bert-finetuned-ner
This model is a fine-tuned version of bert-base-cased on the wnut_17 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4341
- Precision: 0.5380
- Recall: 0.3216
- F1: 0.4026
- Accuracy: 0.9367
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 |
---|---|---|---|---|---|---|---|
No log | 1.0 | 425 | 0.4089 | 0.4827 | 0.2586 | 0.3368 | 0.9314 |
0.2032 | 2.0 | 850 | 0.4356 | 0.5435 | 0.2836 | 0.3727 | 0.9337 |
0.0815 | 3.0 | 1275 | 0.4341 | 0.5380 | 0.3216 | 0.4026 | 0.9367 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.0.0+cu117
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 0
Finetuned from
Dataset used to train shubingxl/bert-finetuned-ner
Evaluation results
- Precision on wnut_17test set self-reported0.538
- Recall on wnut_17test set self-reported0.322
- F1 on wnut_17test set self-reported0.403
- Accuracy on wnut_17test set self-reported0.937