DIMP

Self-supervised Graph Neural Networks via Diverse and Interactive Message Passing.

Dependencies

• torch 1.7.1
• sklearn 0.21.3
• numpy 1.16.0
• scipy 1.5.4
• munkres 1.1.4

Usage

Train and evaluate the node_classify model by executing

python -u execute.py --dataset cora --nb_epochs 2000 --patience 20 --numb_start 4 --numb_end 5 --chu_start 1 --chu_end 2 --chu_strip 0.5 --h1 -0.1 --h2 -0.1 --k_numb1 0 --k_numb2 1

The --dataset argument should be one of [ cora, citeseer, pubmed, amazon_electronics_computers, amazon_electronics_photo, ms_academic_cs, ms_academic_phy ].

Train and evaluate the graph_classify model by executing

python graph_classify/graph_classify.py

The --dataset argument should be one of [ MUTAG, PTC_MR, IMDB-BINARY, IMDB-MULTI, REDDIT-BINARY, REDDIT-MULTI-5K].