sa_BERT_48_qqp
This model is a fine-tuned version of gokuls/bert_base_48 on the GLUE QQP dataset. It achieves the following results on the evaluation set:
- Loss: 0.3425
- Accuracy: 0.8510
- F1: 0.7996
- Combined Score: 0.8253
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: 4e-05
- train_batch_size: 96
- eval_batch_size: 96
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Combined Score |
|---|---|---|---|---|---|---|
| 0.4679 | 1.0 | 3791 | 0.3795 | 0.8222 | 0.7705 | 0.7964 |
| 0.3469 | 2.0 | 7582 | 0.3580 | 0.8447 | 0.7963 | 0.8205 |
| 0.2868 | 3.0 | 11373 | 0.3425 | 0.8510 | 0.7996 | 0.8253 |
| 0.2372 | 4.0 | 15164 | 0.3706 | 0.8561 | 0.8149 | 0.8355 |
| 0.1938 | 5.0 | 18955 | 0.3679 | 0.8625 | 0.8197 | 0.8411 |
| 0.1567 | 6.0 | 22746 | 0.4246 | 0.8639 | 0.8214 | 0.8427 |
| 0.1294 | 7.0 | 26537 | 0.4047 | 0.8585 | 0.8189 | 0.8387 |
| 0.1059 | 8.0 | 30328 | 0.5063 | 0.8579 | 0.8181 | 0.8380 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.0
- Tokenizers 0.13.3