Basic PyTorch implementation of the Levenberg-Marquardt algorithm
Project description
levmarq-torch
A basic PyTorch implementation of the Levenberg-Marquardt algorithm. This solves minimization problems of the form
$$\mathbf{x}^* = \mathrm{argmin}_{\mathbf{x}} |\mathbf{y} - \mathbf{\hat{y}}(\mathbf{x})|^2 , .$$
The implementation is batched over the parameters $\mathbf{x}$ and datapoints $\mathbf{y}$.
Based on implementation 1 from Gavin 2022 and some help from Connor Stone.
Installation
Running
git clone git@github.com:adam-coogan/levmarq-torch.git
cd levmarq-torch
pip install .
will install the levmarq_torch
package.
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
levmarq_torch-0.0.1.tar.gz
(5.0 kB
view hashes)
Built Distribution
Close
Hashes for levmarq_torch-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | ecaf5ce261b7eae50b62fa4cb954fe487964da58658f262520bf9de3ae55d507 |
|
MD5 | edadfcb631a5d7c0792e92a2ed949f41 |
|
BLAKE2b-256 | f0da868a90132aac346f5b81b461e73c5fd2efb72759a50729e8c241958c7027 |