Hell yeah, Cleveland!

I’m not aware of an Arrow or Parquet format, but there is the ONNX format (see onnx.ai). Depending on the model, you could try Posit’s orbital project, which translates your model to SQL (see here orbital.tidymodels.org).

Hella exciting! Great work, dude.

Hell yeah! Good stuff.

Working with data where columns contain nested data structures is easier in R (e.g. JSON). Packages like {purrr} and {dplyr} let you to manipulate columns of arbitrary types using the same APIs. The NSE in R is also very helpful here.

I also think {tidymodels} has the best UX for ML work.

Hell yeah! Do it. Try getting a friend involved that can hold you accountable. My friend and I are aiming to have something written every month.

Hero.

You have a handful of options to try. The package {duckdb}’s already been mentioned, and is good. There’s a similar package {duckplyr} that also works well. The {arrow} package is also great for reading in large amounts of CSVs.

Sick find. Thanks, G!