metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: FakevsRealNews
results: []
FakevsRealNews
This model is a fine-tuned version of distilbert-base-uncased on "Fake and real news dataset" dataset.
Link to Dataset : https://www.kaggle.com/datasets/clmentbisaillon/fake-and-real-news-dataset
It achieves the following results on the evaluation set:
- Loss: 0.0006
- Accuracy: 0.6309
- F1: 0.7677
- Precision: 0.6233
- Recall: 0.9992
Model description
Finetuned Distilbert
Intended uses & limitations
More information needed
Training and evaluation data
The data was split into train-dev-test sets on a ratio of 80:10:10
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-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
- lr_scheduler_warmup_steps: 500
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.0176 | 1.0 | 1956 | 0.0009 | 0.9616 | 0.9695 | 0.9409 | 1.0 |
| 0.0014 | 2.0 | 3912 | 0.0015 | 0.9864 | 0.9890 | 0.9783 | 1.0 |
| 0.0011 | 3.0 | 5868 | 0.0008 | 0.7611 | 0.8363 | 0.7188 | 0.9996 |
| 0.0008 | 4.0 | 7824 | 0.0008 | 0.7872 | 0.8514 | 0.7418 | 0.9992 |
| 0.0006 | 5.0 | 9780 | 0.0006 | 0.6309 | 0.7677 | 0.6233 | 0.9992 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.1.0
- Tokenizers 0.12.1