Check for "/" in feature names #660
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
aliberts
approved these changes
Jan 29, 2025
menhguin
pushed a commit
to menhguin/lerobot
that referenced
this pull request
Feb 9, 2025
JIy3AHKO
pushed a commit
to vertix/lerobot
that referenced
this pull request
Feb 27, 2025
Ke-Wang1017
pushed a commit
to Ke-Wang1017/lerobot
that referenced
this pull request
Mar 26, 2025
Kalcy-U
referenced
this pull request
in Kalcy-U/lerobot
May 13, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What this does
If you create a dataset with feature keys that contain a '/', this will work fine and you can store the dataset. However, the serialized dataset stats will be corrupt because they are flattened using a '/' as separator. Upon training, you will get 'keyErrors' as the original feature names are now no longer available in the dataset stats file.
This is rather annoying to debug when you start training, and at that point you have to recreate your datasets.
This PR adds a simple check to guard against creating a dataset with features that contain a '/' in their key, to make sure this issues is caught upon creation of the dataset.
How it was tested
Explain/show how you tested your changes.
added a test in
tests/test_datasetto ensure that dataset creation fails when the feature keys contain '/'tagging @aliberts and @Cadene since their names are linked to a related #TODO in the codebase.