defeasible-nli-bert-clf
This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.3952
- Accuracy: 0.5
- F1: 0.6667
- Precision: 0.5
- Recall: 1.0
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: 16
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 1.4816 | 1.0 | 1 | 1.3908 | 0.5 | 0.6667 | 0.5 | 1.0 |
| 1.3623 | 2.0 | 2 | 1.3927 | 0.5 | 0.6667 | 0.5 | 1.0 |
| 1.3392 | 3.0 | 3 | 1.3952 | 0.5 | 0.6667 | 0.5 | 1.0 |
Framework versions
- Transformers 5.14.1
- Pytorch 2.5.1+cu121
- Datasets 5.0.1
- Tokenizers 0.22.2