remove useless code

[zhoutianzi666]

🤖 Paddle-CI-Agent | pr_review | 2026-06-12 21:27:01

📋 Review 摘要

PR 概述：复用 DeepseekV3MLAAttention 初始化 DSA attention，并移除空类 docstring。
变更范围：fastdeploy/model_executor/models/deepseek_v3.py
影响面 Tag：[Models]

问题

未发现阻塞性问题。PR 规范问题在下面章节报，不要在这里重复。

历史 Findings 修复情况

Finding	问题	状态
F1	glm_moe_dsa 主 DSA rotary_emb 使用错误 rope_theta	✅ 已修复

📝 PR 规范检查

标题缺少官方 Tag，PR 描述仍是空模板。建议替换为以下内容。

标题建议（可直接复制）：

• [Models] Initialize DeepSeek V32 DSA attention correctly

PR 描述建议（点击展开，可直接复制）

## Motivation
Initialize DeepseekV32DSAAttention through the MLA attention base class and remove empty class docstrings.

## Modifications
- `fastdeploy/model_executor/models/deepseek_v3.py`: pass `fd_config`, `layer_id`, and `prefix` to the parent attention initializer for DSA attention.
- Remove redundant empty class docstrings in DeepSeek V3 model classes.

## Usage or Command
N/A

## Accuracy Tests
N/A

## Checklist

- [x] Add at least a tag in the PR title.
  - Tag list: [`[FDConfig]`,`[APIServer]`,`[Engine]`, `[Scheduler]`, `[PD Disaggregation]`, `[Executor]`, `[Graph Optimization]`, `[Speculative Decoding]`, `[RL]`, `[Models]`, `[Quantization]`, `[Loader]`, `[OP]`, `[KVCache]`, `[DataProcessor]`, `[BugFix]`, `[Docs]`, `[CI]`, `[Optimization]`, `[Feature]`, `[Benchmark]`, `[Others]`, `[XPU]`, `[HPU]`, `[GCU]`, `[DCU]`, `[Iluvatar]`, `[Metax]`]
  - You can add new tags based on the PR content, but the semantics must be clear.
- [ ] Format your code, run `pre-commit` before commit.
- [ ] Add unit tests. Please write the reason in this PR if no unit tests.
- [ ] Provide accuracy results.
- [ ] If the current PR is submitting to the `release` branch, make sure the PR has been submitted to the `develop` branch, then cherry-pick it to the `release` branch with the `[Cherry-Pick]` PR tag.

总体评价

当前代码在父类构造 rotary_emb 前按 glm_moe_dsa 改用 rope_parameters 下的 rope_theta，DeepseekV32DSAAttention 继承初始化后主 attention 与 indexer attention 使用同一个修正后的 RoPE base，历史问题已修复。未发现新的代码阻塞项；仍建议补齐 PR 标题、描述和测试/精度说明。

[chang-wenbin]

LGTM

[codecov-commenter]

Codecov Report

❌ Patch coverage is 16.66667% with 5 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@e287d0b). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/model_executor/models/deepseek_v3.py	16.66%	5 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #8042   +/-   ##
==========================================
  Coverage           ?   67.55%           
==========================================
  Files              ?      475           
  Lines              ?    66571           
  Branches           ?    10258           
==========================================
  Hits               ?    44969           
  Misses             ?    18739           
  Partials           ?     2863           

Flag	Coverage Δ
GPU	77.55% <16.66%> (?)
XPU	6.98% <0.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[PaddlePaddle-bot]

🤖 Paddle-CI-Agent | pr_review | 2026-06-12 21:27:01

📋 Review 摘要

PR 概述：复用 DeepseekV3MLAAttention 初始化 DSA attention，并移除空类 docstring。
变更范围：fastdeploy/model_executor/models/deepseek_v3.py
影响面 Tag：[Models]

问题

未发现阻塞性问题。PR 规范问题在下面章节报，不要在这里重复。

历史 Findings 修复情况

Finding	问题	状态
F1	glm_moe_dsa 主 DSA rotary_emb 使用错误 rope_theta	✅ 已修复

📝 PR 规范检查

标题缺少官方 Tag，PR 描述仍是空模板。建议替换为以下内容。

标题建议（可直接复制）：

• [Models] Initialize DeepSeek V32 DSA attention correctly

PR 描述建议（点击展开，可直接复制）

## Motivation
Initialize DeepseekV32DSAAttention through the MLA attention base class and remove empty class docstrings.

## Modifications
- `fastdeploy/model_executor/models/deepseek_v3.py`: pass `fd_config`, `layer_id`, and `prefix` to the parent attention initializer for DSA attention.
- Remove redundant empty class docstrings in DeepSeek V3 model classes.

## Usage or Command
N/A

## Accuracy Tests
N/A

## Checklist

- [x] Add at least a tag in the PR title.
  - Tag list: [`[FDConfig]`,`[APIServer]`,`[Engine]`, `[Scheduler]`, `[PD Disaggregation]`, `[Executor]`, `[Graph Optimization]`, `[Speculative Decoding]`, `[RL]`, `[Models]`, `[Quantization]`, `[Loader]`, `[OP]`, `[KVCache]`, `[DataProcessor]`, `[BugFix]`, `[Docs]`, `[CI]`, `[Optimization]`, `[Feature]`, `[Benchmark]`, `[Others]`, `[XPU]`, `[HPU]`, `[GCU]`, `[DCU]`, `[Iluvatar]`, `[Metax]`]
  - You can add new tags based on the PR content, but the semantics must be clear.
- [ ] Format your code, run `pre-commit` before commit.
- [ ] Add unit tests. Please write the reason in this PR if no unit tests.
- [ ] Provide accuracy results.
- [ ] If the current PR is submitting to the `release` branch, make sure the PR has been submitted to the `develop` branch, then cherry-pick it to the `release` branch with the `[Cherry-Pick]` PR tag.

总体评价

当前代码在父类构造 rotary_emb 前按 glm_moe_dsa 改用 rope_parameters 下的 rope_theta，DeepseekV32DSAAttention 继承初始化后主 attention 与 indexer attention 使用同一个修正后的 RoPE base，历史问题已修复。未发现新的代码阻塞项；仍建议补齐 PR 标题、描述和测试/精度说明。