locuslab/optnet
OptNet: Differentiable Optimization as a Layer in Neural Networks
This project helps machine learning practitioners build more sophisticated neural networks. It allows you to embed well-defined optimization problems directly into your neural network models, rather than trying to approximate them with standard layers. This means you can combine the power of automatic feature learning with the precision of mathematical optimization, leading to more robust and accurate models for tasks like signal denoising or solving structured puzzles. It's for machine learning engineers and researchers who are developing advanced models.
577 stars. No commits in the last 6 months.
Use this if you are building neural networks for problems where some parts of the solution can be precisely described by mathematical optimization, and you want your network to learn how to incorporate these rules.
Not ideal if your problem doesn't involve sub-problems that can be framed as quadratic programs, or if you are looking for a simple, off-the-shelf solution without custom model building.
Stars
577
Forks
80
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 26, 2020
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/locuslab/optnet"
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.
gpauloski/kfac-pytorch
Distributed K-FAC preconditioner for PyTorch
opthub-org/pytorch-bsf
PyTorch implementation of Bezier simplex fitting
pytorch/xla
Enabling PyTorch on XLA Devices (e.g. Google TPU)