Polars version
Opened this issue · 0 comments
Describe the bug
The new versions of polars >1.x have a different behavior in the .drop method compared to the older ones, like 0.20.x.
Check here:
https://docs.pola.rs/api/python/stable/reference/dataframe/api/polars.DataFrame.drop.html
To Reproduce
For example. check the response matching metric. When you try to drop the 'context' column, which does not exist necessarily exist, you do not want to get an error stating that column 'context' is not included in your input data.
Example: https://docs.uptrain.ai/predefined-evaluations/ground-truth-comparison/response-matching
Expected behavior
Proceed without getting error for optional columns.
Additional context
I just tackled that by freezing my polars version to 0.20.31, while the strict parameter of the new polars version seems to handle that issue, but needs to be added in each use of .drop method.