Skip to content

ci: Install CI dependencies from uv lock#15697

Open
chtruong814 wants to merge 26 commits into
mainfrom
chtruong/uv-lock
Open

ci: Install CI dependencies from uv lock#15697
chtruong814 wants to merge 26 commits into
mainfrom
chtruong/uv-lock

Conversation

@chtruong814
Copy link
Copy Markdown
Collaborator

Summary

  • Replace the CI Docker base with CUDA 13.1.2 and install dependencies with uv from pyproject.toml/uv.lock
  • Remove TE/Megatron-Core wheel stages and install_dep.sh usage from docker/Dockerfile.ci
  • Force Linux torch resolution to torch 2.10.0+cu130 and keep nemo_automodel pinned to the requirements manifest commit

Testing

  • UV_CACHE_DIR=/private/tmp/nemo-uv-cache /private/tmp/nemo-uv-latest/bin/uv lock --check
  • DOCKER_BUILDKIT=1 /opt/homebrew/bin/docker-buildx build --progress=plain -f docker/Dockerfile.ci -t nemo-ci-uv-test --load . (uv install completed; local image load failed due Docker disk space)

Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814 chtruong814 changed the title Install CI dependencies from uv lock ci: Install CI dependencies from uv lock May 13, 2026
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Copy link
Copy Markdown
Collaborator Author

/ok to test 611aa64

Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Copy link
Copy Markdown
Collaborator Author

/ok to test fd0cce2

Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
chtruong814 and others added 2 commits May 21, 2026 22:24
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Piotr Żelasko <pzelasko@nvidia.com>
@copy-pr-bot
Copy link
Copy Markdown

copy-pr-bot Bot commented May 27, 2026

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

pzelasko added 2 commits May 27, 2026 07:56
Signed-off-by: Piotr Żelasko <pzelasko@nvidia.com>
Signed-off-by: Piotr Żelasko <pzelasko@nvidia.com>
@pzelasko
Copy link
Copy Markdown
Collaborator

/ok to test 88de033

@copy-pr-bot
Copy link
Copy Markdown

copy-pr-bot Bot commented May 27, 2026

/ok to test 88de033

@pzelasko, there was an error processing your request: E2

See the following link for more information: https://docs.gha-runners.nvidia.com/cpr/e/2/

@pzelasko
Copy link
Copy Markdown
Collaborator

/ok to test fa5e5f9

Comment thread nemo/collections/asr/data/feature_to_label.py Dismissed
@pzelasko
Copy link
Copy Markdown
Collaborator

/ok to test 074a90a

Comment thread pyproject.toml Outdated
],
]
override-dependencies = [
"torch==2.12.0+cu130 ; sys_platform == 'linux'",
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can we use torch==2.12.0+cu132 since the base container is now with CUDA 13.2?

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done + refreshed uv.lock

@pzelasko
Copy link
Copy Markdown
Collaborator

/ok to test eca699c

@pzelasko
Copy link
Copy Markdown
Collaborator

/ok to test 1eca063

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants