Validation of channel keys for time series annotation

[pmayostendorp]

Is your feature request related to a problem? Please describe. When annotating time series from .csv files, the keys entered into the annotation template must match the csv headers exactly. Literally a single misspelled value in the column tag for the Channel element results in the task failing to properly load data. A blank screen like this will result (I'm using v1.13.1 here, but this is an issue in previous versions as well):

Even the console errors provided are not particularly useful. They are mostly red herrings that resulted in lots of tail chasing for me. For all I know, some other bug reports related to time series could be the result of a simple misspelled key. This issue comes to mind.

Describe the solution you'd like When opening a time series task, there should be a validation step to ensure that the csv headers requested are available in the time series data source. If they are not present, the front end should display an error indicating that the channels requested by the annotation template were not available in the data source. The user should be prompted to examine their data source and their annotation template to ensure they are compatible.

Describe alternatives you've considered Probably some additional documentation around setting up time series templates and ensuring that the user understands that just because the template looks reasonable on the labeling interface settings config page does not mean that data will correctly load in a task.

[heidi-humansignal]

Hi there! Thank you for your detailed feedback regarding the validation of channel keys for time series annotation in Label Studio. We understand that when annotating time series data from CSV files, a single mismatch between the column names in your labeling configuration and the CSV headers can result in tasks failing to load, displaying a blank screen without helpful error messages. This can make it difficult to identify and resolve the issue. We agree that implementing a validation step to ensure the column keys specified in the Channel elements match the CSV headers would greatly improve the user experience. Clear error messages indicating missing or mismatched channels would help users quickly diagnose and fix configuration issues. We've received similar feedback from other users who faced challenges due to minor discrepancies in the labeling configuration or data formatting. For example, in GitHub Issue #3094, users experienced tasks failing to load due to issues with time series data and found the error messages unhelpful. We're actively working on improving Label Studio's validation and error reporting mechanisms. I've forwarded your suggestion to our product team to consider adding this feature in a future release. While I can't provide a specific timeline, please know that we value your input and strive to enhance the usability of Label Studio based on user feedback. In the meantime, please ensure that the column names in your labeling configuration exactly match the CSV headers, including case sensitivity and any whitespace. Double-checking this can help prevent tasks from failing to load due to mismatches. If you have any further questions or need assistance with your labeling configuration, please don't hesitate to ask. We're here to help.

Thanks,

Tyler Conlee Head of Support HumanSignal

Comment by Tyler Conlee Workflow Run