rrrlw / TDAstats

R pipeline for computing persistent homology in topological data analysis. See https://doi.org/10.21105/joss.00860 for more details.
https://rrrlw.github.io/TDAstats
GNU General Public License v3.0
37 stars 9 forks source link

real-world illustration #1

Closed corybrunson closed 5 years ago

corybrunson commented 6 years ago

The toy examples provided in with the package are helpful and wholly appropriate. The package would also benefit, i think, from an illustration using real-world data (as suggested but not required by the JOSS review checklist). Is there a dataset you've used for this purpose that could be included in the package and demonstrated either in the functional documentation or in a separate vignette? This isn't a sticking point for my review, but i do think the package would achieve its aims more effectively with such a case study.

Part of this JOSS review.

rrrlw commented 6 years ago

Yes, this is a great suggestion. Unfortunately, this package was created to cement the methodology of future research, rather than to solve a problem noticed in a prior project. As such, we do not yet have real-world data to which we have applied the functions in this package. I will leave the issue open, and when we do apply this method to real-world data with successful results, I will add the dataset and corresponding vignette/case study to the package.

kaitai commented 5 years ago

Hi -- I'm working on using this with real data and could contribute an illustration if I knew how to input a distance matrix to the TDAstats calculate_homology command. I know that Ripser is happy to work with distance matrices, but I don't see how to use a distance matrix with TDAstats, and so much real-world data requires a distance matrix (financial data, health data, etc.).

Let me know how to input a distance matrix and I'll give you a nice real-world application!

ShotaOchi commented 5 years ago

Does the following code work?

calculate_homology(mat, format = "distmat")
kaitai commented 5 years ago

Thanks for the reply.

calculate_homology(mat, format = "distmat") does work as long as mat is

It does not work if mat is upper-triangular or the output of R's dist function, for what it's worth -- but dist does have a funky output format.

I'll try to submit a real-world example by next Tuesday.

rrrlw commented 5 years ago

Great, looking forward to seeing the real-world example @kaitai. Happy that TDAstats was helpful. And thank you, @ShotaOchi, appreciate the helpful answer.

kaitai commented 5 years ago

I should have been wiser than to promise anything during the first week of the semester. I put an Rmd file on my github page and am happy to contribute it. https://github.com/kaitai/Example-with-TDAstats

rrrlw commented 5 years ago

No worries, all contributions are appreciated. The link you included does not seem to be working for me; does it point to a private repo?

kaitai commented 5 years ago

No, just a typo :) Try now!

rrrlw commented 5 years ago

Looks wonderful, thank you! I will link it to the TDAstats main page soon.

rrrlw commented 5 years ago

The link has been added to the Applications section of the README. Thank you, @kaitai!