bert-finetuned-ner / README.md
librarian-bot's picture
Librarian Bot: Add base_model information to model
1baeab5
metadata
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - conll2003
metrics:
  - precision
  - recall
  - f1
  - accuracy
base_model: bert-base-cased
model-index:
  - name: bert-finetuned-ner
    results:
      - task:
          type: token-classification
          name: Token Classification
        dataset:
          name: conll2003
          type: conll2003
          args: conll2003
        metrics:
          - type: precision
            value: 0.9392329403951519
            name: Precision
          - type: recall
            value: 0.9520363513968361
            name: Recall
          - type: f1
            value: 0.9455913079816131
            name: F1
          - type: accuracy
            value: 0.9864308000235474
            name: Accuracy

bert-finetuned-ner

This model is a fine-tuned version of bert-base-cased on the conll2003 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0634
  • Precision: 0.9392
  • Recall: 0.9520
  • F1: 0.9456
  • Accuracy: 0.9864

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: 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: 3

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.0866 1.0 1756 0.0736 0.9157 0.9322 0.9239 0.9816
0.0382 2.0 3512 0.0663 0.9326 0.9472 0.9398 0.9855
0.0226 3.0 5268 0.0634 0.9392 0.9520 0.9456 0.9864

Framework versions

  • Transformers 4.13.0.dev0
  • Pytorch 1.10.0
  • Datasets 1.16.2.dev0
  • Tokenizers 0.10.3