Skip to content

Latest commit

 

History

History
15 lines (9 loc) · 487 Bytes

README.md

File metadata and controls

15 lines (9 loc) · 487 Bytes
Raw

neighbor_sampling_heterogeneous_graphs

Master's Thesis - Neighborhood Sampling of Heterogeneous Graphs in Graph Neural Networks.

Paper: TODO

Running

There are two parts that should be run separately due to differences in the version of the libraries (mainly PyTorch). The first one is heterogeneous random walks, and the second one is a graph neural network.

Heterogeneous random walks:

  • PyTorch: 1.5.0a0+4ff3872, build from source

Graph neural network:

  • PyTorch: 1.9.0