[Typing][C-73] Add type annotations for python/paddle/incubate/nn/functional/fused_layer_norm.py

[haoyu2022]

PR Category

User Experience

PR Types

Improvements

Description

类型标注：

• python/paddle/incubate/nn/functional/ fused_layer_norm.py

Related links

• [Type Hints] 为公开 API 标注类型提示信息 #65008

@SigureMo @megemini

[paddle-bot]

你的PR提交成功，感谢你对开源项目的贡献!
请关注后续CI自动化测试结果，详情请参考Paddle-CI手册。
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

[luotao1]

PR-CI-Codestyle-Check 没过 @haoyu2022

[megemini]

@haoyu2022 帮忙改一下 Codestyle 吧 ～ 谢谢 ～

[haoyu2022]

@megemini 好的，稍等😊

[paddle-ci-bot]

Sorry to inform you that 48e7400's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually.

[megemini]

根据 residual 使用 overload

from __future__ import annotations

from typing import TYPE_CHECKING, overload

import paddle
from paddle import _C_ops
from paddle.framework import LayerHelper, in_dynamic_mode, in_pir_mode

if TYPE_CHECKING:
    from paddle import Tensor

@overload
def fused_layer_norm(
    x: Tensor,
    norm_weight: Tensor,
    norm_bias: Tensor,
    epsilon: float,
    residual_alpha: float = ...,
    begin_norm_axis: int = ...,
    bias: Tensor | None = ...,
    residual: None = ...,
    quant_scale: float = ...,
    quant_round_type: float = ...,
    quant_max_bound: float = ...,
    quant_min_bound: float = ...,
) -> Tensor: ...

@overload
def fused_layer_norm(
    x: Tensor,
    norm_weight: Tensor,
    norm_bias: Tensor,
    epsilon: float,
    residual_alpha: float = ...,
    begin_norm_axis: int = ...,
    bias: Tensor | None = ...,
    residual: Tensor = ...,
    quant_scale: float = ...,
    quant_round_type: float = ...,
    quant_max_bound: float = ...,
    quant_min_bound: float = ...,
) -> tuple[Tensor, Tensor]: ...

def fused_layer_norm(
    x,
    norm_weight,
    norm_bias,
    epsilon,
    residual_alpha=1.0,
    begin_norm_axis=1,
    bias=None,
    residual=None,
    quant_scale=-1,
    quant_round_type=0,
    quant_max_bound=0,
    quant_min_bound=0,
):
    r"""
    Apply Fused LayerNorm kernel. Also support LayerNorm(bias + residual_alpha * residual + x) fused pattern.

[haoyu2022]

好的，感谢

[megemini]

看一下上次 review 的代码 ... ...

[haoyu2022]

看一下上次 review 的代码 ... ...

我明白意思了。

[megemini]

LGTM ～

[SigureMo]

LGTMeow