tiny-vanilla-target-glue-qnli
This model is a fine-tuned version of google/bert_uncased_L-2_H-128_A-2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4624
- Accuracy: 0.7825
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 200
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6082 | 0.15 | 500 | 0.5375 | 0.7362 |
| 0.5378 | 0.31 | 1000 | 0.5192 | 0.7459 |
| 0.5161 | 0.46 | 1500 | 0.4967 | 0.7672 |
| 0.5097 | 0.61 | 2000 | 0.5182 | 0.7505 |
| 0.5092 | 0.76 | 2500 | 0.4728 | 0.7750 |
| 0.5011 | 0.92 | 3000 | 0.4660 | 0.7866 |
| 0.4889 | 1.07 | 3500 | 0.4476 | 0.7922 |
| 0.48 | 1.22 | 4000 | 0.4619 | 0.7840 |
| 0.4661 | 1.37 | 4500 | 0.4813 | 0.7741 |
| 0.4742 | 1.53 | 5000 | 0.4624 | 0.7825 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2