sklearn.feature_selection
.chi2¶

sklearn.feature_selection.
chi2
(X, y)[source]¶ Compute chisquared stats between each nonnegative feature and class.
This score can be used to select the n_features features with the highest values for the test chisquared statistic from X, which must contain only nonnegative features such as booleans or frequencies (e.g., term counts in document classification), relative to the classes.
Recall that the chisquare test measures dependence between stochastic variables, so using this function “weeds out” the features that are the most likely to be independent of class and therefore irrelevant for classification.
Parameters: X : {arraylike, sparse matrix}, shape = (n_samples, n_features_in)
Sample vectors.
y : arraylike, shape = (n_samples,)
Target vector (class labels).
Returns: chi2 : array, shape = (n_features,)
chi2 statistics of each feature.
pval : array, shape = (n_features,)
pvalues of each feature.
See also
f_classif
 ANOVA Fvalue between labe/feature for classification tasks.
f_regression
 Fvalue between label/feature for regression tasks.
Notes
Complexity of this algorithm is O(n_classes * n_features).