benfulcher/hctsa

Init only with notLengthDependent features

Drodriguez-9 opened this issue · 1 comments

Hi Ben, first of all thanks for this amazing tool.
Could you please explain me how to compute only the notLengthDependent features? In your tutorial it's only explained how to do so with the catch22 set.
Thank you.

Thanks @Drodriguez-9,
Easiest is to do a TS_Init with all features as HCTSA.mat, and then subset:

[IDs_lenDep,IDs_notLenDep] = TS_GetIDs('lengthDependent','HCTSA.mat','ops','Keywords');
TS_Subset('HCTSA.mat',[],IDs_notLenDep,true,'HCTSA_lenDepFiltered.mat')

Then you have the full set initialized as HCTSA.mat and the length-dependent-removed as HCTSA_lenDepFiltered.mat. You can then e.g., just work from the lenDepFiltered version for the subsequent calculation.

I typically compute everything and then subset later for flexibility. If you go this route, then the TS_Subset is the command you use, just on the already-computed version of HCTSA.mat, or there is some capability for filtering for analysis without needing to create lots of filtered .mat files (e.g., in the reducedFeatureSet field of the classification parameter structure, cf. GiveMeDefaultClassificationParams).

Hope this helps and good luck with your research.
Ben