The implementation of Batch OMP. I have written omp algorithm with Numpy, including Origin OMP、Cholesky OMP and Batch OMP in /code/omp.py, but it's a bit slow. You can see the excellent performance of Cython version in /demo.py, however, it is still very slow compared to ompbox10 (matlab). You can clone this repository with git clone --depth 1 [email protected]:muggledy/batch-omp.git.
First, add Eigen library to /third-party by executing git submodule init and git submodule update.
Second, you need to prepare Cython for compiling /code/batch_omp.pyx to .cpp file and gcc(C++), so the .cpp file can be compiled to .pyd(on Windows), which can be directly imported in a Python session. To accomplish the installation, cd into sub-directory /code and run python setup.py build_ext --inplace, you will get a .pyd file.
[1] M. Elad, R. Rubinstein, and M. Zibulevsky, "Efficient Implementation of the K-SVD Algorithm using Batch Orthogonal Matching Pursuit", Technical Report - CS, Technion, April 2008.