718 0.8996 0.8855 0.9836

Framework versions

  • Transformers 4.34.1
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1