distilbert_add_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: 0.4419
- Accuracy: 0.7997
- F1: 0.7305
- Combined Score: 0.7651
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 |
|---|---|---|---|---|---|---|
| 0.5641 | 1.0 | 1422 | 0.5422 | 0.7304 | 0.6553 | 0.6929 |
| 0.5047 | 2.0 | 2844 | 0.5128 | 0.7504 | 0.6575 | 0.7039 |
| 0.487 | 3.0 | 4266 | 0.5020 | 0.7573 | 0.6656 | 0.7114 |
| 0.4729 | 4.0 | 5688 | 0.4907 | 0.7638 | 0.6695 | 0.7166 |
| 0.4502 | 5.0 | 7110 | 0.4759 | 0.7789 | 0.6736 | 0.7262 |
| 0.4139 | 6.0 | 8532 | 0.4635 | 0.7926 | 0.6909 | 0.7417 |
| 0.3728 | 7.0 | 9954 | 0.4419 | 0.7997 | 0.7305 | 0.7651 |
| 0.3334 | 8.0 | 11376 | 0.4760 | 0.8026 | 0.6964 | 0.7495 |
| 0.2998 | 9.0 | 12798 | 0.4597 | 0.8125 | 0.7314 | 0.7719 |
| 0.2704 | 10.0 | 14220 | 0.4692 | 0.8173 | 0.7403 | 0.7788 |
| 0.244 | 11.0 | 15642 | 0.4990 | 0.8195 | 0.7413 | 0.7804 |
| 0.2218 | 12.0 | 17064 | 0.5195 | 0.8198 | 0.7350 | 0.7774 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2