MAPLE
MAPLE copied to clipboard
Potential for multi-threading?
I assume parts of this algorithm could be parallelized.
eg. using multoprocessing library
What are your plans to do this?
I am currently working on improving the code/algorithm (there is still substantial margin to reduce time and memory demands). At the same time I am in contact with other groups to have the idea implemented in C++ as part of other distributions (UShER and IQ-TREE), and possibly also in Java/BEAST. my suspicion is that it makes sense to invest in parallelization once in C++/Java rather than with this initial python version. However, if you think this would be easy to do, let me know (I have not used parallelization libraries in python before)!