gaialab-polarity-v1
This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8959
- F1 Macro: 0.5363
- Accuracy: 0.6435
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: 16
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.1
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | Accuracy |
|---|---|---|---|---|---|
| 1.0964 | 1.0 | 65 | 1.0278 | 0.4173 | 0.4609 |
| 0.9716 | 2.0 | 130 | 0.9479 | 0.4838 | 0.5565 |
| 0.8278 | 3.0 | 195 | 0.8955 | 0.5363 | 0.6435 |
| 0.4912 | 4.0 | 260 | 0.9730 | 0.4922 | 0.5652 |
| 0.4128 | 5.0 | 325 | 1.0079 | 0.5208 | 0.6087 |
| 0.2996 | 6.0 | 390 | 1.0841 | 0.5324 | 0.6261 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2
- Downloads last month
- 8