Trend

A trend in the coordinates, from detrend: a polynomial, or a smooth kernel average of the samples.

bandwidth property

bandwidth

Kernel bandwidth in use; None for a polynomial.

bandwidths property

bandwidths

Candidate bandwidths; None for a polynomial.

categories property

categories

Category names of a categorical trend, else None.

coefficients property

coefficients

Polynomial coefficients (1, x, y, z, x², y², z², xy, xz, yz up to the degree); None for a kernel trend.

degree property

degree

Polynomial degree; None for a kernel trend.

scores property

scores

Leave-one-out error of each candidate bandwidth; None for a polynomial.

from_json staticmethod

from_json(text)

Reads to_json output; raises InvalidInput on another class's JSON or a newer format.

predict method descriptor

predict(coords)

The trend at coords (array, PointSet or BlockModel cells): an array, or for categories a Table of proportions, one column each. A kernel trend is null (NaN) beyond four bandwidths of every sample; fill it (e.g. with the data mean) where a simulation needs a trend everywhere.

to_json method descriptor

to_json()

JSON of the parameters and, once fitted, the fitted state.