Home

Awesome

PyTorch Cluster

PyPI Version Testing Status Linting Status Code Coverage

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:

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

Installation

Anaconda

Update: You can now install pytorch-cluster via Anaconda for all major OS/PyTorch/CUDA combinations 🤗 Given that you have pytorch >= 1.8.0 installed, simply run

conda install pytorch-cluster -c pyg

Binaries

We alternatively provide pip wheels for all major OS/PyTorch/CUDA combinations, see here.

PyTorch 2.5

To install the binaries for PyTorch 2.5.0, simply run

pip install torch-cluster -f https://data.pyg.org/whl/torch-2.5.0+${CUDA}.html

where ${CUDA} should be replaced by either cpu, cu118, cu121, or cu124 depending on your PyTorch installation.

cpucu118cu121cu124
Linux✅✅✅✅
Windows✅✅✅✅
macOS✅

PyTorch 2.4

To install the binaries for PyTorch 2.4.0, simply run

pip install torch-cluster -f https://data.pyg.org/whl/torch-2.4.0+${CUDA}.html

where ${CUDA} should be replaced by either cpu, cu118, cu121, or cu124 depending on your PyTorch installation.

cpucu118cu121cu124
Linux✅✅✅✅
Windows✅✅✅✅
macOS✅

Note: Binaries of older versions are also provided for PyTorch 1.4.0, PyTorch 1.5.0, PyTorch 1.6.0, PyTorch 1.7.0/1.7.1, PyTorch 1.8.0/1.8.1, PyTorch 1.9.0, PyTorch 1.10.0/1.10.1/1.10.2, PyTorch 1.11.0, PyTorch 1.12.0/1.12.1, PyTorch 1.13.0/1.13.1, PyTorch 2.0.0/2.0.1, PyTorch 2.1.0/2.1.1/2.1.2, PyTorch 2.2.0/2.2.1/2.2.2, and PyTorch 2.3.0/2.3.1 (following the same procedure). For older versions, you need to explicitly specify the latest supported version number or install via pip install --no-index in order to prevent a manual installation from source. You can look up the latest supported version number here.

From source

Ensure that at least PyTorch 1.4.0 is installed and verify that cuda/bin and cuda/include are in your $PATH and $CPATH respectively, e.g.:

$ python -c "import torch; print(torch.__version__)"
>>> 1.4.0

$ python -c "import torch; print(torch.__version__)"
>>> 1.1.0

$ echo $PATH
>>> /usr/local/cuda/bin:...

$ echo $CPATH
>>> /usr/local/cuda/include:...

Then run:

pip install torch-cluster

When running in a docker container without NVIDIA driver, PyTorch needs to evaluate the compute capabilities and may fail. In this case, ensure that the compute capabilities are set via TORCH_CUDA_ARCH_LIST, e.g.:

export TORCH_CUDA_ARCH_LIST = "6.0 6.1 7.2+PTX 7.5+PTX"

Functions

Graclus

A greedy clustering algorithm of picking an unmarked vertex and matching it with one its unmarked neighbors (that maximizes its edge weight). The GPU algorithm is adapted from Fagginger Auer and Bisseling: A GPU Algorithm for Greedy Graph Matching (LNCS 2012)

import torch
from torch_cluster import graclus_cluster

row = torch.tensor([0, 1, 1, 2])
col = torch.tensor([1, 0, 2, 1])
weight = torch.tensor([1., 1., 1., 1.])  # Optional edge weights.

cluster = graclus_cluster(row, col, weight)

print(cluster)
tensor([0, 0, 1])

VoxelGrid

A clustering algorithm, which overlays a regular grid of user-defined size over a point cloud and clusters all points within a voxel.

import torch
from torch_cluster import grid_cluster

pos = torch.tensor([[0., 0.], [11., 9.], [2., 8.], [2., 2.], [8., 3.]])
size = torch.Tensor([5, 5])

cluster = grid_cluster(pos, size)

print(cluster)
tensor([0, 5, 3, 0, 1])

FarthestPointSampling

A sampling algorithm, which iteratively samples the most distant point with regard to the rest points.

import torch
from torch_cluster import fps

x = torch.tensor([[-1., -1.], [-1., 1.], [1., -1.], [1., 1.]])
batch = torch.tensor([0, 0, 0, 0])
index = fps(x, batch, ratio=0.5, random_start=False)

print(index)
tensor([0, 3])

kNN-Graph

Computes graph edges to the nearest k points.

Args:

import torch
from torch_cluster import knn_graph

x = torch.tensor([[-1., -1.], [-1., 1.], [1., -1.], [1., 1.]])
batch = torch.tensor([0, 0, 0, 0])
edge_index = knn_graph(x, k=2, batch=batch, loop=False)

print(edge_index)
tensor([[1, 2, 0, 3, 0, 3, 1, 2],
        [0, 0, 1, 1, 2, 2, 3, 3]])

Radius-Graph

Computes graph edges to all points within a given distance.

Args:

import torch
from torch_cluster import radius_graph

x = torch.tensor([[-1., -1.], [-1., 1.], [1., -1.], [1., 1.]])
batch = torch.tensor([0, 0, 0, 0])
edge_index = radius_graph(x, r=2.5, batch=batch, loop=False)

print(edge_index)
tensor([[1, 2, 0, 3, 0, 3, 1, 2],
        [0, 0, 1, 1, 2, 2, 3, 3]])

Nearest

Clusters points in x together which are nearest to a given query point in y. batch_{x,y} vectors need to be sorted.

import torch
from torch_cluster import nearest

x = torch.Tensor([[-1, -1], [-1, 1], [1, -1], [1, 1]])
batch_x = torch.tensor([0, 0, 0, 0])
y = torch.Tensor([[-1, 0], [1, 0]])
batch_y = torch.tensor([0, 0])
cluster = nearest(x, y, batch_x, batch_y)

print(cluster)
tensor([0, 0, 1, 1])

RandomWalk-Sampling

Samples random walks of length walk_length from all node indices in start in the graph given by (row, col).

import torch
from torch_cluster import random_walk

row = torch.tensor([0, 1, 1, 1, 2, 2, 3, 3, 4, 4])
col = torch.tensor([1, 0, 2, 3, 1, 4, 1, 4, 2, 3])
start = torch.tensor([0, 1, 2, 3, 4])

walk = random_walk(row, col, start, walk_length=3)

print(walk)
tensor([[0, 1, 2, 4],
        [1, 3, 4, 2],
        [2, 4, 2, 1],
        [3, 4, 2, 4],
        [4, 3, 1, 0]])

Running tests

pytest

C++ API

torch-cluster also offers a C++ API that contains C++ equivalent of python models.

export Torch_DIR=`python -c 'import torch;print(torch.utils.cmake_prefix_path)'`
mkdir build
cd build
# Add -DWITH_CUDA=on support for the CUDA if needed
cmake ..
make
make install