feat(pt): support CPU parallel training with PT #4224

iProzd · 2024-10-16T10:37:35Z

Summary by CodeRabbit

New Features
- Enhanced backend selection for distributed training, allowing for flexible use of NCCL or Gloo based on availability.
Bug Fixes
- Corrected indentation for improved code clarity.
Documentation
- Added comments to explain the logic behind backend selection.

for more information, see https://pre-commit.ci

coderabbitai · 2024-10-16T10:41:06Z

📝 Walkthrough

Walkthrough

The changes in this pull request focus on the get_trainer function in deepmd/pt/entrypoints/main.py, enhancing the initialization of the distributed process group by implementing a flexible backend selection process. The new logic checks for the availability of NCCL and Gloo backends, using NCCL if available, and Gloo as a fallback. If neither backend is found, a RuntimeError is raised. The local_rank variable's indentation has been corrected, and comments have been added for clarity. The overall structure of the train function remains unchanged, with improved handling for model configuration.

Changes

File	Change Summary
deepmd/pt/entrypoints/main.py	Modified `get_trainer` function for flexible backend selection (NCCL or Gloo); corrected indentation of `local_rank` assignment; added comments for clarity. `prepare_trainer_input_single` function remains unchanged but is called in a clearer control flow.

Assessment against linked issues

Objective	Addressed	Explanation
Support CPU parallel training with PT (#4132)	✅

Possibly related PRs

feat(pt): support DeepEval.eval_descriptor #4214: The changes in the DeepEval class and its methods may relate to the overall structure and functionality of the training process, particularly in how descriptors are evaluated, which could be relevant to the modifications in the get_trainer function's control flow.

Suggested reviewers

wanghan-iapcm

Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media?

❤️ Share

🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

Review comments: Directly reply to a review comment made by CodeRabbit. Example:
- I pushed a fix in commit <commit_id>, please review it.
- Generate unit testing code for this file.
- Open a follow-up GitHub issue for this discussion.
Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query. Examples:
- @coderabbitai generate unit testing code for this file.
- @coderabbitai modularize this function.
PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
- @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
- @coderabbitai read src/utils.ts and generate unit testing code.
- @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.
- @coderabbitai help me debug CodeRabbit configuration file.

Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments.

CodeRabbit Commands (Invoked using PR comments)

@coderabbitai pause to pause the reviews on a PR.
@coderabbitai resume to resume the paused reviews.
@coderabbitai review to trigger an incremental review. This is useful when automatic reviews are disabled for the repository.
@coderabbitai full review to do a full review from scratch and review all the files again.
@coderabbitai summary to regenerate the summary of the PR.
@coderabbitai resolve resolve all the CodeRabbit review comments.
@coderabbitai configuration to show the current CodeRabbit configuration for the repository.
@coderabbitai help to get help.

Other keywords and placeholders

Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
Add @coderabbitai anywhere in the PR title to generate the title automatically.

CodeRabbit Configuration File (`.coderabbit.yaml`)

You can programmatically configure CodeRabbit by adding a .coderabbit.yaml file to the root of your repository.
Please see the configuration documentation for more information.
If your editor has YAML language server enabled, you can add the path at the top of this file to enable auto-completion and validation: # yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json

Documentation and Community

Visit our Documentation for detailed information on how to use CodeRabbit.
Join our Discord Community to get help, request features, and share feedback.
Follow us on X/Twitter for updates and announcements.

coderabbitai

Actionable comments posted: 1

📜 Review details

Configuration used: CodeRabbit UI
Review profile: CHILL

📥 Commits

Files that changed from the base of the PR and between 5050f61 and 0b0d943.

📒 Files selected for processing (1)

deepmd/pt/entrypoints/main.py (1 hunks)

🧰 Additional context used

deepmd/pt/entrypoints/main.py

codecov · 2024-10-16T11:43:03Z

Codecov Report

Attention: Patch coverage is 0% with 8 lines in your changes missing coverage. Please review.

Project coverage is 83.51%. Comparing base (5050f61) to head (0b0d943).

Files with missing lines	Patch %	Lines
deepmd/pt/entrypoints/main.py	0.00%	8 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##            devel    #4224      +/-   ##
==========================================
- Coverage   83.52%   83.51%   -0.02%     
==========================================
  Files         542      542              
  Lines       52544    52550       +6     
  Branches     3043     3047       +4     
==========================================
- Hits        43888    43886       -2     
- Misses       7709     7715       +6     
- Partials      947      949       +2

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

njzjz · 2024-10-16T14:41:20Z

deepmd/pt/entrypoints/main.py

+        nccl_available = dist.is_nccl_available()
+        gloo_available = dist.is_gloo_available()
+        # nccl first
+        if nccl_available:
+            backend = "nccl"
+        elif gloo_available:
+            backend = "gloo"


I have a question: when one installs the GPU version but doesn't have a GPU (or set CUDA_VISIBLE_DEVICES to empty), will backend be gloo?

njzjz · 2024-10-17T18:45:36Z

deepmd/pt/entrypoints/main.py

+        if nccl_available:
+            backend = "nccl"
+        elif gloo_available:
+            backend = "gloo"


It seems that one can set the backend to cpu:gloo,cuda:nccl. See https://pytorch.org/docs/stable/distributed.html#torch.distributed.init_process_group

iProzd and others added 2 commits October 16, 2024 18:37

feat(pt): support CPU parallel training with PT

6acfd3b

[pre-commit.ci] auto fixes from pre-commit.com hooks

0b0d943

for more information, see https://pre-commit.ci

iProzd linked an issue Oct 16, 2024 that may be closed by this pull request

[Feature Request] support CPU parallel training with PT #4132

Open

github-actions bot added the Python label Oct 16, 2024

coderabbitai bot reviewed Oct 16, 2024

View reviewed changes

deepmd/pt/entrypoints/main.py Show resolved Hide resolved

iProzd requested review from njzjz and wanghan-iapcm October 16, 2024 12:24

wanghan-iapcm approved these changes Oct 16, 2024

View reviewed changes

njzjz reviewed Oct 16, 2024

View reviewed changes

njzjz reviewed Oct 17, 2024

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(pt): support CPU parallel training with PT #4224

feat(pt): support CPU parallel training with PT #4224

iProzd commented Oct 16, 2024 •

edited by coderabbitai bot

Loading

coderabbitai bot commented Oct 16, 2024 •

edited

Loading

Walkthrough

Changes

Assessment against linked issues

Possibly related PRs

Suggested reviewers

Chat

CodeRabbit Commands (Invoked using PR comments)

Other keywords and placeholders

CodeRabbit Configuration File (`.coderabbit.yaml`)

Documentation and Community

coderabbitai bot left a comment

codecov bot commented Oct 16, 2024

njzjz Oct 16, 2024

njzjz Oct 17, 2024

feat(pt): support CPU parallel training with PT #4224

Are you sure you want to change the base?

feat(pt): support CPU parallel training with PT #4224

Conversation

iProzd commented Oct 16, 2024 • edited by coderabbitai bot Loading

Summary by CodeRabbit

coderabbitai bot commented Oct 16, 2024 • edited Loading

Walkthrough

Changes

Assessment against linked issues

Possibly related PRs

Suggested reviewers

Chat

CodeRabbit Commands (Invoked using PR comments)

Other keywords and placeholders

CodeRabbit Configuration File (.coderabbit.yaml)

Documentation and Community

coderabbitai bot left a comment

Choose a reason for hiding this comment

codecov bot commented Oct 16, 2024

Codecov Report

njzjz Oct 16, 2024

Choose a reason for hiding this comment

njzjz Oct 17, 2024

Choose a reason for hiding this comment

iProzd commented Oct 16, 2024 •

edited by coderabbitai bot

Loading

coderabbitai bot commented Oct 16, 2024 •

edited

Loading

CodeRabbit Configuration File (`.coderabbit.yaml`)