<!-- 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. -->
ner-deBERTa-v2-x-large
This model is a fine-tuned version of microsoft/deberta-v3-base on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3963
- Precision: 0.7384
- Recall: 0.7378
- F1: 0.7381
- Accuracy: 0.9461
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 219 | 0.4082 | 0.6932 | 0.7087 | 0.7009 | 0.9386 |
| No log | 2.0 | 439 | 0.4299 | 0.7467 | 0.6948 | 0.7198 | 0.9426 |
| 0.0094 | 3.0 | 658 | 0.4086 | 0.7435 | 0.7072 | 0.7249 | 0.9441 |
| 0.0094 | 4.0 | 878 | 0.3873 | 0.7426 | 0.7420 | 0.7423 | 0.9461 |
| 0.0054 | 4.99 | 1095 | 0.3963 | 0.7384 | 0.7378 | 0.7381 | 0.9461 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
