LinSATNet offers a neural network layer to enforce the satisfiability of positive linear constraints to the output of neural networks. The gradient through the layer is exactly computed. This package now works with PyTorch.
Project description
LinSATNet
Official implementation of our ICML 2023 paper "LinSATNet: The Positive Linear Satisfiability Neural Networks".
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
LinSATNet-0.0.1.tar.gz
(6.3 kB
view hashes)
Built Distribution
Close
Hashes for LinSATNet-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 8ed11d932313fd0037106a22b480ccbbedccdc91a18d76e911a98ab2f8dea330 |
|
MD5 | f5aef1f497b75b10cce1af4c851164b4 |
|
BLAKE2b-256 | bbd9ce0c71a098a87550177b1218767c197fd1c1d276828c1aa125ccc6fae410 |