未验证 提交 c4815707 编写于 作者: H hutuxian 提交者: GitHub

fix doc for adamw (#32438)

上级 d03b0b16
...@@ -59,7 +59,7 @@ class AdamW(Adam): ...@@ -59,7 +59,7 @@ class AdamW(Adam):
weight_decay (float|Tensor, optional): The weight decay coefficient, it can be float or Tensor. The default value is 0.01. weight_decay (float|Tensor, optional): The weight decay coefficient, it can be float or Tensor. The default value is 0.01.
apply_decay_param_fun (function|None, optional): If it is not None, apply_decay_param_fun (function|None, optional): If it is not None,
only tensors that makes apply_decay_param_fun(Tensor.name)==True only tensors that makes apply_decay_param_fun(Tensor.name)==True
will be updated. It only works when we want to specify tensors. will be updated with weight decay. It only works when we want to specify tensors.
Default: None. Default: None.
grad_clip (GradientClipBase, optional): Gradient cliping strategy, it's an instance of grad_clip (GradientClipBase, optional): Gradient cliping strategy, it's an instance of
some derived class of ``GradientClipBase`` . There are three cliping strategies some derived class of ``GradientClipBase`` . There are three cliping strategies
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册