Using JAX to Improve Separable Image Filters
Optimizing the filters to improve the filtered images for computer vision tasks.
jax numpy computer-vision separable-filters
Objectives & Highlights

• How to use JAX (“hyped” new Python ML / autodifferentiation library). • Basic application that is follow-up to my previous blog post on using SVD for low-rank approximations and separable image filters – we will look at “optimizing” the filters to improve the filtered images.

Don't forget to add the tag @bartwronski in your comments.

This project's author does not have a MWML account yet. If you are @bartwronski, then sign up to gain ownership of this project and edit this page.
Share this project
Similar projects
Flax: Google’s Open Source Approach To Flexibility In ML
A gentle introduction to Flax: a neural network library for JAX that is designed for flexibility.
Getting started with JAX (MLPs, CNNs & RNNs)
Learn the building blocks of JAX and use them to build some standard Deep Learning architectures (MLP, CNN, RNN, etc.).
From PyTorch to JAX
Towards neural net frameworks that purify stateful code.
You don't know JAX
In this tutorial, we'll cover each of these transformations in turn by demonstrating their use on one of the core problems of AGI: learning the Exclusive ...