# SPDX-License-Identifier: MPL-2.0
# Copyright (C) 2020- The University of Tokyo
#
# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at https://mozilla.org/MPL/2.0/.
import numpy as np
[docs]
class Cov:
"""
Covariance
Attributes
==========
params: float
half of log of variance
nparams: int
number of parameters
sigma2: float
variance
prec: float
inv. of variance
"""
def __init__(self, params=None):
self.params = params
if self.params is None:
self.params = np.log(1)
self.nparams = 1
self.sigma2, self.prec = self._trans_params(params)
[docs]
def get_cov(self, N, params=None):
"""
compute the covariance of prior
Parameters
==========
N: int
dimension
params:
half of log of variance
(default: self.params)
Returns
=======
numpy.ndarray
NxN covariance matrix
"""
if params is None:
params = self.params
sigma2, prec = self._trans_params(params)
return np.identity(N) * sigma2
[docs]
def get_prec(self, N, params=None):
"""
compute the precision of prior
Parameters
==========
N: int
dimension
params:
half of log of variance
(default: self.params)
Returns
=======
numpy.ndarray
inverse of covariance matrix
"""
if params is None:
params = self.params
sigma2, prec = self._trans_params(params)
return np.identity(N) * prec
[docs]
def set_params(self, params):
"""
set the parameter
Parameters
==========
params: float
half of log of variance
"""
self.params = params
self.sigma2, self.prec = self._trans_params(params)
def _trans_params(self, params=None):
"""
transform the parameter into variance and precision
Parameters
==========
params: float
half of log of variance
(default: self.params)
Returns
=======
sigma2: float
variance
prec: float
precision (inv. of variance)
"""
if params is None:
params = np.copy(self.params)
sigma2 = np.exp(2 * params)
prec = 1 / sigma2
return sigma2, prec