Package transsys :: Module optim :: Class GradientOptimisationRecord
[hide private]
[frames] | no frames]

Class GradientOptimisationRecord

source code

        object --+    
                 |    
OptimisationRecord --+
                     |
                    GradientOptimisationRecord

Record containing data from a step of gradient optimisation.

Instance Methods [hide private]
 
__init__(self, obj, stepsize, numEvaluations, gradient)
Constructor.
source code
 
__str__(self)
str(x)
source code

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

Class Variables [hide private]
  table_header = 'obj stepsize numEvaluations gradient_entropy g...
string providing a suitable header for a table containing instance representations generated by the __str__ method
Instance Variables [hide private]
  gradient_entropy
entropy of the gradient vector
  gradient_max
maximum of the gradient vector
  numEvaluations
number of evaluations (cumulative)
  obj
current objective function value
Properties [hide private]

Inherited from object: __class__

Method Details [hide private]

__init__(self, obj, stepsize, numEvaluations, gradient)
(Constructor)

source code 

Constructor.

Parameters:
  • obj - current objective function value
  • numEvaluations - cumulative number of evaluations
  • gradient - gradient vector as a list, normalised
Overrides: object.__init__

__str__(self)
(Informal representation operator)

source code 

str(x)

Overrides: object.__str__
(inherited documentation)

Class Variable Details [hide private]

table_header

string providing a suitable header for a table containing instance representations generated by the __str__ method
Value:
'obj stepsize numEvaluations gradient_entropy gradient_max'