ChrisJar
ChrisJar
Yup, so did I @rjzamora
Yep! I can tackle this
I'm still looking into the relational conversion error (seems to be an issue with the column name), but yeah I think I'm ready for a first round of reviews.
I haven't been able to create a minimal reproducer, but here's a traceback from using this PR to run: ```python import pandas as pd from dask_sql import Context df =...
@charlesbluca Thanks! This solved the issue
Ah I see. Is there any other way to square the elements of a SeriesGroupBy object?
For query 5 it appears that using [sklearn](https://scikit-learn.org/stable/modules/classes.html#module-sklearn.linear_model) as a direct replacement for cuml is slightly faster than adjusting to use [dask-ml](https://ml.dask.org/modules/generated/dask_ml.linear_model.LogisticRegression.html): | Run | Sklearn | Dask-ml | |-----|-------------|-------------|...
Those changes are actually not needed for this PR at all. I ended up taking a different path.
> Should we close this as #235 got merged in. This PR specifically addresses the DataFrame versions of some of the queries
@drobison00 Thanks for the reviews! I have a couple questions: How do you think we should go about preloading the vocab files in the case that the user doesn't want...