Implementation of Collective Learning-GNN in PyTorch

This is the code for A Collective Learning Framework to Boost GNN Expressiveness

Author: Mengyue Hang

Requirements:

 PyTorch 1.2.0
 Python 3.6

 networkx==2.4
 numpy==1.17.3
 scipy==1.3.1

Usage:

 for unlabeled test data: cd clgnn; python train_unlabeled.py -h
 for partially-labeled test data: cd clgnn; python train_labeled.py -h

 A full list of parameters is shown in help message with -h.

 We provide Cora as example dataset. You can put your own dataset in the data/ for testing.

 e.g. to test GCN and CL-GCN (with our collective learning framework) performance on      unlabeled test data:
 python train_unlabeled.py --model_choice gcn_rand --iterations 2

 to test tk and CL-tk performance on partially-labeled test data:
 python train_labeled.py --model_choice tk_rand --baseline
 python train_labeled.py --model_choice tk_rand