Open mfalkiewicz opened 6 years ago
@mfalkiewicz - should #13 still be a PR?
@satra: no, I merged them both together and deleted #13.
also tests are not passing - since the markov computation has changed.
i would recommend creating a second markov
option with your changes, unless you think the current implementation is an error. this will help keep the api consistent. and you can create a new set of embedding tests with the alternate calculation.
Refactoring the kernel normalization and transition matrix computation as independent function will allow for more code re-use.