Package pylearn :: Package sandbox :: Package rbm :: Module model
[hide private]

Module model

source code

The model for an autoassociator for sparse inputs, using Ronan Collobert + Jason Weston's sampling trick (2008).

Classes [hide private]
Model
Functions [hide private]
 
sigmoid(v) source code
 
sample(v) source code
 
crossentropy(output, target)
Compute the crossentropy of binary output wrt binary target.
source code

Imports: parameters, numpy, dot, random, pylearn


Function Details [hide private]

sigmoid(v)

source code 
To Do:
  • Move to pylearn.more_numpy
  • Fix to avoid floating point overflow.

sample(v)

source code 

To Do: Move to pylearn.more_numpy

crossentropy(output, target)

source code 

Compute the crossentropy of binary output wrt binary target.

Note: We do not sum, crossentropy is computed by component.

To Do:
  • Rewrite as a scalar, and then broadcast to tensor.
  • Move to pylearn.more_numpy
  • Fix to avoid floating point overflow.