distilbert_sa_GLUE_Experiment_qqp_256
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: 0.4425
- Accuracy: 0.8030
- F1: 0.7323
- Combined Score: 0.7677
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 |
|---|---|---|---|---|---|---|
| 0.53 | 1.0 | 1422 | 0.5023 | 0.7557 | 0.6592 | 0.7075 |
| 0.479 | 2.0 | 2844 | 0.4823 | 0.7679 | 0.6483 | 0.7081 |
| 0.4522 | 3.0 | 4266 | 0.4788 | 0.7741 | 0.6474 | 0.7108 |
| 0.4263 | 4.0 | 5688 | 0.4753 | 0.7829 | 0.6911 | 0.7370 |
| 0.4009 | 5.0 | 7110 | 0.4536 | 0.7906 | 0.7194 | 0.7550 |
| 0.3772 | 6.0 | 8532 | 0.4497 | 0.7949 | 0.7200 | 0.7574 |
| 0.3548 | 7.0 | 9954 | 0.4453 | 0.8010 | 0.7201 | 0.7606 |
| 0.3332 | 8.0 | 11376 | 0.4425 | 0.8030 | 0.7323 | 0.7677 |
| 0.3132 | 9.0 | 12798 | 0.4654 | 0.7938 | 0.7375 | 0.7657 |
| 0.2951 | 10.0 | 14220 | 0.4551 | 0.8056 | 0.7423 | 0.7739 |
| 0.2777 | 11.0 | 15642 | 0.4675 | 0.8120 | 0.7374 | 0.7747 |
| 0.2625 | 12.0 | 17064 | 0.4946 | 0.8082 | 0.7451 | 0.7766 |
| 0.2473 | 13.0 | 18486 | 0.5041 | 0.8102 | 0.7469 | 0.7786 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2