distilbert-base-uncased-finetuned-cola
This model is a fine-tuned version of distilbert-base-uncased on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 1.0043
- Matthews Correlation: 0.5408
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: 8.291414308409856e-06
- train_batch_size: 4
- eval_batch_size: 16
- seed: 18
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
|---|---|---|---|---|
| 0.5098 | 1.0 | 2138 | 0.6569 | 0.3562 |
| 0.4689 | 2.0 | 4276 | 0.6456 | 0.5238 |
| 0.3279 | 3.0 | 6414 | 0.8929 | 0.5047 |
| 0.2835 | 4.0 | 8552 | 0.9614 | 0.5251 |
| 0.226 | 5.0 | 10690 | 1.0043 | 0.5408 |
Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 4
Dataset used to train ducatte/distilbert-base-uncased-finetuned-cola
Evaluation results
- Matthews Correlation on gluevalidation set self-reported0.541