mindspore.ops.Logit

class mindspore.ops.Logit(eps=- 1.0)[source]

Calculate the logit of a tensor element-wise. Element in x is clamped to [eps, 1-eps].

Warning

This is an experimental API that is subject to change or deletion.

Refer to mindspore.ops.logit() for more details.

Parameters: eps (float, optional) – The epsilon. The input clamp bound is defined as [eps, 1-eps]. Default: -1.0 .

Inputs:

x (Tensor) - The input tensor of type float16, float32 or float64.

Outputs:

Tensor, with the same shape and dtype as the x.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import numpy as np
>>> from mindspore import Tensor, ops
>>> x = Tensor(np.array([0.1, 0.2, 0.3]).astype(np.float32))
>>> op = ops.Logit(eps=1e-5)
>>> output = op(x)
>>> print(output)
[-2.1972246 -1.3862944 -0.8472978]