pubmed-abs-noise-mixed-v2
This model is a fine-tuned version of facebook/bart-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9626
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: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.5067 | 0.11 | 500 | 1.3888 |
| 1.373 | 0.21 | 1000 | 1.3029 |
| 1.2141 | 0.32 | 1500 | 1.2323 |
| 1.2791 | 0.43 | 2000 | 1.2643 |
| 1.0758 | 0.54 | 2500 | 1.1578 |
| 1.2029 | 0.64 | 3000 | 1.1410 |
| 1.109 | 0.75 | 3500 | 1.1115 |
| 1.1173 | 0.86 | 4000 | 1.0894 |
| 1.068 | 0.96 | 4500 | 1.0772 |
| 0.917 | 1.07 | 5000 | 1.0755 |
| 0.8664 | 1.18 | 5500 | 1.0731 |
| 0.9063 | 1.28 | 6000 | 1.0501 |
| 0.9207 | 1.39 | 6500 | 1.0393 |
| 0.9118 | 1.5 | 7000 | 1.0257 |
| 0.8558 | 1.61 | 7500 | 1.0107 |
| 0.8938 | 1.71 | 8000 | 1.0114 |
| 0.816 | 1.82 | 8500 | 0.9993 |
| 0.8644 | 1.93 | 9000 | 0.9856 |
| 0.7265 | 2.03 | 9500 | 1.0233 |
| 0.7326 | 2.14 | 10000 | 0.9946 |
| 0.6854 | 2.25 | 10500 | 0.9921 |
| 0.6851 | 2.35 | 11000 | 0.9826 |
| 0.6521 | 2.46 | 11500 | 0.9788 |
| 0.6671 | 2.57 | 12000 | 0.9734 |
| 0.7138 | 2.68 | 12500 | 0.9707 |
| 0.7599 | 2.78 | 13000 | 0.9653 |
| 0.6576 | 2.89 | 13500 | 0.9642 |
| 0.68 | 3.0 | 14000 | 0.9626 |
Framework versions
- Transformers 4.36.1
- Pytorch 2.0.1
- Datasets 2.16.1
- Tokenizers 0.15.0