<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert-base-casedfinetuned-fake-news-detection
This model is a fine-tuned version of distilbert-base-cased on the Fake and Reals News dataset. It achieves the following results on the evaluation set:
- Loss: 0.0019
- F1: 0.9998
- Accuracy: 0.9998
The Fake and Reals News dataset was used. It was stratified split into a train-val-test set (60/20/20).
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: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy |
|---|---|---|---|---|---|
| No log | 1.0 | 1684 | 0.0021 | 0.9998 | 0.9998 |
| No log | 2.0 | 3368 | 0.0019 | 0.9998 | 0.9998 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.0.0
- Tokenizers 0.11.6
