This Repository contains the project of COL865 Special topics in Commputer Application (on Multimodal and Tabular data)
This is an implementation of Frocc combined with GNNs. The code for ResGNN and BGNN are taken from their official implementation
The main code added are inside bgnn/bgnn/models/ in the files FRMCC.py, Frc_mlp.py. Also scripts have been provided in bgnn/scripts
Running instructions are in inside the bgnn folder's README.md file
Boost then Convolve: Gradient Boosting Meets Graph Neural Networks FROCC: Fast Random projection-based One-Class Classification