bert-small-finetuned-wnut17-ner
This model is a fine-tuned version of google/bert_uncased_L-4_H-512_A-8 on the wnut_17 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3649
- Precision: 0.6259
- Recall: 0.4043
- F1: 0.4913
- Accuracy: 0.9255
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 | 0.3578 | 0.6382 | 0.3481 | 0.4505 | 0.9229 |
| 0.2359 | 2.0 | 850 | 0.3708 | 0.6535 | 0.3768 | 0.4780 | 0.9245 |
| 0.1231 | 3.0 | 1275 | 0.3649 | 0.6259 | 0.4043 | 0.4913 | 0.9255 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1