Package trunk :: Package BIP :: Package Bayes :: Package Samplers :: Module MCMC :: Class _Sampler

Class _Sampler

object --+
         |
        _Sampler

Base classe for all samplers Holds common logic and

Instance Methods

[hide private]

__init__(self, parpriors=[], parnames=[])
x.__init__(...) initializes x; see x.__class__.__doc__ for signature

source code

best_prop_index(self)
Returns the index of the best fitting proposal, i.e., the one which with max Likelihood

source code

DIC(self)
Calculates the deviance information criterion

source code

dimensions(self)

source code

po(self)
Pool of processes for parallel execution of tasks Remember to call self.term_pool() when done.

source code

shut_down(self, reason='')
Finalizes the sampler, nicely closing the resources allocated

source code

term_pool(self)

source code

gr_R(self, end, start)

source code

gr_convergence(self, relevantHistoryEnd, relevantHistoryStart)
Gelman-Rubin Convergence

source code

_accept(self, last_lik, lik)
Decides whether to accept a proposal

source code

setup_xmlrpc_plotserver(self)
Sets up the server for real-time chain watch

source code

shutdown_xmlrpc_plotserver(self)

source code

_every_plot(self)
plotting function for generating a plot at every step

source code

_watch_chain(self, j)

source code

_tune_likvar(self, ar)

source code

check_constraints(self, theta)
Check if given theta vector complies with all constraints

source code

_propose(self, step, po=None)
Generates proposals. returns two lists

source code

Inherited from object: __delattr__, __format__, __getattribute__, __hash__, __new__, __reduce__, __reduce_ex__, __repr__, __setattr__, __sizeof__, __str__, __subclasshook__

Class Variables

[hide private]

_po = None

_dimensions = None

trace_acceptance = False

trace_convergence = False

seqhist = None

liklist = []

e = 1e-20

_j = -1

_R = np.inf

Properties

[hide private]

Inherited from object: __class__

Method Details

[hide private]

init(self, parpriors=[], parnames=[])
(Constructor)

source code

x.__init__(...) initializes x; see x.__class__.__doc__ for signature

Overrides: object.__init__: (inherited documentation)

best_prop_index(self)

source code

Returns the index of the best fitting proposal, i.e., the one which with max Likelihood

Decorators:

@property

DIC(self)

source code

Calculates the deviance information criterion

Decorators:

@property

dimensions(self)

source code

Decorators:

@property

po(self)

source code

Pool of processes for parallel execution of tasks Remember to call self.term_pool() when done.

Decorators:

@property

shut_down(self, reason='')

source code

Finalizes the sampler, nicely closing the resources allocated

Parameters:

reason - : comment stating why the sampling is being shutdown.

_accept(self, last_lik, lik)

source code

Decides whether to accept a proposal

Decorators:

@np.vectorize

check_constraints(self, theta)

source code

Check if given theta vector complies with all constraints

Parameters:

theta - : parameter vector

Returns:

True if theta passes all constraints, False otherwise

_propose(self, step, po=None)

source code

Generates proposals. returns two lists

Parameters:

step - : Position in the markov chain history.
po - : Process pool for parallel proposal generation

Returns:

theta: List of proposed self.dimensional points in parameter space
prop: List of self.nchains proposed phis.

Class _Sampler

__init__(self, parpriors=[], parnames=[]) (Constructor)

best_prop_index(self)

DIC(self)

dimensions(self)

po(self)

shut_down(self, reason='')

_accept(self, last_lik, lik)

check_constraints(self, theta)

_propose(self, step, po=None)

init(self, parpriors=[], parnames=[])
(Constructor)