Justin Zhao

@tgaddair there's a chance that #2455 may help, but it looks like @Jeffwan is getting the same super large number of training steps even when he tried setting `batch_size=128` manually....

@tgaddair Ah, thanks for the catch! We should be sure to include this in our documentation.

As part of this change, I've made a few changes to the `OneOf`, requiring that at most one field has `allow_none=True`. ```python if default is None: # If the default...

Thanks @js-dieu! Congrats on your new laptop. Hopefully it's not a complicated fix.

@js-dieu Here's my output: ``` >>> import memory_profiler as mp import psutil as ps print(ps.__version__) print(mp.__version__) >>> import psutil as ps >>> >>> print(ps.__version__) 5.9.1 >>> print(mp.__version__) 0.60.0 ```

Hi @jiangweiatgithub, we just released v0.6, which addressed a handful of issues with Ray. Would you be able to try upgrading the version of Ludwig to see if the ray...

Thanks @abaletinskikh for posting the issue and for posting a workaround. Can confirm that version compatibility is particularly important for torch libraries, and indeed the latest compatible version of torchtext...

Hi @dragosmc, We had to pin pydantic to

The regression tests are flaky, however it looks like there's another dependency issue going on with torch nightly.

Hi @sergsb, Thanks for sharing your experience. The Ludwig team is focused on building first class support for natively supported models on HF. As I understand, supporting models that require...