OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

[Easy] Optimize clip_grad param description (#151532)

Browse Source 
Fix missing optional description in `clip_grad_norm_` and `clip_grad_value_`

## Test Result

### Before

![image](https://github.com/user-attachments/assets/3393dd4b-a730-4dd4-8304-9b895ac669d4)

![image](https://github.com/user-attachments/assets/220c4738-a728-474b-b06d-b5be7660d150)

### After

![image](https://github.com/user-attachments/assets/5637bb68-3b6d-49a3-8ee1-3af636950aa0)

![image](https://github.com/user-attachments/assets/c0f1d966-a9ba-4fac-a874-9d4955f6e0d6)

Pull Request resolved: https://github.com/pytorch/pytorch/pull/151532
Approved by: https://github.com/Skylion007, https://github.com/albanD
This commit is contained in:
zeshengzong 2025-04-17 16:47:35 +00:00 committed by PyTorch MergeBot
parent c3a18f6126
commit fe90a5c140
1 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
torch/nn/utils/clip_grad.py
View File

@ -197,12 +197,12 @@ def clip_grad_norm_(
parameters (Iterable[Tensor] or Tensor): an iterable of Tensors or a
single Tensor that will have gradients normalized
max_norm (float): max norm of the gradients
norm_type (float): type of the used p-norm. Can be ``'inf'`` for
infinity norm.
error_if_nonfinite (bool): if True, an error is thrown if the total
norm_type (float, optional): type of the used p-norm. Can be ``'inf'`` for
infinity norm. Default: 2.0
error_if_nonfinite (bool, optional): if True, an error is thrown if the total
norm of the gradients from :attr:`parameters` is ``nan``,
``inf``, or ``-inf``. Default: False (will switch to True in the future)
foreach (bool): use the faster foreach-based implementation.
``inf``, or ``-inf``. Default: False
foreach (bool, optional): use the faster foreach-based implementation.
If ``None``, use the foreach implementation for CUDA and CPU native tensors and silently
fall back to the slow implementation for other device types.
Default: ``None``
@ -258,7 +258,7 @@ def clip_grad_value_(
clip_value (float): maximum allowed value of the gradients.
The gradients are clipped in the range
:math:`\left[\text{-clip\_value}, \text{clip\_value}\right]`
foreach (bool): use the faster foreach-based implementation
foreach (bool, optional): use the faster foreach-based implementation
If ``None``, use the foreach implementation for CUDA and CPU native tensors and
silently fall back to the slow implementation for other device types.
Default: ``None``