distilbert-base-uncased_fakenews_identification

This model is a fine-tuned version of distilbert-base-uncased on the below dataset. https://www.kaggle.com/clmentbisaillon/fake-and-real-news-dataset It achieves the following results on the evaluation set:

• Loss: 0.0059
• Accuracy: 0.999
• F1: 0.9990

Label Description

LABEL_0 - Fake News LABEL_1 - Real News

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.0014	1.0	1000	0.0208	0.9965	0.9965
0.0006	2.0	2000	0.0041	0.9994	0.9994
0.0006	3.0	3000	0.0044	0.9992	0.9993
0.0	4.0	4000	0.0059	0.999	0.9990

Framework versions

• Transformers 4.16.0
• Pytorch 1.10.0+cu111
• Datasets 2.0.0
• Tokenizers 0.11.6