facebookresearch/theseus
A library for differentiable nonlinear optimization
This library helps robotics and computer vision researchers develop advanced models by integrating complex optimization problems directly into their neural networks. You can input data like sensor readings or image pixels, define your optimization goals, and get optimized parameters or refined model predictions out. It's designed for researchers building sophisticated machine learning systems where traditional optimization needs to be part of the learning process.
2,008 stars. No commits in the last 6 months.
Use this if you are a researcher or advanced practitioner who needs to embed sophisticated, differentiable nonlinear optimization routines, like those found in robotics or computer vision algorithms, directly into PyTorch-based neural networks.
Not ideal if you are looking for a simple, off-the-shelf machine learning model or a general-purpose optimization library without the need for deep integration into neural network architectures.
Stars
2,008
Forks
143
Language
Python
License
MIT
Category
Last pushed
Jan 16, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/facebookresearch/theseus"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
nschaetti/EchoTorch
A Python toolkit for Reservoir Computing and Echo State Network experimentation based on...
metaopt/torchopt
TorchOpt is an efficient library for differentiable optimization built upon PyTorch.
opthub-org/pytorch-bsf
PyTorch implementation of Bezier simplex fitting
gpauloski/kfac-pytorch
Distributed K-FAC preconditioner for PyTorch
pytorch/xla
Enabling PyTorch on XLA Devices (e.g. Google TPU)