Home       Trees       Indices       Help   
CSB v1.2.5.d7fd331
Package csb :: Package statistics :: Package samplers :: Package mc :: Module singlechain :: Class HMCSampler
[frames] | no frames]

Class HMCSampler

source code

     object --+            
              |            
AbstractSampler --+        
                  |        
         AbstractMC --+    
                      |    
  AbstractSingleChainMC --+
                          |
                         HMCSampler

Hamilton Monte Carlo (HMC, also called Hybrid Monte Carlo by the inventors, Duane, Kennedy, Pendleton, Duncan 1987).

Nested Classes

Inherited from AbstractSampler: __metaclass__

Instance Methods
 
__init__(self, pdf, state, gradient, timestep, nsteps, mass_matrix=None, integrator=<class 'csb.numeric.integrators.FastLeapFrog'>, temperature=1.0)
x.__init__(...) initializes x; see help(type(x)) for signature		 source code

Inherited from AbstractSingleChainMC: sample

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

Class Variables
  __abstractmethods__ = frozenset([])
Properties
  mass_matrix
  nsteps
  timestep

Inherited from AbstractSingleChainMC: acceptance_rate, energy, last_move_accepted, temperature

Inherited from AbstractMC: state

Inherited from object: __class__

Method Details

__init__(self, pdf, state, gradient, timestep, nsteps, mass_matrix=None, integrator=<class 'csb.numeric.integrators.FastLeapFrog'>, temperature=1.0)
(Constructor)

source code 

x.__init__(...) initializes x; see help(type(x)) for signature

Parameters:
  • pdf (csb.statistics.pdf.AbstractDensity) - Probability density function to be sampled from
  • state (State) - Inital state
  • gradient (AbstractGradient) - Gradient of the negative log-probability
  • timestep (float) - Timestep used for integration
  • nsteps (int) - Number of integration steps to be performed in each iteration
  • mass_matrix (n-dimensional InvertibleMatrix with n being the dimension of the configuration space, that is, the dimension of the position / momentum vectors) - Mass matrix
  • integrator (AbstractIntegrator) - Subclass of AbstractIntegrator to be used for integrating Hamiltionian equations of motion
  • temperature (float) - Pseudo-temperature of the Boltzmann ensemble p(x) = 1/N * exp(-1/T * E(x)) with the pseudo-energy defined as E(x) = -log(p(x)) where p(x) is the PDF under consideration
Overrides: object.__init__
(inherited documentation)

Property Details

mass_matrix

Get Method:
unreachable.mass_matrix(self)
Set Method:
unreachable.mass_matrix(self, value)

nsteps

Get Method:
unreachable.nsteps(self)
Set Method:
unreachable.nsteps(self, value)

timestep

Get Method:
unreachable.timestep(self)
Set Method:
unreachable.timestep(self, value)

   Home       Trees       Indices       Help   
CSB v1.2.5.d7fd331
Generated by Epydoc 3.0.1 on Tue Jul 4 20:19:06 2017 http://epydoc.sourceforge.net