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: 0.3540
- Precision: 0.5162
- Recall: 0.5373
- F1: 0.5265
- Accuracy: 0.8882
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 |
---|---|---|---|---|---|---|---|
0.6731 | 1.0 | 878 | 0.4315 | 0.5372 | 0.3880 | 0.4505 | 0.8681 |
0.3888 | 2.0 | 1756 | 0.3721 | 0.4748 | 0.5319 | 0.5017 | 0.8802 |
0.3067 | 3.0 | 2634 | 0.3540 | 0.5162 | 0.5373 | 0.5265 | 0.8882 |
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
- Downloads last month
- 16
Finetuned from
Dataset used to train harsh1304/distilbert-base-uncased-finetuned-ner
Evaluation results
- Precision on conll2003validation set self-reported0.516
- Recall on conll2003validation set self-reported0.537
- F1 on conll2003validation set self-reported0.527
- Accuracy on conll2003validation set self-reported0.888