How to use the chaospy.distributions.evaluation.get_forward_cache function in chaospy

def _fwd_cache(self, cache):
        dist = evaluation.get_forward_cache(self.prm["dist"], cache)
        if not isinstance(dist, Dist):
            return numpy.arcsin(dist)
        return self

def _inv_cache(self, cache):
        dist = evaluation.get_forward_cache(self.prm["dist"], cache)
        if not isinstance(dist, Dist):
            return numpy.tan(dist)
        return self

def _inv_cache(self, cache):
        left = evaluation.get_forward_cache(self.prm["left"], cache)
        right = evaluation.get_forward_cache(self.prm["right"], cache)
        if not isinstance(left, Dist) and not isinstance(right, Dist):
            return left+right
        return self

def _fwd_cache(self, cache):
        dist = evaluation.get_forward_cache(self.prm["dist"], cache)
        if not isinstance(dist, Dist):
            return numpy.arctanh(dist)
        return self

Example:
            >>> print(chaospy.Uniform().fwd([-0.5, 0.5, 1.5, 2.5]))
            [0.  0.5 1.  1. ]
            >>> print(chaospy.Pow(chaospy.Uniform(), 2).fwd([-0.5, 0.5, 1.5, 2.5]))
            [0.         0.70710678 1.         1.        ]
            >>> print(chaospy.Pow(chaospy.Uniform(1, 2), -1).fwd([0.4, 0.6, 0.8, 1.2]))
            [0.         0.33333333 0.75       1.        ]
            >>> print(chaospy.Pow(2, chaospy.Uniform()).fwd([-0.5, 0.5, 1.5, 2.5]))
            [0.        0.        0.5849625 1.       ]
            >>> print(chaospy.Pow(2, chaospy.Uniform(-1, 0)).fwd([0.4, 0.6, 0.8, 1.2]))
            [0.         0.26303441 0.67807191 1.        ]
            >>> print(chaospy.Pow(2, 3).fwd([7, 8, 9]))
            [0. 1. 1.]
        """
        left = evaluation.get_forward_cache(left, cache)
        right = evaluation.get_forward_cache(right, cache)

        if isinstance(left, Dist):
            if isinstance(right, Dist):
                raise StochasticallyDependentError(
                    "under-defined distribution {} or {}".format(left, right))

        elif not isinstance(right, Dist):
            return numpy.inf

        else:
            assert numpy.all(left > 0), "imaginary result"

            y = (numpy.log(numpy.abs(xloc) + 1.*(xloc <= 0)) /
                 numpy.log(numpy.abs(left)+1.*(left == 1)))

            out = evaluation.evaluate_forward(right, y, cache=cache.copy())

def _inv_cache(self, cache):
        dist = evaluation.get_forward_cache(self.prm["dist"], cache)
        if not isinstance(dist, Dist):
            return 10**dist
        return self

def _inv_cache(self, cache):
        dist = evaluation.get_forward_cache(self.prm["dist"], cache)
        if not isinstance(dist, Dist):
            return numpy.arccos(dist)
        return self

def _inv_cache(self, cache):
        dist = evaluation.get_forward_cache(self.prm["dist"], cache)
        if not isinstance(dist, Dist):
            return numpy.sinh(dist)
        return self

def _fwd_cache(self, cache):
        left = evaluation.get_forward_cache(self.prm["left"], cache)
        right = evaluation.get_forward_cache(self.prm["right"], cache)
        if not isinstance(left, Dist) and not isinstance(right, Dist):
            return left*right
        return self

def _pdf(self, xloc, left, right, cache):
        """
        Probability density function.

        Example:
            >>> print(chaospy.Uniform().pdf([-2, 0, 2, 4]))
            [0. 1. 0. 0.]
            >>> print(chaospy.Add(chaospy.Uniform(), 2).pdf([-2, 0, 2, 4]))
            [0. 0. 1. 0.]
            >>> print(chaospy.Add(2, chaospy.Uniform()).pdf([-2, 0, 2, 4]))
            [0. 0. 1. 0.]
            >>> print(chaospy.Add(1, 1).pdf([-2, 0, 2, 4])) # Dirac logic
            [ 0.  0. inf  0.]
        """
        left = evaluation.get_forward_cache(left, cache)
        right = evaluation.get_forward_cache(right, cache)

        if isinstance(left, Dist):
            if isinstance(right, Dist):
                raise evaluation.DependencyError(
                    "under-defined distribution {} or {}".format(left, right))
        elif not isinstance(right, Dist):
            return numpy.inf
        else:
            left, right = right, left

        xloc = (xloc.T-numpy.asfarray(right).T).T
        output = evaluation.evaluate_density(left, xloc, cache=cache)
        assert output.shape == xloc.shape
        return output