distilbert_sa_GLUE_Experiment_qqp_192
This model is a fine-tuned version of distilbert-base-uncased on the GLUE QQP dataset. It achieves the following results on the evaluation set:
- Loss: 0.4568
- Accuracy: 0.7910
- F1: 0.7234
- Combined Score: 0.7572
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: 256
- eval_batch_size: 256
- 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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Combined Score |
|---|---|---|---|---|---|---|
| 0.5339 | 1.0 | 1422 | 0.5031 | 0.7551 | 0.6484 | 0.7018 |
| 0.4835 | 2.0 | 2844 | 0.4866 | 0.7650 | 0.6504 | 0.7077 |
| 0.4587 | 3.0 | 4266 | 0.4792 | 0.7694 | 0.6422 | 0.7058 |
| 0.4369 | 4.0 | 5688 | 0.4851 | 0.7745 | 0.6716 | 0.7230 |
| 0.4155 | 5.0 | 7110 | 0.4705 | 0.7791 | 0.6970 | 0.7380 |
| 0.3961 | 6.0 | 8532 | 0.4633 | 0.7858 | 0.7093 | 0.7476 |
| 0.3772 | 7.0 | 9954 | 0.4572 | 0.7908 | 0.7176 | 0.7542 |
| 0.3593 | 8.0 | 11376 | 0.4568 | 0.7910 | 0.7234 | 0.7572 |
| 0.3422 | 9.0 | 12798 | 0.4661 | 0.7927 | 0.7227 | 0.7577 |
| 0.3265 | 10.0 | 14220 | 0.4596 | 0.7983 | 0.7290 | 0.7636 |
| 0.3119 | 11.0 | 15642 | 0.4635 | 0.7977 | 0.7255 | 0.7616 |
| 0.2961 | 12.0 | 17064 | 0.4857 | 0.8008 | 0.7309 | 0.7659 |
| 0.2831 | 13.0 | 18486 | 0.4987 | 0.8037 | 0.7314 | 0.7676 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2