Skip to content
TGM: PyTorch Geometry
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
docker dockerise travis ci script Jan 17, 2019
examples refactor docs page and tutorials Jan 9, 2019
test fix and refactor test_warp_perspective Apr 3, 2019
torchgeometry Fix inf Apr 2, 2019
.coveragerc add codecov Oct 2, 2018
.gitignore add make file commands to travis Mar 11, 2019
.travis.yml add make file commands to travis Mar 11, 2019
CONTRIBUTING.rst fix import __version__ in docs/ Jan 17, 2019
LICENSE Update LICENSE Sep 28, 2018
Makefile fix typo in makefile Mar 11, 2019
README.rst update readme.rst with codecov report Jan 19, 2019 Added first tests checking cpu / gpu Mar 7, 2019
mypy_files.txt add normalize confusion matrix and return as torch.float32 Mar 26, 2019 fix sdk_dir in conda environment Oct 1, 2018
readthedocs.yml Create readthedocs.yml Jan 19, 2019
setup.cfg update to pytorch v1.0.1 Feb 11, 2019 use pytorch-cpu==1.0.1 in travis Mar 25, 2019
tox.ini Added lint exception Mar 8, 2019 dockerise travis ci script Jan 17, 2019

README.rst Documentation Status

The PyTorch Geometry package is a geometric computer vision library for PyTorch.

It consists of a set of routines and differentiable modules to solve generic geometry computer vision problems. At its core, the package uses PyTorch as its main backend both for efficiency and to take advantage of the reverse-mode auto-differentiation to define and compute the gradient of complex functions.


From pip:

pip install torchgeometry

From source:

python install

From source using pip:

pip install git+

Quick Usage

import torch
import torchgeometry as tgm

x_rad = tgm.pi * torch.rand(1, 3, 3)
x_deg = tgm.rad2deg(x_rad)

torch.allclose(x_rad, tgm.deg2rad(x_deg))  # True


Run our Jupyter notebooks examples to learn to use the library.


If you are using torchgeometry in your research-related documents, it is recommended that you cite the poster.

 author    = {E. Riba, M. Fathollahi, W. Chaney, E. Rublee and G. Bradski}
 title     = {torchgeometry: when PyTorch meets geometry},
 booktitle = {PyTorch Developer Conference},
 year      = {2018},
 url       = {}


We appreciate all contributions. If you are planning to contribute back bug-fixes, please do so without any further discussion. If you plan to contribute new features, utility functions or extensions, please first open an issue and discuss the feature with us. Please, consider to read the CONTRIBUTING notes.

You can’t perform that action at this time.