Package topo :: Package learningfns :: Module basic :: Class Oja

Class Oja

                                  object --+        
                                           |        
base.parameterizedobject.ParameterizedObject --+    
                                               |    
                base.functionfamilies.LearningFn --+
                                                   |
                                                  Oja

Oja's rule (Oja, 1982; Dayan and Abbott, 2001, equation 8.16.)

Hebbian rule with soft multiplicative normalization, tending the weights toward a constant sum-squared value over time. Thus this function does not normally need a separate output_fn for normalization.

Nested Classes

[hide private]

Inherited from base.parameterizedobject.ParameterizedObject: __metaclass__

Instance Methods

[hide private]

__call__(self, input_activity, unit_activity, weights, single_connection_learning_rate)
Apply this learning function given the input and output activities and current weights.

source code

Inherited from base.parameterizedobject.ParameterizedObject: __getstate__, __init__, __repr__, __setstate__, __str__, as_uninitialized, debug, defaults, force_new_dynamic_value, get_param_values, get_value_generator, inspect_value, message, print_param_values, script_repr, verbose, warning

Inherited from base.parameterizedobject.ParameterizedObject (private): _check_params, _instantiate_param, _set_name, _setup_params

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

Class Methods

[hide private]

Inherited from base.parameterizedobject.ParameterizedObject: params, print_param_defaults

Class Variables

[hide private]

alpha = Number(default= 0.1, bounds= (0, None))

Inherited from base.parameterizedobject.ParameterizedObject: name, print_level

Properties

[hide private]

Inherited from object: __class__

Method Details

[hide private]

call(self, input_activity, unit_activity, weights, single_connection_learning_rate)
(Call operator)

source code

Apply this learning function given the input and output activities and current weights.

Must be implemented by subclasses.

Overrides: base.functionfamilies.LearningFn.__call__: (inherited documentation)

Class Oja

__call__(self, input_activity, unit_activity, weights, single_connection_learning_rate) (Call operator)

call(self, input_activity, unit_activity, weights, single_connection_learning_rate)
(Call operator)