search

dirmeier / diffusr

Network diffusion algorithms in R.
GNU General Public License v3.0
21 stars 6 forks source link

Support for igraph or sparse matrix? #13

Open fl-yu opened 3 years ago

fl-yu commented 3 years ago

Thank you for the development of such a valuable tool. I am wondering if the 'graph' used in the random walk could be supported by using an object of igraph package or spare matrix? I think it would be a great benefit for a large background network.

dirmeier commented 3 years ago

Ciao @fl-yu , that should be easily doable. I don't have much time on my hands rn, so do you maybe want to give it a try and submit a PR? Otherwise I can implement this once I have so more time. Cheers, S

randef1ned commented 7 months ago

I have forked this R package and added capability to efficiently handle sparse matrix. I have also reorganized the code; so reduced the running time by approximately 10%. I also enabled oneAPI MKL by default, so users may gain another 5% boost. I modified and tested my code on both Fedora 39 and Windows 10, and it worked smoothly well.

However, I have not done everything well due to my limited time. There are still some computational bugs in Ubuntu. The docs are not prepared well. The coverage rate of unit test in newly-added code is low.

If you are still interested in this package, I suggest you try my fork at https://github.com/randef1ned/diffusr @fl-yu.

Cheers!