Changeset 10179:cf5dc2ee7eec in orange
- 02/13/12 09:32:43 (22 months ago)
- 1 edited
r9994 r10179 6 6 ================================== 7 7 8 This module includes functions for data sampling and splitting, and for 9 testing learners. It implements cross-validation, leave-one out, random 10 sampling and learning curves. All functions return their results in the same 11 format - an instance of :obj:`ExperimentResults`, described at the end of the 12 page, or, in case of learning curves, a list of :obj:`ExperimentResults`. This 13 object(s) can be passed to statistical function for model evaluation 14 (classification accuracy, Brier score, ROC analysis...) available in 15 module :obj:`Orange.evaluation.scoring`. 8 There are many ways to test prediction models on data. Orange includes 9 methods for cross-validation, leave-one out, random sampling and learning 10 curves. This methods handle learning of models and prediction of new 11 examples; they return :obj:`ExperimentResults` which can be passed to 12 :obj:`~Orange.evaluation.scoring` functions to evaluate model. 13 14 16 15 17 16 Your scripts will thus basically conduct experiments using methods of … … 22 21 description of :obj:`TestedExample` and :obj:`ExperimentResults` are 23 22 available at the end of this page. 24 25 .. note:: Orange has been "de-randomized". Running the same script twice 26 will generally give the same results, unless special care is taken to 27 randomize it. This is opposed to the previous versions where special 28 care needed to be taken to make experiments repeatable. 29 See argument obj:`random_generator` for the explanation. 30 23 31 24 Example scripts in this section suppose that the data is loaded and a
Note: See TracChangeset for help on using the changeset viewer.