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: 0.0617
- Precision: 0.9262
- Recall: 0.9380
- F1: 0.9321
- Accuracy: 0.9840
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 |
|---|---|---|---|---|---|---|---|
| 0.2465 | 1.0 | 878 | 0.0727 | 0.9175 | 0.9199 | 0.9187 | 0.9808 |
| 0.0527 | 2.0 | 1756 | 0.0610 | 0.9245 | 0.9361 | 0.9303 | 0.9834 |
| 0.0313 | 3.0 | 2634 | 0.0617 | 0.9262 | 0.9380 | 0.9321 | 0.9840 |
Framework versions
- Transformers 4.12.5
- Pytorch 1.8.0
- Datasets 1.16.1
- Tokenizers 0.10.3
- Downloads last month
- 126
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Dataset used to train momo/distilbert-base-uncased-finetuned-ner
Evaluation results
- Precision on conll2003self-reported0.926
- Recall on conll2003self-reported0.938
- F1 on conll2003self-reported0.932
- Accuracy on conll2003self-reported0.984