Joppe Geluykens

Will we be replacing dictionary accesses with the schema class in a follow up PR?

Can this be closed now that #2370 is merged? @connor-mccorm @dantreiman

+1

Hi @camaya7, thanks for trying out time series with Ludwig! I tried reproducing your issue using the following code, but did get the expected output: ```python from io import StringIO...

Glad to hear that you have clear steps forward @camaya7! I will close this issue, but feel free to open a new one or [start a discussion](https://github.com/ludwig-ai/ludwig/discussions) if you have...

Stratified split is due to the highly imbalanced dataset. Good call on the simplification, will add!

EDIT: Looks like the LudwigProfiler is not serializable because of the thread that is started when entering its context. Debugging serializability: ```python >>> from ray.util import inspect_serializability >>> from ludwig.benchmarking.profiler_callbacks...

Maybe something like this? 

This is awesome, thanks @krfricke ! Is there any chance of getting a PyPi release for 2.1 as well?

@yimingsun12138 did you find a solution for this?