<!-- 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. -->
all-15-bert-finetuned-ner
This model is a fine-tuned version of bert-base-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0081
- Precision: 0.9630
- Recall: 0.9661
- F1: 0.9646
- Accuracy: 0.9987
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: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.014 | 1.0 | 6693 | 0.0080 | 0.9048 | 0.9363 | 0.9203 | 0.9976 |
0.007 | 2.0 | 13386 | 0.0070 | 0.9116 | 0.9459 | 0.9284 | 0.9976 |
0.0034 | 3.0 | 20079 | 0.0050 | 0.9514 | 0.9529 | 0.9522 | 0.9985 |
0.0027 | 4.0 | 26772 | 0.0065 | 0.9360 | 0.9618 | 0.9487 | 0.9982 |
0.002 | 5.0 | 33465 | 0.0062 | 0.9485 | 0.9555 | 0.9520 | 0.9984 |
0.0008 | 6.0 | 40158 | 0.0069 | 0.9498 | 0.9468 | 0.9483 | 0.9983 |
0.0013 | 7.0 | 46851 | 0.0059 | 0.9591 | 0.9618 | 0.9605 | 0.9987 |
0.0007 | 8.0 | 53544 | 0.0072 | 0.9635 | 0.9594 | 0.9614 | 0.9986 |
0.0003 | 9.0 | 60237 | 0.0076 | 0.9656 | 0.9621 | 0.9638 | 0.9987 |
0.0006 | 10.0 | 66930 | 0.0080 | 0.9598 | 0.9625 | 0.9611 | 0.9986 |
0.0007 | 11.0 | 73623 | 0.0072 | 0.9584 | 0.9651 | 0.9618 | 0.9986 |
0.0 | 12.0 | 80316 | 0.0073 | 0.9606 | 0.9658 | 0.9632 | 0.9987 |
0.0001 | 13.0 | 87009 | 0.0072 | 0.9649 | 0.9636 | 0.9642 | 0.9987 |
0.0 | 14.0 | 93702 | 0.0078 | 0.9629 | 0.9665 | 0.9647 | 0.9987 |
0.0 | 15.0 | 100395 | 0.0081 | 0.9630 | 0.9661 | 0.9646 | 0.9987 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.1