Computing KDE can still be slow for some numax/dnu

grd349 / PBjam

A repo for our peak baggin code and tips on jam

MIT License

17 stars 6 forks source link

Closed nielsenmb closed 4 years ago

nielsenmb commented 5 years ago

For some values of numax/dnu the prior sample is still very dense.

Suggest selecting a subsample of stars if number within the range used for the KDE exceeds X.

nielsenmb commented 4 years ago

Fixed with PR #198

Downsamples to KDEsize (default = 100) if the interval in numax spans more than 100.