splitter.py 文件源码

def _info_nominal(self, x, y):
        """ Info for nominal feature feature_values
        :math: p(a)H(a) :math: from
        https://en.wikipedia.org/wiki/ID3_algorithm

        Parameters
        ----------
        x : np.array of shape [n remaining examples]
            containing feature values
        y : np.array of shape [n remaining examples]
            containing relevent class

        Returns
        -------
        : float
            information for remaining examples given feature
        """
        info = 0
        n = x.shape[0]
        items, count = unique(x)
        for value, p in zip(items, count):
            info += p * self._entropy(y[x == value])
        return CalcRecord(CalcRecord.NOM,
                          info * np.true_divide(1, n),
                          attribute_counts=count)