MParzenANLL: Average 'MParzenTest()' over all test points

In rudazhang/plmr: Probabilistic Learning on Manifolds in R

Description

Useful for tuning hyperparameters: d, k, sigma.

Usage

MParzenANLL(Y, X, d, k, sigma)

Arguments

Y	Matrix of test data, n columns.
X	Matrix of training data, n columns.
d	Dimension of manifold.
k	Number of nearest neighbors to use in local covariance analysis.
sigma	Maximum noise level added to the normal space of each kernel, i.e. marginal standard deviation of an isotropic Gaussian. Should be greater than zero.

Value

Average negative log likelihood of the test data given the MParzen model on the training data and the hyperparameters.

rudazhang/plmr documentation built on Aug. 30, 2021, 5:43 p.m.