Compressed Sparse Format

SciSharp / NumSharp

High Performance Computation for N-D Tensors in .NET, similar API to NumPy.

https://github.com/SciSharp

Apache License 2.0

1.36k stars 189 forks source link

Compressed Sparse Format #190

Open Oceania2018 opened 5 years ago

Oceania2018 commented 5 years ago

We should implement the compressed ndarray, it is used widely in scikit-learn. The csr_matrix is included in SciPy not NumPy, but I think we should move it into NumSharp. @dotChris90 What do you think of it?

Compressed Sparse Column Format (CSC) Compressed Sparse Row Format (CSR)

https://www.scipy-lectures.org/advanced/scipy_sparse/storage_schemes.html

Deep-Blue-2013 commented 5 years ago

CSC format saved memory a lot. Looking forward to see this feature.

sebhofer commented 5 years ago

Would be great to have an sparse matrix implementation with good performance! Math.Net has some support for sparse matrices, but performance is not great. Also, would be useful to have support for different storage formats (as you already hinted at above).

As an additional reference: Here is an implementation improving on the performance of Math.Net.

Oceania2018 commented 5 years ago

@sebhofer Thanks for the info, it's important to us.