Bug#1052148: ITP: pytorch-cluster -- PyTorch extension library of optimized graph cluster algorithms

[Andrius Merkys <merkys@debian.org>]

Package: wnpp
Owner: Andrius Merkys <merkys@debian.org>
Severity: wishlist
Control: block 970575 by -1

* Package name    : pytorch-cluster
  Version         : 1.6.1
  Upstream Author : Matthias Fey
* URL             : https://github.com/rusty1s/pytorch_cluster
* License         : Expat
  Programming Lang: C
  Description     : PyTorch extension library of optimized graph 
cluster algorithms (Python 3)

This package consists of a small extension library of highly optimized 
graph cluster algorithms for the use in PyTorch. The package consists of 
the following clustering algorithms:

  * Graclus from Dhillon et al.: Weighted Graph Cuts without 
Eigenvectors: A
    Multilevel Approach
  * Voxel Grid Pooling from, e.g., Simonovsky and Komodakis: Dynamic
    Edge-Conditioned Filters in Convolutional Neural Networks on Graphs
  * Iterative Farthest Point Sampling from, e.g. Qi et al.: PointNet++: 
Deep
    Hierarchical Feature Learning on Point Sets in a Metric Space
  * k-NN and Radius graph generation
  * Clustering based on nearest points
  * Random Walk Sampling from, e.g., Grover and Leskovec: node2vec: 
Scalable
    Feature Learning for Networks

All included operations work on varying data types and are implemented 
both for CPU and GPU.

pytorch-cluster is a test dependency for pytorch-geometric (ITP bug 
#970575).

Remark: This package is to be maintained with Debian Deep Learning Team at
   https://salsa.debian.org/deeplearning-team/pytorch-cluster