giotto-ai / giotto-tda

A high-performance topological machine learning toolbox in Python
https://giotto-ai.github.io/gtda-docs
Other
845 stars 173 forks source link

Compute mask sizes more carefully in DensityFiltration #512

Open ulupo opened 3 years ago

ulupo commented 3 years ago

Currently, the logic in

https://github.com/giotto-ai/giotto-tda/blob/d08900294467417a1ddc29a19d8c353583c567f1/gtda/images/filtrations.py#L1166-L1169

assumes a fair amount form the metric used. In particular, it assumes that the metric has a counterpart in 1D and that a cubical "container" for the appropriate mask can be found purely by considering this 1D counterpart.