distilbert-qa-checkpoint-v4
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8092
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.0541 | 1.0 | 1083 | 0.9490 |
| 0.0494 | 2.0 | 2166 | 0.9200 |
| 0.0913 | 3.0 | 3249 | 0.6719 |
| 0.0935 | 4.0 | 4332 | 0.6882 |
| 0.0768 | 5.0 | 5415 | 0.6854 |
| 0.0732 | 6.0 | 6498 | 0.7032 |
| 0.0768 | 7.0 | 7581 | 0.6902 |
| 0.0755 | 8.0 | 8664 | 0.8092 |
Framework versions
- Transformers 4.27.2
- Pytorch 1.13.1+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3