Home       Trees       Indices       Help   
CSB
Package csb :: Package statistics :: Package pdf :: Class AbstractDensity
[frames] | no frames]

Class AbstractDensity

source code

object --+
         |
        AbstractDensity
Known Subclasses:

Defines the interface and common operations for all probability density functions. This is a generic class which can operate on parameters of any type (e.g. simple floats or custom parameter objects).

Subclasses must complete the implementation by implementing the AbstractDensity.log_prob method. Subclasses could also consider--but are not obliged to--override the AbstractDensity.random method. If any of the density parameters need validation, subclasses are expected to override the AbstractDensity._validate method and raise ParameterValueError on validation failure. Note that implementing parameter validation in property setters has almost no effect and is discouraged.

Nested Classes
  __metaclass__
Metaclass for defining Abstract Base Classes (ABCs).
Instance Methods
 
__call__(self, x) source code
 
__getitem__(self, param) source code
 
__init__(self)
x.__init__(...) initializes x; see help(type(x)) for signature		 source code
 
__setitem__(self, param, value) source code
 
__str__(self)
str(x)		 source code
 
estimate(self, data)
Estimate and load the parameters of the distribution from sample data using the current AbstractEstimator strategy.		 source code
array
evaluate(self, x)
Evaluate the probability of observing values x.		 source code
 
get_params(self) source code
array
log_prob(self, x)
Evaluate the logarithm of the probability of observing values x.		 source code
 
random(self, size=None)
Generate random samples from the probability distribution.		 source code
 
set_params(self, *values, **named_params) source code

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

Class Variables
  __abstractmethods__ = frozenset(['log_prob'])
Properties
  estimator
  parameters
Get a list of all distribution parameter names.

Inherited from object: __class__

Method Details

__init__(self)
(Constructor)

source code 

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

Overrides: object.__init__
(inherited documentation)

__str__(self)
(Informal representation operator)

source code 

str(x)

Overrides: object.__str__
(inherited documentation)

estimate(self, data)

source code 

Estimate and load the parameters of the distribution from sample data using the current AbstractEstimator strategy.

Parameters:
  • data (array) - sample values
Raises:
  • NotImplementedError - when no estimator is available for this distribution
  • IncompatibleEstimatorError - when the current estimator is not compatible with this pdf

evaluate(self, x)

source code 

Evaluate the probability of observing values x.

Parameters:
  • x (array) - values
Returns: array

log_prob(self, x)

source code 

Evaluate the logarithm of the probability of observing values x.

Parameters:
  • x (array) - values
Returns: array

random(self, size=None)

source code 

Generate random samples from the probability distribution.

Parameters:
  • size (int) - number of values to sample

Property Details

estimator

Get Method:
unreachable.estimator(self)
Set Method:
unreachable.estimator(self, strategy)

parameters

Get a list of all distribution parameter names.

Get Method:
unreachable.parameters(self) - Get a list of all distribution parameter names.

   Home       Trees       Indices       Help   
CSB
Generated by Epydoc 3.0.1 on Mon Apr 16 11:31:44 2018 http://epydoc.sourceforge.net