Python wrapper?

[MaxBenChrist]

Hi Ben :),

there was a question regarding synthetic time series generators in python, see

MaxBenChrist/awesome_time_series_in_python#7

Can you provide a python wrapper for this TimeSeriesGeneration package? Would love to include it in my list.

[benfulcher]

I'm not sure how to do the python wrapping of Matlab; Santi Villalba helped do this for hctsa a few years ago. But I'd be open to someone working this up! It's all very simple stuff that would be straightforward to reimplement

[MaxBenChrist]

If you are interested, I think we can try to convert the matlab files directly into python code as you are mostly using basic arithmetic functions.

Here is a list of options to convert matlab to python code
https://stackoverflow.com/questions/9845292/a-tool-to-convert-matlab-code-to-python

especially this one looks promising

https://github.com/victorlei/smop

I will have a look at it later

[MaxBenChrist]

I started converting the matlab code to python, see

https://github.com/MaxBenChrist/TimeSeriesGeneration

[benfulcher]

Nice! Look forward to seeing it develop 👍

[MaxBenChrist]

Nice! Look forward to seeing it develop 👍

Shall we make a python package out of it? We can add an folder with the matlab source files inside that package. Or we open a new repo

[benfulcher]

Up to you! Happy for you to do what you like with it, if you properly credit the original source. I should add a license...

[MaxBenChrist]

I would be in favor of having another clean repository, with both repositorys referencing each other.

I think I will have to rewrite quite a lot of the calculations. At the moment, smop is replacing the vectors by custom ndarrays (such that the indexing starts at 1 and not at 0). This dependency on smop is unecessary and I want to get rid of it.

[benfulcher]

Go for it!

…

On 14 Sep 2017, at 00:22, Maximilian Christ ***@***.***> wrote: I would be in favor of having another clean repository, with both repositorys referencing each other. I think I will have to rewrite quite a lot of the calculations. At the moment, smop is replacing the vectors by custom ndarrays (such that the indexing starts at 1 and not at 0). This dependency on smop is unecessary and I want to get rid of it. ― You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.