Changeset 40:d1cc6a016f2c in orange-reliability


Ignore:
Timestamp:
10/02/13 16:22:32 (7 months ago)
Author:
markotoplak
Branch:
default
Message:

Added stubs of ICV and Stacking to the documentation

Files:
2 edited

Legend:

Unmodified
Added
Removed
  • docs/rst/Orange.evaluation.reliability.rst

    r39 r40  
    8888.. autoclass:: ParzenWindowDensityBased 
    8989 
     90Internal cross validation (ICV) 
     91------------------------------- 
     92 
     93.. autoclass:: ICV 
     94 
     95 
     96Stacked generalization (Stacking) 
     97------------------------------- 
     98 
     99.. autoclass:: Stacking 
     100 
    90101Reference Estimate for Classification (:math:`O_{ref}`) 
    91102------------------------------------------------------- 
  • orangecontrib/reliability/__init__.py

    r38 r40  
    219219        Name (string) of reliability estimation method used. 
    220220 
    221     .. attribute:: icv_method 
    222  
    223         An integer ID of reliability estimation method that performed best, 
    224         as determined by ICV, and of which estimate is stored in the 
    225         :obj:`estimate` field. (:obj:`None` when ICV was not used.) 
    226  
    227     .. attribute:: icv_method_name 
    228  
    229         Name (string) of reliability estimation method that performed best, 
    230         as determined by ICV. (:obj:`None` when ICV was not used.) 
    231  
    232     """ 
    233     def __init__(self, estimate, signed_or_absolute, method, icv_method= -1): 
     221    """ 
     222    def __init__(self, estimate, signed_or_absolute, method): 
    234223        self.estimate = estimate 
    235224        self.signed_or_absolute = signed_or_absolute 
    236225        self.method = method 
    237226        self.method_name = METHOD_NAME[method] 
    238         self.icv_method = icv_method 
    239         self.icv_method_name = METHOD_NAME[icv_method] if icv_method != -1 else "" 
    240227        self.text_description = None 
    241228 
     
    435422    for a single classifier. If instances for measuring predictions 
    436423    are given as a parameter, this class can only compute their reliability, 
    437     which allows less memory use.  
     424    which saves memory.  
    438425 
    439426    """ 
Note: See TracChangeset for help on using the changeset viewer.