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.

Running

git clone git@github.com:adam-coogan/levmarq-torch.git
cd levmarq-torch
pip install .

will install the levmarq_torch package.

These details have not been verified by PyPI

This version

0.0.1

May 4, 2023

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Uploaded May 4, 2023 Source

Uploaded May 4, 2023 Python 3

Hashes for levmarq_torch-0.0.1.tar.gz
Algorithm	Hash digest
SHA256	`f40c59d1f837ccb2f1587411835f807bf853f6c31c6780a706cfcd46f5c8432a`
MD5	`9bd6090510361d03ed2714045326fd35`
BLAKE2b-256	`05f30c8ce59a7f3221887f1a96cb0a486e0ef361e4ecdb671fdf5e72dffc5a4c`

Hashes for levmarq_torch-0.0.1-py3-none-any.whl
Algorithm	Hash digest
SHA256	`ecaf5ce261b7eae50b62fa4cb954fe487964da58658f262520bf9de3ae55d507`
MD5	`edadfcb631a5d7c0792e92a2ed949f41`
BLAKE2b-256	`f0da868a90132aac346f5b81b461e73c5fd2efb72759a50729e8c241958c7027`