Home       Trees       Indices       Help   
Package crino :: Module criterion :: Class Criterion
[hide private]
[frames] | no frames]

Class Criterion

source code

Known Subclasses:
The Criterion class handles the loss computation between ŷ (the outputs vector of a Module) and y (the targets vector). This loss has to be differentiable, in order to perform a gradient descent.

Attention: This is an abstract class, it must be derived to be used.

Instance Methods [hide private]
 
__init__(self, outputs, targets)
Constructs a new Criterion object.		 source code
 
prepare(self)
Computes the symbolic expression of the loss.		 source code
Instance Variables [hide private]
TensorVariable outputs
The symbolic outputs vector of the network, denoted ŷ.
TensorVariable targets
The symbolic targets vector, denoted y, that will be estimated by the outputs.
TensorVariable expression
The symbolic expression that expresses the loss between the outputs and the targets vectors.
Method Details [hide private]

__init__(self, outputs, targets)
(Constructor)

source code 
Constructs a new Criterion object.
Parameters:

prepare(self)

source code 
Computes the symbolic expression of the loss.

Attention: It must be implemented in derived classes.


   Home       Trees       Indices       Help   
Generated by Epydoc 3.0.2 on Thu Aug 20 13:34:14 2015 http://epydoc.sourceforge.net