Changeset 11377:1f01a3c46101 in orange


Ignore:
Timestamp:
03/07/13 15:01:37 (14 months ago)
Author:
Ales Erjavec <ales.erjavec@…>
Branch:
default
Message:

Added notes to svm learner documentation about discrete feature translation.

File:
1 edited

Legend:

Unmodified
Added
Removed
  • Orange/classification/svm/__init__.py

    r10955 r11377  
    8686    :param shrinking: use shrinking heuristics 
    8787    :type shrinking: bool 
    88     :param normalization: normalize the input data prior to learning 
     88    :param normalization: normalize the input data prior to learning into 
     89        range [0..1] and replace discrete features with indicator columns 
     90        one for each value of the feature using 
     91        :class:`~Orange.data.continuization.DomainContinuizer` class 
    8992        (default ``True``) 
    9093    :type normalization: bool 
     
    826829        :type bias: float 
    827830 
    828         :param normalization: Normalize the input data prior to learning 
    829             (default True) 
     831        :param normalization: Normalize the input data into range [0..1] prior 
     832            to learning (default ``True``) 
    830833        :type normalization: bool 
     834 
     835        .. note:: If the training data contains discrete features they are 
     836            replaced by indicator columns one for each value of the feature 
     837            regardless of the value of `normalization`. This is different 
     838            then in :class:`SVMLearner` where this is done only if 
     839            `normalization` is ``True``. 
    831840 
    832841        Example 
Note: See TracChangeset for help on using the changeset viewer.