Fake_News_BERT_Classifier

This model is a fine-tuned version of distilbert-base-uncased trained on a Fake News Dataset It achieves the following results on the evaluation set:

• Loss: 0.2542

• Accuracy: 0.9688

• LABEL_0 = Fake news

• LABEL_1 = Real News

Model description

More information needed

Intended uses & limitations

This model was created for the purposes of UW IMT 575 project.

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1284	1.0	4490	0.1488	0.9582
0.12	2.0	8980	0.1704	0.9627
0.0741	3.0	13470	0.1971	0.9674
0.0202	4.0	17960	0.2265	0.9677
0.0465	5.0	22450	0.2542	0.9688

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3