mindspore.ops.clamp

mindspore.ops.clamp(input, min=None, max=None)[source]

Clamp all elements of the input tensor within the range [min, max].

\[\begin{split}out_i= \left\{ \begin{array}{align} max & \text{ if } input_i\ge max \\ input_i & \text{ if } min \lt input_i \lt max \\ min & \text{ if } input_i \le min \\ \end{array}\right.\end{split}\]

Note

  • min and max cannot be None at the same time;

  • If min is None , there is no lower bound.

  • if max is None , there is no upper bound.

  • If min is greater than max, the value of all elements in Tensor will be set to max;

Parameters

  • input (Tensor) – The input tensor.

  • min (Union(Tensor, float, int), optional) – The minimum value. Default None .

  • max (Union(Tensor, float, int), optional) – The maximum value. Default None .

Returns

Tensor

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> # case 1: `min` and `max` are integer
>>> input = mindspore.tensor([[1, 25, 5, 7], [4, 11, 6, 21]])
>>> mindspore.ops.clamp(input, 5, 20)
Tensor(shape=[2, 4], dtype=Int64, value=
[[ 5, 20,  5,  7],
 [ 5, 11,  6, 20]])
>>>
>>> # case 2: If `min` and `max` are tensors, their shapes need to be broadcastable with input.
>>> min = mindspore.tensor([2, 4, 6, 8])
>>> max = mindspore.tensor([10, 12, 14, 18])
>>> mindspore.ops.clamp(input, min, max)
Tensor(shape=[2, 4], dtype=Int64, value=
[[ 2, 12,  6,  8],
 [ 4, 11,  6, 18]])