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: 0.0608
- Precision: 0.9290
- Recall: 0.9371
- F1: 0.9331
- 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 |
---|---|---|---|---|---|---|---|
0.2276 | 1.0 | 878 | 0.0685 | 0.9204 | 0.9246 | 0.9225 | 0.9814 |
0.0498 | 2.0 | 1756 | 0.0622 | 0.9238 | 0.9358 | 0.9298 | 0.9833 |
0.0298 | 3.0 | 2634 | 0.0608 | 0.9290 | 0.9371 | 0.9331 | 0.9840 |
Framework versions
- Transformers 4.11.2
- Pytorch 1.9.0+cu102
- Datasets 1.12.1
- Tokenizers 0.10.3
- Downloads last month
- 99
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Dataset used to train ueb1/distilbert-base-uncased-finetuned-ner
Evaluation results
- Precision on conll2003self-reported0.929
- Recall on conll2003self-reported0.937
- F1 on conll2003self-reported0.933
- Accuracy on conll2003self-reported0.984