Hey guys, I see that there's a

df.sample()

api to get a fractional sample of the dataframe. Is there an API that I can pass an integer to (ie. number of rows)? I just want to materialize n-random-rows as a daft dataframe, not show them in a notebook.

Hey @Neil Wadhvana, currrently theres no built in method to do this. If you're alright to sacrifice some efficiency, you can do a

df.count_rows()

and then sample with

df.sample(fraction=desired_num_rows/num_total_rows

. Also, I believe @Desmond Cheong is working on implementing sampling by

max_rows

,https://github.com/Eventual-Inc/Daft/issues/3332 .

That's exactly what I'm doing right now! Cool, thanks for the heads up!