<!-- 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. -->
deberta-v3-large-finetuned-ner
This model is a fine-tuned version of microsoft/deberta-v3-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4130
- Precision: 0.8219
- Recall: 0.8955
- F1: 0.8571
- Accuracy: 0.9310
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: cosine
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 45 | 1.0375 | 0.4072 | 0.2743 | 0.3278 | 0.7192 |
| No log | 2.0 | 90 | 0.7673 | 0.4724 | 0.3914 | 0.4281 | 0.7522 |
| No log | 3.0 | 135 | 0.6973 | 0.4892 | 0.6637 | 0.5633 | 0.7757 |
| No log | 4.0 | 180 | 0.6645 | 0.5209 | 0.7237 | 0.6058 | 0.7961 |
| No log | 5.0 | 225 | 0.4692 | 0.6618 | 0.7041 | 0.6823 | 0.8644 |
| No log | 6.0 | 270 | 0.4469 | 0.6902 | 0.7552 | 0.7213 | 0.8776 |
| No log | 7.0 | 315 | 0.4761 | 0.6713 | 0.8123 | 0.7351 | 0.8745 |
| No log | 8.0 | 360 | 0.3956 | 0.7524 | 0.8063 | 0.7784 | 0.9055 |
| No log | 9.0 | 405 | 0.4272 | 0.7298 | 0.8332 | 0.7781 | 0.8976 |
| No log | 10.0 | 450 | 0.4285 | 0.7520 | 0.8577 | 0.8014 | 0.9096 |
| No log | 11.0 | 495 | 0.4022 | 0.7764 | 0.8693 | 0.8202 | 0.9147 |
| 0.4557 | 12.0 | 540 | 0.3584 | 0.8090 | 0.8640 | 0.8356 | 0.9287 |
| 0.4557 | 13.0 | 585 | 0.4022 | 0.8102 | 0.8733 | 0.8405 | 0.9253 |
| 0.4557 | 14.0 | 630 | 0.4149 | 0.8067 | 0.8902 | 0.8464 | 0.9268 |
| 0.4557 | 15.0 | 675 | 0.4160 | 0.8188 | 0.8919 | 0.8538 | 0.9290 |
| 0.4557 | 16.0 | 720 | 0.4015 | 0.8173 | 0.8932 | 0.8536 | 0.9302 |
| 0.4557 | 17.0 | 765 | 0.4084 | 0.8215 | 0.8945 | 0.8565 | 0.9309 |
| 0.4557 | 18.0 | 810 | 0.4133 | 0.8219 | 0.8955 | 0.8571 | 0.9307 |
| 0.4557 | 19.0 | 855 | 0.4131 | 0.8217 | 0.8955 | 0.8570 | 0.9310 |
| 0.4557 | 20.0 | 900 | 0.4130 | 0.8219 | 0.8955 | 0.8571 | 0.9310 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
