mindspore.ops.clip_by_global_norm(x, clip_norm=1.0, use_norm=None)[source]

Clips tensor values by the ratio of the sum of their norms.


Input ‘x’ should be a tuple or list of tensors. Otherwise, it will raise an error.

  • x (Union(tuple[Tensor], list[Tensor])) – Input data to clip. The shape of each Tensor in tuple is \((N,*)\) where \(*\) means, any number of additional dimensions.

  • clip_norm (Union(float, int)) – The clipping ratio, it should be greater than 0. Default: 1.0

  • use_norm (None) – The global norm. Default: None. Currently only none is supported.


tuple[Tensor], a clipped Tensor. It has the same data type as x and each Tensor in the output tuple is the same as the original input shape.

Supported Platforms:

Ascend GPU CPU


>>> from mindspore import Tensor, ops
>>> import numpy as np
>>> x1 = np.array([[2., 3.], [1., 2.]]).astype(np.float32)
>>> x2 = np.array([[1., 4.], [3., 1.]]).astype(np.float32)
>>> input_x = (Tensor(x1), Tensor(x2))
>>> out = ops.clip_by_global_norm(input_x, 1.0)
>>> print(out)
(Tensor(shape=[2, 2], dtype=Float32, value=
[[ 2.98142403e-01,  4.47213590e-01],
 [ 1.49071202e-01,  2.98142403e-01]]), Tensor(shape=[2, 2], dtype=Float32, value=
[[ 1.49071202e-01,  5.96284807e-01],
 [ 4.47213590e-01,  1.49071202e-01]]))