csalgs: Compressed sensing algorithms in Python

This is a small collection of compressed sensing/low rank matrix recovery algorithms in Python. It's neither complete nor very elaborate -- it's mainly just for learning exisiting algorithms or for testing purposes. Use at your own risk :)

Content

• csalg.tt: Low-rank tensor recovery for the tensor train format
  • iht.py: Iterative hard thresholding (projected gradient descent)
  • altmin.py: Alternating Least Squares
  • _altmin_gpu.py: A CUDA implementation of alternating least squares
• csalgs.lowrank: Low-rank matrix recovery
  • gradient.py: Gradient based schemes such as Iterative hard thresholding (projected gradient descent) and conjugated gradient descent
  • convex.py: Convex optimization methods (nuclear norm minimization and constrained l2 minimization)
  • altmin.py: Alternating Least Squares
• csalg.cs: Compressed Sensing (Recovery of sparse vectors)
  • iht.py: Iterative hard thresholding (projected gradient descent)

LICENSE

Distributed under the terms of the GPLv3 license (see LICENSE).