metadata
language:
- en
library_name: tf-keras
pipeline_tag: text-classification
Model description
NLP model for fake news classification using Roberta pre-trained model
Intended uses & limitations
Fake news classification
Training and evaluation data
Trained on 80% of English fake news classification dataset: https://www.kaggle.com/datasets/sadikaljarif/fake-news-detection-dataset-english
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
| Hyperparameters | Value |
|---|---|
| name | Adam |
| clipnorm | 0.8 |
| learning_rate | 2.9999999242136255e-05 |
| decay | 0.025 |
| beta_1 | 0.8999999761581421 |
| beta_2 | 0.9990000128746033 |
| epsilon | 9e-09 |
| amsgrad | False |
| training_precision | float32 |
Model Plot
View Model Plot
