bert-finetuned-large
This model is a fine-tuned version of google-bert/bert-large-uncased on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0517
- Precision: 0.9433
- Recall: 0.9539
- F1: 0.9485
- Accuracy: 0.9894
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 |
|---|---|---|---|---|---|---|---|
| 0.0639 | 1.0 | 1756 | 0.0610 | 0.9126 | 0.9349 | 0.9236 | 0.9851 |
| 0.0281 | 2.0 | 3512 | 0.0524 | 0.9420 | 0.9504 | 0.9461 | 0.9883 |
| 0.0148 | 3.0 | 5268 | 0.0517 | 0.9433 | 0.9539 | 0.9485 | 0.9894 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 117
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for varunnagda/bert-finetuned-large
Base model
google-bert/bert-large-uncasedDataset used to train varunnagda/bert-finetuned-large
Evaluation results
- Precision on conll2003validation set self-reported0.943
- Recall on conll2003validation set self-reported0.954
- F1 on conll2003validation set self-reported0.949
- Accuracy on conll2003validation set self-reported0.989