distilbert_agnews_padding10model
This model is a fine-tuned version of distilbert-base-uncased on the ag_news dataset. It achieves the following results on the evaluation set:
- Loss: 0.6613
- Accuracy: 0.9445
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 | Accuracy |
|---|---|---|---|---|
| 0.1788 | 1.0 | 7500 | 0.1962 | 0.9405 |
| 0.1396 | 2.0 | 15000 | 0.1946 | 0.9446 |
| 0.1159 | 3.0 | 22500 | 0.2336 | 0.9433 |
| 0.0802 | 4.0 | 30000 | 0.2599 | 0.9437 |
| 0.055 | 5.0 | 37500 | 0.3196 | 0.9432 |
| 0.0408 | 6.0 | 45000 | 0.4017 | 0.9434 |
| 0.0338 | 7.0 | 52500 | 0.4113 | 0.9412 |
| 0.0258 | 8.0 | 60000 | 0.4533 | 0.9416 |
| 0.0159 | 9.0 | 67500 | 0.4573 | 0.9442 |
| 0.0154 | 10.0 | 75000 | 0.4980 | 0.9420 |
| 0.0171 | 11.0 | 82500 | 0.4935 | 0.9420 |
| 0.0105 | 12.0 | 90000 | 0.5304 | 0.9399 |
| 0.0079 | 13.0 | 97500 | 0.5437 | 0.9439 |
| 0.0061 | 14.0 | 105000 | 0.5889 | 0.9429 |
| 0.0056 | 15.0 | 112500 | 0.5444 | 0.9426 |
| 0.0059 | 16.0 | 120000 | 0.6274 | 0.9429 |
| 0.0004 | 17.0 | 127500 | 0.6264 | 0.9428 |
| 0.0014 | 18.0 | 135000 | 0.6138 | 0.9441 |
| 0.0002 | 19.0 | 142500 | 0.6431 | 0.9447 |
| 0.0005 | 20.0 | 150000 | 0.6613 | 0.9445 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3