ConditionResult#

class tinygp.gp.ConditionResult(log_probability: JAXArray, gp: GaussianProcess)[source]#

Bases: NamedTuple

The result of conditioning a GaussianProcess on data

This has two entries, log_probability and gp, that are described below.

count(value, /)#

Return number of occurrences of value.

gp: GaussianProcess#

A GaussianProcess describing the conditional distribution

This will have a mean and covariance conditioned on the observed data, but it is otherwise a fully functional GP that can sample from or condition further (although that’s probably not going to be very efficient).

index(value, start=0, stop=9223372036854775807, /)#

Return first index of value.

Raises ValueError if the value is not present.

log_probability: JAXArray#

The log probability of the conditioned model

In other words, this is the marginal likelihood for the kernel parameters, given the observed data, or the multivariate normal log probability evaluated at the given data.