fix: Incorrect URI Prefix Stripping in MLFlowLogger

[awindmann]

What does this PR do?

Addresses the issue described in issue 20279. Essentially, str.lstrip() is used incorrectly, causing the MLFlowLogger to log in the wrong location under certain conditions. This can be fixed by stripping the uri prefix with another function. As str.removeprefix() was only introduced with Python 3.9.0, this pull request uses a more general method to strip the string, which should be okay because the method is contained by the if statement.

Background

The save_dir property in the MLFlowLogger class currently uses str.lstrip() to remove the LOCAL_FILE_URI_PREFIX ("file:") from the tracking URI. This method does not strip the prefix, but all characters that match those in the prefix ("f", "i", "l", "e", ":"). Thus, setting save_dir="logs" would be incorrectly modified to "ogs", leading to the creation of an "ogs" folder where logging subsequently occurs.

Minimal Example

from pytorch_lightning.loggers import MLFlowLogger

logger = MLFlowLogger(
    save_dir='logs'
)
print(logger.save_dir)  # ogs

Fixes #20279

Before submitting

• Was this discussed/agreed via a GitHub issue? (not for typos and docs)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure your PR does only one thing, instead of bundling different changes together?
• Did you make sure to update the documentation with your changes? (if necessary)
• Did you write any new necessary tests? (not for typos and docs)
• Did you verify new and existing tests pass locally with your changes?
• Did you list all the breaking changes introduced by this pull request?
• Did you update the CHANGELOG? (not for typos, docs, test updates, or minor internal changes/refactors)

PR review

Anyone in the community is welcome to review the PR.
Before you start reviewing, make sure you have read the review guidelines. In short, see the following bullet-list:

Reviewer checklist

• Is this pull request ready for review? (if not, please submit in draft mode)
• Check that all items from Before submitting are resolved
• Make sure the title is self-explanatory and the description concisely explains the PR
• Add labels and milestones (and optionally projects) to the PR so it can be classified

---

📚 Documentation preview 📚: https://pytorch-lightning--20365.org.readthedocs.build/en/20365/

[lantiga]

Good catch!