Changeset 7584:69e833700856 in orange
- 02/05/11 00:00:21 (3 years ago)
- 1 edited
r7560 r7584 37 37 Many machine learning techniques generate a set different solutions or have to 38 38 choose, as for instance in classification tree induction, between different 39 attributes. The most trivial solution is to iterate through the candidates, 39 es. The most trivial solution is to iterate through the candidates, 40 40 compare them and remember the optimal one. The problem occurs, however, when 41 41 there are multiple candidates that are equally good, and the naive approaches … … 59 59 60 60 The following snippet loads the data set lymphography and prints out the 61 attribute with the highest information gain. 61 e with the highest information gain. 62 62 63 63 part of `misc-selection-bestonthefly.py`_ (uses `lymphography.tab`_) … … 66 66 :lines: 7-16 67 67 68 Our candidates are tuples gain ratios and attributes, so we set 68 Our candidates are tuples gain ratios and es, so we set 69 69 :obj:`callCompareOn1st` to make the compare function compare the first element 70 70 (gain ratios). We could achieve the same by initializing the object like this:
Note: See TracChangeset for help on using the changeset viewer.