search

rusty1s / pytorch_cluster

PyTorch Extension Library of Optimized Graph Cluster Algorithms
MIT License
813 stars 146 forks source link

Implement weighted random walks #141

Open pbielak opened 2 years ago

pbielak commented 2 years ago

In the Node2vec paper, the transition probabilities are computed based on both edge weights and the hyperparameters p and q. The current implementation does not include the edge weights of a graph. This feature was already requested in https://github.com/rusty1s/pytorch_cluster/issues/88 and https://github.com/rusty1s/pytorch_cluster/issues/115. Based on the rejection sampling method used for the current implementation [1], I implemented a version that includes edge weights.

The PR for this feature implementation is https://github.com/rusty1s/pytorch_cluster/pull/140

[1] https://louisabraham.github.io/articles/node2vec-sampling.html

github-actions[bot] commented 1 year ago

This issue had no activity for 6 months. It will be closed in 2 weeks unless there is some new activity. Is this issue already resolved?