Compound comparisons via univariate ANOVA.
This measure compute an ANOVA F-score per each feature, for each one-vs-rest comparision for all unique labels in a dataset. Each F-score vector for each comparision is included in the return datasets as a separate samples. Corresponding p-values are avialable in feature attributes named ‘fprob_X’, where X is the name of the actual comparision label. Note that p-values are only available, if SciPy is installed. The comparison labels for each F-vectore are also stored as ‘targets’ sample attribute in the returned dataset.
Notes
Available conditional attributes:
(Conditional attributes enabled by default suffixed with +)
Parameters : | space : str
enable_ca : None or list of str
disable_ca : None or list of str
null_dist : instance of distribution estimator
auto_train : bool
force_train : bool
postproc : Node instance, optional
descr : str
|
---|