Changeset 10335:18f3ac9e1ec6 in orange for Orange/multitarget/tree.py
 Timestamp:
 02/22/12 12:07:50 (2 years ago)
 Branch:
 default
 File:

 1 edited
Legend:
 Unmodified
 Added
 Removed

Orange/multitarget/tree.py
r10092 r10335 62 62 class MultitargetVariance(Orange.feature.scoring.Score): 63 63 """ 64 A multitarget score that ranks features based on the variance of the 65 subsets. A weighted distance can be used to compute the variance. 64 A multitarget score that ranks features based on the average class 65 variance of the subsets. 66 67 To compute it, a prototype has to be defined for each subset. Here, it 68 is just the mean vector of class variables. Then the sum of squared 69 distances to the prototypes is computed in each subset. The final score 70 is obtained as the average of subset variances (weighted, to account for 71 subset sizes). 72 73 Weights can be passed to the constructor to normalize classes with values 74 of different magnitudes or to increase the importance of some classes. In 75 this case, class values are first scaled according to the given weights. 66 76 """ 67 77 68 78 def __init__(self, weights=None): 69 79 """ 70 :param weights: Weights of the features used when computing distances.71 If None, all weights are set to 1.80 :param weights: Weights of the class variables used when computing 81 distances. If None, all weights are set to 1. 72 82 :type weigts: list 73 83 """
Note: See TracChangeset
for help on using the changeset viewer.