:ref:`api_paddle_fluid_clip_ClipGradByValue` ). If you want better convergence, it is recommended
to use :ref:`api_paddle_fluid_clip_ClipGradByGlobalNorm` . Default None, meaning there is no gradient clipping.
exclude_from_weight_decay_fn (function, optional): whether to skip weight decay for a parameter when this function returns True while take the parameter as input.
always_adapt (bool, optional): whether to use Layer-wise LR adaptation. By default, skip adaptation on parameters that are
excluded from weight decay, unless always_adapt == True, then always enable LR adaptation.
name(str|None): For detailed information, please refer to
:ref:`api_guide_Name` . Usually name is no need to set and None by default.