Skip to content

Latest commit

 

History

History
30 lines (23 loc) · 602 Bytes

README.md

File metadata and controls

30 lines (23 loc) · 602 Bytes

Fake News Detector

Code implementation for fake news detection using GAT, GCN and GraphSAGE

Requirements

  • Numpy
  • Pytorch (torch_geometric)
  • scikit-learn (sklearn.metrics)

Dataset

Download dataset using this link

Run

python run.py

Configuration

For changing models you can pass --model='sage' or set value manually. The allowed values for models are:

  1. sage
  2. gcn
  3. gat

Other available args are as follows:

  • --batch_size
  • --lr
  • --hidden_size
  • --epochs
  • --feature