erfinv:
  description: |
    Returns the result of the inverse error function with `input`, which is defined in the range `(-1, 1)` as:

    .. math::

        erfinv(erf(x)) = x

    where :math:`x` is the `input`.

    Args:
        input (Tensor): The input tensor to compute with.
            Supported dtypes:

            - Ascend: float16, float32, int8, int16, int32, int64, uint8, bool.
            - GPU/CPU: float16, float32 or float64.

    Returns:
        Tensor, has the same shape and dtype as `input`.

    :raise TypeError: If dtype of `input` is not as follows

        - Ascend: float16, float32, int8, int16, int32, int64, uint8, bool.
        - GPU/CPU: float16, float32 or float64.

    Supported Platforms:
        ``Ascend`` ``GPU`` ``CPU``

    Examples:
        >>> import mindspore
        >>> import numpy as np
        >>> from mindspore import Tensor, ops
        >>> input = Tensor(np.array([0, 0.5, -0.9]), mindspore.float32)
        >>> output = ops.erfinv(input)
        >>> print(output)
        [ 0.          0.47695306 -1.1630805 ]