distilbert-base-cased-finetuned-ner_0301_J_DATA
This model is a fine-tuned version of distilbert-base-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0356
- Precision: 0.9588
- Recall: 0.9664
- F1: 0.9626
- Accuracy: 0.9933
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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.2272 | 1.0 | 705 | 0.0711 | 0.8795 | 0.9249 | 0.9016 | 0.9777 |
| 0.0551 | 2.0 | 1410 | 0.0439 | 0.9222 | 0.9563 | 0.9389 | 0.9878 |
| 0.0253 | 3.0 | 2115 | 0.0397 | 0.9312 | 0.9563 | 0.9436 | 0.9898 |
| 0.0277 | 4.0 | 2820 | 0.0500 | 0.9492 | 0.9641 | 0.9566 | 0.9898 |
| 0.018 | 5.0 | 3525 | 0.0414 | 0.9524 | 0.9652 | 0.9588 | 0.9902 |
| 0.0154 | 6.0 | 4230 | 0.0383 | 0.9397 | 0.9608 | 0.9501 | 0.9892 |
| 0.0133 | 7.0 | 4935 | 0.0454 | 0.9408 | 0.9619 | 0.9512 | 0.9888 |
| 0.0073 | 8.0 | 5640 | 0.0343 | 0.9496 | 0.9709 | 0.9601 | 0.9917 |
| 0.0071 | 9.0 | 6345 | 0.0295 | 0.9524 | 0.9652 | 0.9588 | 0.9923 |
| 0.0053 | 10.0 | 7050 | 0.0307 | 0.9449 | 0.9619 | 0.9533 | 0.9929 |
| 0.0048 | 11.0 | 7755 | 0.0221 | 0.9304 | 0.9585 | 0.9442 | 0.9935 |
| 0.0022 | 12.0 | 8460 | 0.0338 | 0.9450 | 0.9630 | 0.9539 | 0.9923 |
| 0.0024 | 13.0 | 9165 | 0.0263 | 0.9578 | 0.9675 | 0.9626 | 0.9938 |
| 0.0017 | 14.0 | 9870 | 0.0345 | 0.9547 | 0.9697 | 0.9622 | 0.9935 |
| 0.0016 | 15.0 | 10575 | 0.0356 | 0.9588 | 0.9664 | 0.9626 | 0.9933 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.13.0+cu117
- Datasets 2.8.0
- Tokenizers 0.12.1