def dot()

in flink-ml-python/pyflink/ml/linalg.py [0:0]

• 21 lines of code
• 7 McCabe index (conditional complexity)

    def dot(self, other: Union[Vector, np.ndarray, Sized]) -> np.ndarray:
        """
        Dot product of two Vectors.

        Examples:
        ::

            >>> sparse = SparseVector(4, [1, 3], [3.0, 4.0])
            >>> sparse.dot(sparse)
            25.0
            >>> sparse.dot(array.array('d', [1., 2., 3., 4.]))
            22.0
            >>> sparse2 = SparseVector(4, [2], [1.0])
            >>> sparse.dot(sparse2)
            0.0
            >>> sparse.dot(np.array([[1, 1], [2, 2], [3, 3], [4, 4]]))
            array([22., 22.])
        """
        if isinstance(other, np.ndarray):
            if other.ndim not in (2, 1):
                raise ValueError('Cannot call dot with %d-dimensional array' % other.ndim)
            assert len(self) == other.shape[0], "dimension mismatch"
            return np.dot(self._values, other[self._indices])

        assert len(self) == len(other), "dimension mismatch"

        if isinstance(other, DenseVector):
            return np.dot(other.to_array()[self._indices], self._values)
        elif isinstance(other, SparseVector):
            self_cmind = np.in1d(self._indices, other._indices, assume_unique=True)
            self_values = self._values[self_cmind]
            if self_values.size == 0:
                return np.float_(0.0)  # type: ignore
            else:
                other_cmind = np.in1d(other._indices, self._indices, assume_unique=True)
                return np.dot(self_values, other._values[other_cmind])
        else:
            if isinstance(other, (array.array, np.ndarray, list, tuple, range)):
                return self.dot(DenseVector(other))
            raise ValueError('Cannot call with the type %s' % (type(other)))