reco-ner
This model is a fine-tuned version of microsoft/deberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0668
- Precision: 0.8125
- Recall: 0.8790
- F1: 0.8444
- Accuracy: 0.9819
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.4516 | 1.0 | 626 | 0.4047 | 0.4332 | 0.4564 | 0.4445 | 0.8980 |
| 0.3677 | 2.0 | 1252 | 0.2774 | 0.4918 | 0.5731 | 0.5293 | 0.9193 |
| 0.2892 | 3.0 | 1878 | 0.2133 | 0.6139 | 0.6581 | 0.6353 | 0.9384 |
| 0.2736 | 4.0 | 2504 | 0.1772 | 0.6248 | 0.6854 | 0.6537 | 0.9488 |
| 0.221 | 5.0 | 3130 | 0.1503 | 0.6295 | 0.7328 | 0.6772 | 0.9560 |
| 0.1569 | 6.0 | 3756 | 0.1283 | 0.6821 | 0.8108 | 0.7409 | 0.9623 |
| 0.1534 | 7.0 | 4382 | 0.0995 | 0.7412 | 0.8119 | 0.7749 | 0.9708 |
| 0.089 | 8.0 | 5008 | 0.0846 | 0.7695 | 0.8353 | 0.8010 | 0.9760 |
| 0.0923 | 9.0 | 5634 | 0.0743 | 0.7881 | 0.8740 | 0.8289 | 0.9789 |
| 0.0711 | 10.0 | 6260 | 0.0668 | 0.8125 | 0.8790 | 0.8444 | 0.9819 |
Framework versions
- Transformers 4.22.2
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1