mindspore-ai / mindspore-ai/hyper-parallel

[Bug]:[master]hyper-parallel主线代码重构,2BV2_LLM未适配,报错:No module named 'hyper_parallel.core.shard.api,需要打对应的patch

Open Beginner friendly
#159 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
53
Forks
63
Avg merge
23h 45m
Merged PRs (30d)
63

Description

Checklist
  • 1. I have searched the existing issues (https://gitcode.com/mindspore/hyper-parallel/issues)
  • 2. I have read the relevant documentation.
  • 3. I have created a minimal reproduction case that clearly demonstrates the issue, including a complete code example and the error message with full traceback and error logs.
🐛 Describe the bug

一、镜像信息
swr-aifm-exp-mt7nab.swr-pro.myhuaweicloud.com/aifm.infra.training/train_910b-pytorch_2.6.0-cann_8.5.1.b010-py_3.11-hce_2.0.2509-aarch64-snt9b23-20260123:1.2.1

二、环境信息
硬件:A2

三、报错信息
Traceback (most recent call last):
File "/home/code/AI_Infra_TrainTest/testcases/TrainningScenarios/test_pangu_2B_llm_fsdp/pangu_sophon_pytorch/PanGu/hyper_parallel/init.py", line 34, in
from hyper_parallel.core.shard.api import shard_module
ModuleNotFoundError: No module named 'hyper_parallel.core.shard.api'

具体打patch操作:在原有patch的基础上,需要修改pangu_sophon_pytorch/PanGu/hyper_parallel/init.py ,把34、35行去掉
image.png

四、复现步骤

五、预期结果
网络能正常训练,精度零误差对齐。

Expected behavior

网络能正常训练,精度零误差对齐。

Additional context
Environment info

A2

Thanks for contributing 🎉!

schema_version: 1
source: gitcode
gitcode_repo: mindspore/hyper-parallel
gitcode_issue: 372
source_url: https://gitcode.com/mindspore/hyper-parallel/issues/372

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start in pangu_sophon_pytorch/PanGu/hyper_parallel/init.py and inspect the imports around lines 34–35, along with the existing patch referenced in the issue. Reproduce the ModuleNotFoundError in the A2 environment, then verify that the 2B V2 LLM training network runs successfully with zero accuracy deviation after the import adjustment.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
distributed-systems
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
74/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.