Bratmobile
Public Member Functions | Protected Attributes | List of all members
ThresholdLearner Class Referenceabstract
Collaboration diagram for ThresholdLearner:
Collaboration graph
[legend]

Public Member Functions

void set_learning_rate (float f)
 
Bundle get_Di_weights ()
 
Bundle get_Dn_weights ()
 
void log ()
 
void make_log ()
 
virtual void Di_tune (const Bundle &error, const Bundle &x)=0
 Tune weights for Di. More...
 
virtual Bundle update_bundle (const Bundle &error, const Bundle &x)=0
 Updates a threshold according to a custom learning rule. More...
 
Threshold get_weighted (const Threshold &t)
 

Protected Attributes

Bundle Di_weights
 
Bundle Dn_weights
 
float mu =0.001
 

Member Function Documentation

◆ Di_tune()

virtual void ThresholdLearner::Di_tune ( const Bundle error,
const Bundle x 
)
pure virtual

Tune weights for Di.

Parameters
errorthe error

◆ update_bundle()

virtual Bundle ThresholdLearner::update_bundle ( const Bundle error,
const Bundle x 
)
pure virtual

Updates a threshold according to a custom learning rule.

Parameters
error
xthe input
Returns
Bundle

The documentation for this class was generated from the following file: