ASCL.net

Astrophysics Source Code Library

Making codes discoverable since 1999

ASCL Code Record

[ascl:2005.010] NNKCDE: Nearest Neighbor Kernel Conditional Density Estimation

NNKCDE is a simple and easily interpretable Conditional Density Estimation (CDE) method. It computes a kernel density estimate of y using the k nearest neighbors of the evaluation point x. The model has only two tuning parameters: the number of nearest neighbors k and the bandwidth h of the smoothing kernel in y-space. Both tuning parameters are chosen in a principled way by minimizing the CDE loss on validation data.

Code site:
https://github.com/lee-group-cmu/NNKCDE
Used in:
https://ui.adsabs.harvard.edu/abs/2020A%26C....3000362D
Described in:
https://ui.adsabs.harvard.edu/abs/2018arXiv180505480I
Bibcode:
2020ascl.soft05010I

Views: 2525

ascl:2005.010
Add this shield to your page
Copy the above HTML to add this shield to your code's website.