<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert-base-uncased-finetuned-ner
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1080
- Precision: 0.9631
- Recall: 0.9740
- F1: 0.9685
- Accuracy: 0.9774
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 14
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.1714 | 1.0 | 1492 | 0.1178 | 0.9209 | 0.9470 | 0.9337 | 0.9613 |
0.1018 | 2.0 | 2984 | 0.1062 | 0.9391 | 0.9546 | 0.9468 | 0.9668 |
0.0826 | 3.0 | 4476 | 0.0878 | 0.9460 | 0.9577 | 0.9518 | 0.9705 |
0.0621 | 4.0 | 5968 | 0.0945 | 0.9507 | 0.9663 | 0.9584 | 0.9728 |
0.0567 | 5.0 | 7460 | 0.0891 | 0.9562 | 0.9690 | 0.9625 | 0.9752 |
0.0422 | 6.0 | 8952 | 0.0805 | 0.9572 | 0.9699 | 0.9635 | 0.9776 |
0.0337 | 7.0 | 10444 | 0.0940 | 0.9575 | 0.9685 | 0.9629 | 0.9752 |
0.0282 | 8.0 | 11936 | 0.0940 | 0.9584 | 0.9717 | 0.9650 | 0.9756 |
0.0239 | 9.0 | 13428 | 0.0939 | 0.9579 | 0.9717 | 0.9647 | 0.9769 |
0.0213 | 10.0 | 14920 | 0.0980 | 0.9618 | 0.9745 | 0.9681 | 0.9775 |
0.0192 | 11.0 | 16412 | 0.1075 | 0.9606 | 0.9731 | 0.9668 | 0.9764 |
0.0156 | 12.0 | 17904 | 0.1046 | 0.9623 | 0.9734 | 0.9678 | 0.9774 |
0.0134 | 13.0 | 19396 | 0.1055 | 0.9614 | 0.9735 | 0.9674 | 0.9769 |
0.0125 | 14.0 | 20888 | 0.1080 | 0.9631 | 0.9740 | 0.9685 | 0.9774 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.13.0+cu117
- Datasets 2.8.0
- Tokenizers 0.12.1