Skip to content

[DOC] Recommend uv and venv for developer setup - #2365

Merged
phoeenniixx merged 4 commits into
sktime:mainfrom
BLVCK-MAMBA-6:docs/2364-uv-venv
Oct 17, 2026
Merged

phoeenniixx merged 4 commits into
sktime:mainfrom
BLVCK-MAMBA-6:docs/2364-uv-venv

Conversation

@BLVCK-MAMBA-6

@BLVCK-MAMBA-6 BLVCK-MAMBA-6 commented Aug 1, 2026 •

Copy link
Copy Markdown
Contributor

Reference Issues/PRs

Fixes #2364.

What does this implement/fix? Explain your changes.

Updates the developer environment setup documentation to:

  • recommend uv as the primary environment manager;
  • document virtual environment creation and activation on Unix and Windows;
  • retain standard-library venv as a fallback;
  • document editable installation with the dev dependencies;
  • show how to install all optional dependencies when required.

The existing Conda package-installation guidance remains unchanged.

What should a reviewer concentrate their feedback on?

  • Whether the uv and venv commands are accurate across supported platforms.
  • Whether the distinction between the recommended workflow and fallback is clear.

Did you add any tests for the change?

No automated tests were added because this is a documentation-only change.

The updated RST document was parsed successfully with docutils, and git diff --check passed.

Any other comments?

None.

PR checklist

  • The PR title starts with either [ENH], [MNT], [DOC], or [BUG].
  • Added/modified tests — not applicable because this is documentation-only.
  • Used pre-commit hooks when committing.

@codecov

codecov Bot commented Aug 1, 2026 •

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (main@24a4d5c). Learn more about missing BASE report.

Additional details and impacted files
@@           Coverage Diff           @@
##             main    #2365   +/-   ##
=======================================
  Coverage        ?   88.81%           
=======================================
  Files           ?      218           
  Lines           ?    11670           
  Branches        ?        0           
=======================================
  Hits            ?    10365           
  Misses          ?     1305           
  Partials        ?        0           
Flag Coverage Δ
cpu 88.81% <ø> (?)
pytest 88.81% <ø> (?)

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.

@phoeenniixx phoeenniixx left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Thanks! This is great
I have a doubt though - should we completely remove conda installation instructions? We can recommend uv but it should not mean that we remove installation instructions for conda ?
@fkiraly What do you think?

@phoeenniixx phoeenniixx added documentation Improvements or additions to documentation ptf-v1 Related to `pytorch-forecasting` v1 ptf-v2 Related to `pytorch-forecasting` v2 labels Aug 3, 2026
@BLVCK-MAMBA-6

BLVCK-MAMBA-6 commented Aug 3, 2026 •

Copy link
Copy Markdown
Contributor Author

Yes, I agree that Conda should remain available as an installation option.

This PR does not remove the general Conda installation instructions, the Conda command is still present near the top of installation.rst. I only replaced the Conda-first workflow in the developer setup section, based on #2364.

My understanding was that uv should become the recommended environment manager for contributors, while Conda remains available as an alternative installation method. I’ll wait for @fkiraly’s view before making any further changes, and I can make this distinction more explicit in the documentation if needed.

@BLVCK-MAMBA-6

Copy link
Copy Markdown
Contributor Author

I’ve now made the distinction explicit in the documentation: Conda remains documented as an installation option for the published package, while uv is recommended specifically for setting up a contributor development environment.

@BLVCK-MAMBA-6

Copy link
Copy Markdown
Contributor Author

Hi @phoeenniixx, just following up on this. I clarified the Conda/uv distinction in 072d2ec: Conda remains available for package installation, while uv is recommended specifically for contributor development environments.

All checks are passing. When you have a chance, could you please take another look? Thanks!

@BLVCK-MAMBA-6

Copy link
Copy Markdown
Contributor Author

Hi @phoeenniixx, just circling back on this one!

I’ve updated the docs to clearly separate the workflows: uv is recommended specifically for contributor developer setups, while Conda remains fully documented as a package installation option.

All checks are passing and the branch is clean. Whenever you have a spare moment, I’d appreciate a re-review so we can get this merged. Thanks!

@phoeenniixx phoeenniixx left a comment •

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Thanks!
This is nice

(will merge once the tests pass)

@phoeenniixx

phoeenniixx commented Oct 16, 2026 •

Copy link
Copy Markdown
Member

And apologies for the late review.
There is a long queue of PRs, and old ones tend to get stalled because of this. Please feel free to tag core devs on discord in the future, if you dont get any review for 1-2 weeks.

Thanks again for the help!

@BLVCK-MAMBA-6

Copy link
Copy Markdown
Contributor Author

Thank you so much, @phoeenniixx! No worries at all about the delay, I appreciate the review and approval.

I'll definitely join the Discord server. As an AI/ML engineer, I'm really glad to dive deeper into pytorch-forecasting and continue contributing to the codebase.

Looking forward to working with everyone!

@BLVCK-MAMBA-6

Copy link
Copy Markdown
Contributor Author

Hi @phoeenniixx, it looks like the CodeQL and Test / code-quality workflow checks have been sitting queued for a bit. Could you please take a quick look or re-trigger the actions? Thank you

@phoeenniixx

Copy link
Copy Markdown
Member

Thanks for the PR @BLVCK-MAMBA-6!
The runners were busy with other PRs and this PR was in a queue, that is why it took some time to run the tests. Everything passed, so i will merge this.

@phoeenniixx
phoeenniixx merged commit dcfde09 into sktime:main Oct 17, 2026
38 checks passed
@BLVCK-MAMBA-6

Copy link
Copy Markdown
Contributor Author

Awesome! Thank you so much @phoeenniixx for merging this.

Glad to contribute.

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

Labels

documentation Improvements or additions to documentation ptf-v1 Related to `pytorch-forecasting` v1 ptf-v2 Related to `pytorch-forecasting` v2

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[DOC] replace conda as recommended env manager with uv and/or venv

2 participants