distilbert-base-uncased-finetuned-SOMETHING
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: 0.7799
- Matthews Correlation: 0.5637
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- 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 |
|---|---|---|---|---|
| 0.5212 | 1.0 | 535 | 0.4654 | 0.4818 |
| 0.3438 | 2.0 | 1070 | 0.5179 | 0.4950 |
| 0.2279 | 3.0 | 1605 | 0.6018 | 0.5404 |
| 0.173 | 4.0 | 2140 | 0.7799 | 0.5637 |
| 0.1335 | 5.0 | 2675 | 0.8336 | 0.5541 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.12.1
- Downloads last month
- 4
Model tree for frtna/distilbert-base-uncased-finetuned-SOMETHING
Base model
distilbert/distilbert-base-uncasedDataset used to train frtna/distilbert-base-uncased-finetuned-SOMETHING
Evaluation results
- Matthews Correlation on gluevalidation set self-reported0.564