def __init__(self, distribution, kernel, Z, threshold, spread):
if not isinstance(distribution, Distribution):
raise TypeError("Target must be a Distribution object")
if not isinstance(kernel, Kernel):
raise TypeError("Kernel must be a Kernel object")
if not type(Z) is numpy.ndarray:
raise TypeError("History must be a numpy array")
if not len(Z.shape) == 2:
raise ValueError("History must be a 2D numpy array")
if not Z.shape[1] == distribution.dimension:
raise ValueError(
"History dimension does not match target dimension")
if not Z.shape[0] > 0:
raise ValueError("History must contain at least one point")
if not type(threshold) is float:
raise TypeError("Threshold must be a float")
if not type(spread) is float:
raise TypeError("Spread must be a float")
if not (spread > 0. and spread < 1.):
raise ValueError("Spread must be a probability")
MCMCSampler.__init__(self, distribution)
self.kernel = kernel
self.Z = Z
self.threshold = threshold
self.spread = spread
def __init__(self, distribution, kernel, Z, threshold, spread):
if not isinstance(distribution, Distribution):
raise TypeError("Target must be a Distribution object")
if not isinstance(kernel, Kernel):
raise TypeError("Kernel must be a Kernel object")
if not type(Z) is numpy.ndarray:
raise TypeError("History must be a numpy array")
if not len(Z.shape) == 2:
raise ValueError("History must be a 2D numpy array")
if not Z.shape[1] == distribution.dimension:
raise ValueError("History dimension does not match target dimension")
if not Z.shape[0] > 0:
raise ValueError("History must contain at least one point")
if not type(threshold) is float:
raise TypeError("Threshold must be a float")
if not type(spread) is float:
raise TypeError("Spread must be a float")
if not (spread > 0. and spread < 1.):
raise ValueError("Spread must be a probability")
MCMCSampler.__init__(self, distribution)
self.kernel = kernel
self.Z = Z
self.threshold = threshold
self.spread = spread
def __init__(self, full_conditionals):
if not isinstance(full_conditionals, FullConditionals):
raise TypeError("Gibbs require full conditional distribution instance")
MCMCSampler.__init__(self, full_conditionals)
# pdf is constant, and therefore symmetric
self.is_symmetric = True
def __init__(self, distribution, kernel, Z, nu2=0.1, gamma=None):
MCMCSampler.__init__(self, distribution)
self.kernel = kernel
self.nu2 = nu2
if gamma is not None:
self.gamma = gamma
else:
self.gamma=0.2
self.Z = Z
def __init__(self, distribution, scale=None, cov=None):
MCMCSampler.__init__(self, distribution)
if scale is None:
self.scale = (2.38 ** 2) / distribution.dimension
else:
self.scale = scale
if cov is None:
self.cov = eye(distribution.dimension)
else:
self.cov = cov
def __init__(self, distribution, scale=None, cov=None):
MCMCSampler.__init__(self, distribution)
if scale is None:
self.scale = (2.38**2) / distribution.dimension
else:
self.scale = scale
if cov is None:
self.cov = eye(distribution.dimension)
else:
self.cov = cov
def __init__(self, distribution, spread, flip_at_least_one=True):
if not isinstance(distribution, Distribution):
raise TypeError("Target must be a Distribution object")
if not type(spread) is float:
raise TypeError("Spread must be a float")
if not (spread > 0. and spread < 1.):
raise ValueError("Spread must be a probability")
MCMCSampler.__init__(self, distribution)
self.spread = spread
self.flip_at_least_one = flip_at_least_one
def __init__(self, distribution, spread, flip_at_least_one=True):
if not isinstance(distribution, Distribution):
raise TypeError("Target must be a Distribution object")
if not type(spread) is float:
raise TypeError("Spread must be a float")
if not (spread > 0. and spread < 1.):
raise ValueError("Spread must be a probability")
MCMCSampler.__init__(self, distribution)
self.spread = spread
self.flip_at_least_one = flip_at_least_one
def __str__(self):
s=self.__class__.__name__+ "=["
s += "kernel="+ str(self.kernel)
s += ", nu2="+ str(self.nu2)
s += ", gamma="+ str(self.gamma)
s += ", " + MCMCSampler.__str__(self)
s += "]"
return s
def __str__(self):
s = self.__class__.__name__ + "=["
s += "kernel=" + str(self.kernel)
s += ", threshold=" + str(self.threshold)
s += ", spread=" + str(self.spread)
s += ", " + MCMCSampler.__str__(self)
s += "]"
return s
def __str__(self):
s = self.__class__.__name__ + "=["
s += "kernel=" + str(self.kernel)
s += ", threshold=" + str(self.threshold)
s += ", spread=" + str(self.spread)
s += ", " + MCMCSampler.__str__(self)
s += "]"
return s
def __str__(self):
s=self.__class__.__name__+ "=["
s += "globalscale="+ str(self.globalscale)
s += ", sample_discard="+ str(self.sample_discard)
s += ", sample_lag="+ str(self.sample_lag)
s += ", accstar="+ str(self.accstar)
s += ", " + MCMCSampler.__str__(self)
s += "]"
return s
def __init__(self, distribution, \
mean_est=None, cov_est=None, \
sample_discard=500, sample_lag=20, accstar=0.234):
MCMCSampler.__init__(self, distribution)
self.globalscale = (2.38 ** 2) / distribution.dimension
if mean_est is None:
mean_est=2*ones(distribution.dimension)
if cov_est is None:
cov_est=0.05 * eye(distribution.dimension)
assert (len(mean_est) == distribution.dimension)
assert (len(cov_est) == distribution.dimension)
self.mean_est = mean_est
self.cov_est = cov_est
self.sample_discard = sample_discard
self.sample_lag = sample_lag
self.accstar = accstar
def __str__(self):
s = self.__class__.__name__ + "=["
s += "spread=" + str(self.spread)
s += ", " + MCMCSampler.__str__(self)
s += "]"
return s
def __str__(self):
s = self.__class__.__name__ + "=["
s += "scale=" + str(self.scale)
s += ", " + MCMCSampler.__str__(self)
s += "]"
return s
def __str__(self):
s = self.__class__.__name__ + "=["
s += MCMCSampler.__str__(self)
s += "]"
return s