GGPSampler

class glimix_core.random.GGPSampler(lik, mean, cov)

Sample from a Generalised Gaussian Process.

Outcome modelled via

\[\mathbf y \sim \int \prod_i \mathrm{ExpFam}(y_i ~|~ g_i(z_i)) \mathcal N(\mathbf z ~|~ \mathbf m; \mathrm K) \mathrm d\mathbf z.\]

Parameters

• link (str) – Likelihood name.

• mean (function) – Mean function.

• cov (function) – Covariance function.

Example

>>> from numpy.random import RandomState
>>>
>>> from glimix_core.example import offset_mean
>>> from glimix_core.example import linear_eye_cov
>>> from glimix_core.random import GGPSampler
>>> from glimix_core.lik import DeltaProdLik
>>>
>>> random = RandomState(1)
>>>
>>> mean = offset_mean()
>>> cov = linear_eye_cov()
>>>
>>> lik = DeltaProdLik()
>>>
>>> y = GGPSampler(lik, mean, cov).sample(random)
>>> print(y[:5])  
[-2.42181498  0.50720447 -1.01053967  0.736624    1.64019063]

__init__(lik, mean, cov)

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__(lik, mean, cov)	Initialize self.
sample([random_state])	Sample from the specified distribution.