Kevin Kho

Hey @lukeb88, Thanks for the well-written issue. You've obviously spent a lot of time learning FugueSQL. I can answer the first question immediately while I look into the second. First,...

For the second question, there is something we should fix. It is a bug for the behavior to be inconsistent across execution engines, but I can explain why this happens...

One last thing I forgot to mention is that the function ```python # schema: *, col3:int def make_new_col(df: pd.DataFrame) -> pd.DataFrame: '''''' df['col3'] = df['col1'] + df['col2'] return df ```...

@lukeb88 , that is very well articulated. They shouldn't be using Pandas for larger datasets ideally. Thanks for that! And your example of using `SELECT` inline is very elegant!

Thank you for the suggestion @aripollak . There are indeed multiple pages where the same thing is mentioned. To answer you quickly though, 1. Results is a subset of persistence....

Hi @vspinu , it would require quite a bit of refactoring and would actually be a breaking change for many users as it breaks the core interface (will explain). The...

This got done. We are on conda-forge: https://anaconda.org/conda-forge/fugue

Closing due to staleness doc changes. We can revive the issue if typos are found here: https://nixtla.github.io/statsforecast/docs/how-to-guides/autoarima_vs_prophet.html

Connected to https://github.com/Nixtla/statsforecast/issues/520

I was just looking into stuff and it looks like the current Numba Ahead-of-Time compilation will be deprecated. See [here](https://numba.readthedocs.io/en/stable/user/pycc.html#compiling-code-ahead-of-time). There will be something new this year I guess. I...