Torch Delaunay - The Delaunay triangulation library for PyTorch
This is a fast library for computing Delaunay triangulation of 2-dimensional points.
The implementation is based on a sweep-algorithm, introduced by David Sinclair1 and later improved by Volodymyr Agafonkin2.
Here is an example of triangulation output produced by Torch Delaunay library for a random set of
points:
Installation
The library is distributed as PyPI package, to install the package, execute the following command:
pip install torch_delaunay
You can use the torch_delaunay library for a fast computation of Delaunay triangulation for
points defined as PyTorch tensors.
Usage
import torch
from torch_delaunay.functional import shull2d
import matplotlib.pyplot as plt
# Compute Delaunay triangulation for randomly-generated 2-dimensional points.
points = torch.rand((100, 2))
simplices = shull2d(points)
# Plot the result
plt.triplot(points[:, 0], points[:, 1], triangles=simplices)
plt.title("Delaunay Triangulation")
Benchmarks
Benchmarks are collected on Apple M1 Pro chip with 32 GiB of memory.
| Prob. Distribution | Data Type | Samples | Time | CPU | Iterations |
|---|---|---|---|---|---|
| Uniform | Float64 | 10000 | 4.18 ms | 4.11 ms | 171 |
| Uniform | Float64 | 100000 | 49.6 ms | 48.3 ms | 15 |
| Uniform | Float64 | 1000000 | 600 ms | 592 ms | 1 |
| Uniform | Float32 | 10000 | 4.37 ms | 4.29 ms | 161 |
| Uniform | Float32 | 100000 | 47.4 ms | 46.4 ms | 15 |
| Uniform | Float32 | 1000000 | 573 ms | 568 ms | 1 |
| Normal | Float64 | 10000 | 4.60 ms | 4.49 ms | 156 |
| Normal | Float64 | 100000 | 48.5 ms | 47.4 ms | 15 |
| Normal | Float64 | 1000000 | 568 ms | 563 ms | 1 |
| Normal | Float32 | 10000 | 4.42 ms | 4.33 ms | 165 |
| Normal | Float32 | 100000 | 46.5 ms | 45.4 ms | 15 |
| Normal | Float32 | 1000000 | 550 ms | 547 ms | 1 |
License
The Torch Delaunay is distributed under GPLv3 license. See the LICENSE file for full license text.
-
David Sinclair - S-hull: a fast radial sweep-hull routine for Delaunay triangulation. ↩
-
Volodymyr Agafonkin - Delaunator: An incredibly fast and robust JavaScript library for Delaunay triangulation of 2D points. ↩
1 maintainer
Release files for torch-delaunay 1.3.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distributions (wheels)
Total release size: 66.0 MB
Release files / torch_delaunay-1.3.0-cp312-cp312-win_amd64.whl
| Download URL | torch_delaunay-1.3.0-cp312-cp312-win_amd64.whl |
|---|---|
| Size | 144.2 kB |
| Tags | CPython 3.12 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
3a2fb96ce8444f47bfc49d63ff84401eb36476fdcdf7c2c076da400c5e8988a4
|
|
BLAKE2b-256 checksum How to use checksums |
143c2825fca33e3ee0baa1714736d01c061ea6b33b352dfa203df3d3b183a35b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp312-cp312-manylinux_2_28_x86_64.whl
| Download URL | torch_delaunay-1.3.0-cp312-cp312-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 8.1 MB |
| Tags | CPython 3.12 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
2a43268cad8f336a016aa5aa4d138247cc37f1f041040413afd0b9f7ac1f579d
|
|
BLAKE2b-256 checksum How to use checksums |
ab8ee4f0319f3be5a2557ae282a36f9c564891a1f9aaca4290df273c6f5fd63e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp312-cp312-manylinux_2_28_aarch64.whl
| Download URL | torch_delaunay-1.3.0-cp312-cp312-manylinux_2_28_aarch64.whl |
|---|---|
| Size | 8.1 MB |
| Tags | CPython 3.12 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
df287715a5fd500525a2313dbb1b6644d60b244bd6229ad450c30cffcaf5600f
|
|
BLAKE2b-256 checksum How to use checksums |
8527debc3bc1c9b6fcf3a44cfda25ea0f1dfa2c467e43fec96f1ad14130a90ab
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp312-cp312-macosx_11_0_arm64.whl
| Download URL | torch_delaunay-1.3.0-cp312-cp312-macosx_11_0_arm64.whl |
|---|---|
| Size | 139.8 kB |
| Tags | CPython 3.12 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
d679d0e6d15550336c380d34fc3504b22d14866e2e1badda59ba3e00f3f4c51a
|
|
BLAKE2b-256 checksum How to use checksums |
4108442ee618112e31fe31005164e8b1acb65246f3691595d1ff01a7d3982b3e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp311-cp311-win_amd64.whl
| Download URL | torch_delaunay-1.3.0-cp311-cp311-win_amd64.whl |
|---|---|
| Size | 143.8 kB |
| Tags | CPython 3.11 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
d10d8288e1fc549b3296e20d41a758de26fdbb4789aead796366410fb6e6ac88
|
|
BLAKE2b-256 checksum How to use checksums |
df6ff51157f4cec09557e212488ea4bb9dc80c4a2c6448162a3ad09d0f6ae342
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp311-cp311-manylinux_2_28_x86_64.whl
| Download URL | torch_delaunay-1.3.0-cp311-cp311-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 8.2 MB |
| Tags | CPython 3.11 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
4a39b851890188247d7f0aef9a22f8665ce226b9b894099aca8f17f41188c442
|
|
BLAKE2b-256 checksum How to use checksums |
75239e1b69fe49ee5b37badae8305ff7a8917d1fe0917912ffeb6aea5a95ede6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp311-cp311-manylinux_2_28_aarch64.whl
| Download URL | torch_delaunay-1.3.0-cp311-cp311-manylinux_2_28_aarch64.whl |
|---|---|
| Size | 8.1 MB |
| Tags | CPython 3.11 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
8204faba943af62294b97d1c1c822a36295f638f717c84a715bfdc7f9e2340c8
|
|
BLAKE2b-256 checksum How to use checksums |
af79c7a6be27543cb35bd7fc482dea916f26c0b884cca96e78caf30b462c05eb
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp311-cp311-macosx_11_0_arm64.whl
| Download URL | torch_delaunay-1.3.0-cp311-cp311-macosx_11_0_arm64.whl |
|---|---|
| Size | 141.2 kB |
| Tags | CPython 3.11 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
369900da64081e1b6b1ec6907451919ba52500fca918d376e714d849a499e32c
|
|
BLAKE2b-256 checksum How to use checksums |
c216ad2fa647810d8c01eda4c7788896de2c784ef30b7e8d044bd435a31a0a4d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp310-cp310-win_amd64.whl
| Download URL | torch_delaunay-1.3.0-cp310-cp310-win_amd64.whl |
|---|---|
| Size | 142.9 kB |
| Tags | CPython 3.10 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
b007ea80a95fa38e706de1c3e77e0c1ea2229ec0a2b0676cd9fe1956a09b7209
|
|
BLAKE2b-256 checksum How to use checksums |
747b50af535de668592eb2a4ed59e2896afd781b2b34ba4689d5795e501cc670
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp310-cp310-manylinux_2_28_x86_64.whl
| Download URL | torch_delaunay-1.3.0-cp310-cp310-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 8.1 MB |
| Tags | CPython 3.10 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
44b2334edb05d4d86d373842b60dbe52187c3f87dc81a6cfea7d97fd61c43b27
|
|
BLAKE2b-256 checksum How to use checksums |
19f65e187ed99920aa07b4ec148edaf8bb6cdc3b08c29e7816f943ab2f138769
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp310-cp310-manylinux_2_28_aarch64.whl
| Download URL | torch_delaunay-1.3.0-cp310-cp310-manylinux_2_28_aarch64.whl |
|---|---|
| Size | 8.0 MB |
| Tags | CPython 3.10 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
ebaa720a6b481216888d94355200127bf67c786281626b6027688eb5232826db
|
|
BLAKE2b-256 checksum How to use checksums |
0ca444478a9a9de2a73e121aabd22fa60481c7464abad9e3efe980e403d8d478
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp310-cp310-macosx_11_0_arm64.whl
| Download URL | torch_delaunay-1.3.0-cp310-cp310-macosx_11_0_arm64.whl |
|---|---|
| Size | 139.6 kB |
| Tags | CPython 3.10 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
c54782bb135193528cefce2282efab0b4fdc3cddbb91de502554a4e5dc3211ce
|
|
BLAKE2b-256 checksum How to use checksums |
cc6738ab708731ab1d2efab79d251da7f1884f812a995f0e8e1dade4d48c3e89
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp39-cp39-win_amd64.whl
| Download URL | torch_delaunay-1.3.0-cp39-cp39-win_amd64.whl |
|---|---|
| Size | 143.4 kB |
| Tags | CPython 3.9 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
7fb7520868c137c58040ac4cbd1c0beaf8f96b5bab4ad1091e1480630b7ecfe0
|
|
BLAKE2b-256 checksum How to use checksums |
e007a1f5147b472d1b08a7ec9393b74943bc618e95ae80dd47dc5dd1d3a2586d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp39-cp39-manylinux_2_28_x86_64.whl
| Download URL | torch_delaunay-1.3.0-cp39-cp39-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 8.1 MB |
| Tags | CPython 3.9 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
31463ae9729ba8dda8d6074b945e194293d7ba0d4a8f62a6046ab634ab219e39
|
|
BLAKE2b-256 checksum How to use checksums |
18faf4d61968e4d9bd06000f562590c69164ef6db609b9d675c7591d5bc8efbc
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp39-cp39-manylinux_2_28_aarch64.whl
| Download URL | torch_delaunay-1.3.0-cp39-cp39-manylinux_2_28_aarch64.whl |
|---|---|
| Size | 8.1 MB |
| Tags | CPython 3.9 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
073f1b9eae5235024b47e709286532fffd5bb6479cdd8fcb099e7f417d2ae51d
|
|
BLAKE2b-256 checksum How to use checksums |
5777daa52351cb3f3c97ca3bb3aba4b2577a59faa42b92d6a416da761c04fb2d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency logRelease files / torch_delaunay-1.3.0-cp39-cp39-macosx_11_0_arm64.whl
| Download URL | torch_delaunay-1.3.0-cp39-cp39-macosx_11_0_arm64.whl |
|---|---|
| Size | 139.8 kB |
| Tags | CPython 3.9 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
f1df56718c62558947bb85a494427be8ceed234246936651d07c49c01d327ef1
|
|
BLAKE2b-256 checksum How to use checksums |
0e7d9166f51c109ed9c34c40898a851d0e0e37c460c3c44dbdf30f6d27c2f16b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 18, 2026.
Transparency log
