deberta-v3-large-dapt-tapt-scientific-papers-pubmed-finetuned-DAGPap22
This model is a fine-tuned version of domenicrosati/deberta-v3-large-dapt-scientific-papers-pubmed-tapt on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0002
- Accuracy: 0.9998
- F1: 0.9999
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: 6e-06
- 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
- lr_scheduler_warmup_steps: 50
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.1884 | 1.0 | 669 | 0.0248 | 0.9951 | 0.9964 |
| 0.0494 | 2.0 | 1338 | 0.0084 | 0.9987 | 0.9990 |
| 0.0199 | 3.0 | 2007 | 0.0051 | 0.9991 | 0.9993 |
| 0.0079 | 4.0 | 2676 | 0.0030 | 0.9993 | 0.9995 |
| 0.0 | 5.0 | 3345 | 0.0026 | 0.9994 | 0.9996 |
| 0.0 | 6.0 | 4014 | 0.0014 | 0.9996 | 0.9997 |
| 0.0 | 7.0 | 4683 | 0.0015 | 0.9996 | 0.9997 |
| 0.0 | 8.0 | 5352 | 0.0011 | 0.9996 | 0.9997 |
| 0.0143 | 9.0 | 6021 | 0.0000 | 1.0 | 1.0 |
| 0.0 | 10.0 | 6690 | 0.0035 | 0.9991 | 0.9993 |
| 0.0 | 11.0 | 7359 | 0.0004 | 0.9998 | 0.9999 |
| 0.0 | 12.0 | 8028 | 0.0002 | 0.9998 | 0.9999 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1