distilbert-base-uncased-finetuned-ner
This model is a fine-tuned version of distilbert-base-uncased on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0594
- Precision: 0.9259
- Recall: 0.9390
- F1: 0.9324
- Accuracy: 0.9838
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 |
---|---|---|---|---|---|---|---|
0.2425 | 1.0 | 878 | 0.0684 | 0.9107 | 0.9216 | 0.9161 | 0.9805 |
0.0514 | 2.0 | 1756 | 0.0585 | 0.9253 | 0.9366 | 0.9309 | 0.9838 |
0.0311 | 3.0 | 2634 | 0.0594 | 0.9259 | 0.9390 | 0.9324 | 0.9838 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1
- Downloads last month
- 1
Finetuned from
Dataset used to train dev4952/distilbert-base-uncased-finetuned-ner
Evaluation results
- Precision on conll2003validation set self-reported0.926
- Recall on conll2003validation set self-reported0.939
- F1 on conll2003validation set self-reported0.932
- Accuracy on conll2003validation set self-reported0.984