refactor MLHMM / baum_welch

[franknoe]

MLHMM / baum_welch are redundant and should be refactored:

• MLHMM._transitionMatrixMLE is probable not used and definitely not needed
• MLHMM._generateInitialModel is not just a call and go to the init method
• MLHMM only delegates the work to baum_welch, so perhaps the content of baum_welch can be migrated into MLHMM

[jchodera]

We should definitely eliminate redundancy and dead code.

Regarding refactoring: we should try to make the code as close to the structure documented in the paper as possible. Optimally this would be to move functions into MLHMM, and have readable Python fallback implementations of anything that calls out to C code.

[franknoe]

I agree. I'll take care of it.

Am 17/03/15 um 14:36 schrieb John Chodera:

We should definitely eliminate redundancy and dead code.

Regarding refactoring: we should try to make the code as close to the structure documented in the paper as possible. Optimally this would be to move functions into MLHMM, and have readable Python fallback implementations of anything that calls out to C code.

— Reply to this email directly or view it on GitHub https://github.com/choderalab/bhmm/issues/12#issuecomment-82355838.

---

Prof. Dr. Frank Noe Head of Computational Molecular Biology group Freie Universitaet Berlin

Phone: (+49) (0)30 838 75354 Web: research.franknoe.de

Mail: Arnimallee 6, 14195 Berlin, Germany

[franknoe]

done