Changeset 7363:deacf0a01d90 in orange


Ignore:
Timestamp:
02/04/11 00:21:29 (3 years ago)
Author:
miha <miha.stajdohar@…>
Branch:
default
Convert:
a3566824cc65ef1bd0588ee7e4e09acc1fdc376d
Message:
 
File:
1 edited

Legend:

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

    r7318 r7363  
    88.. index:: Support Vector Machines Classification 
    99 
    10 Interface to the LibSVM library (a library for support vector machines 
    11 - http://www.csie.ntu.edu.tw/~cjlin/papers/libsvm.ps.gz) 
    12  
    13 .. note:: On some data-sets SVM can perform very badly. It is a known fact that 
    14           SVM can be very sensitive to the proper choice of the parameters. 
    15           If you are having problems with the learner's accuracy try scaling the 
     10A collection of classes that wrap the  
     11`LibSVM library <http://www.csie.ntu.edu.tw/~cjlin/libsvm/>`_ (a library for  
     12`support vector machines <http://en.wikipedia.org/wiki/Support_vector_machine>`_) 
     13 
     14.. note:: On some data-sets SVM can perform very poorly. SVM can be very  
     15          sensitive to the proper choice of the parameters. If you are having  
     16          problems with the learner's accuracy try scaling the 
    1617          data and using different parameters or choose an easier approach 
    1718          and use the :obj:`SVMLearnerEasy` class which does this automatically 
     
    8081 
    8182======================================= 
    82 SVM based Recursive Feature Elimination 
     83SVM-Based Recursive Feature Elimination 
    8384======================================= 
    8485 
     
    9394.. _iris.tab: code/iris.tab 
    9495.. _vehicle.tab: code/vehicle.tab 
     96 
     97References 
     98==========  
     99 
     100C.-W. Hsu, C.-C. Chang, C.-J. Lin. A practical guide to support vector  
     101classification 
    95102 
    96103""" 
     
    167174    :type degree: int 
    168175    :param gamma: kernel parameter (Polynomial/RBF/Sigmoid) 
    169         (default 1/number_of_examples) 
     176        (default 1/number_of_instances) 
    170177    :type gamma: float 
    171178    :param coef0: kernel parameter (Polynomial/Sigmoid) (default 0) 
     
    378385     
    379386    .. note:: Note that meta attributes don't need to be registered with 
    380         the data-set domain, or present in all the examples. Use this if you 
     387        the data-set domain, or present in all the instances. Use this if you 
    381388        are learning from large sparse data-sets. 
    382389     
Note: See TracChangeset for help on using the changeset viewer.