bert-finetuned-ner
This model is a fine-tuned version of bert-base-cased on the tweetner7 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4113
- Precision: 0.6748
- Recall: 0.6566
- F1: 0.6656
- Accuracy: 0.8729
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 | 312 | 0.4366 | 0.7320 | 0.5947 | 0.6562 | 0.8730 |
0.5402 | 2.0 | 624 | 0.4120 | 0.7271 | 0.6127 | 0.6650 | 0.8763 |
0.5402 | 3.0 | 936 | 0.4113 | 0.6748 | 0.6566 | 0.6656 | 0.8729 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.1
- Datasets 2.10.1
- Tokenizers 0.12.1
- Downloads last month
- 21
Evaluation results
- Precision on tweetner7self-reported0.675
- Recall on tweetner7self-reported0.657
- F1 on tweetner7self-reported0.666
- Accuracy on tweetner7self-reported0.873