Local Entropy over Network Graph

These are methods to compute information-theoretic entropy over a network graph. Inputs are a network graph (NN object name) and a feature name. Optionally, users can provide a condition-feature name, which will then compute conditional entropy (X | Y)

Usage

localEntropy.Region(object, nn, feature, condition = NULL, method = "emp")

localEntropy(object, nn, feature, condition = NULL, method = "emp", ...)

Arguments

object

A SpatialMap or Region object

nn

The name of a Nearest Neighbors object

feature

The name of a feature, e.g. from cellMetadata

condition

The name of another feature, e.g. from cellMetadata. If provided, this feature will be used to compute conditional entropy.

method

Argument passed to infotheo::condentropy

...

Arguments passed to .smapply

Value

A SpatialMap or Region object with new data added.