[Python] How to give dataframe inside isin()

I want convert a sql query like

SELECT * FROM df WHERE id IN (SELECT id FROM an_df)


into dask equivalent. So, I am trying this:

d=df[df['id'].isin(an_df['id'])]


But it is thorwing NotImplementedError: passing a 'dask.datframe.core.DataFrame' to'isin'

Then I converted this an_df['id'] to list like

d=df[df['id'].isin(list(an_df['id']))] or d=df[df['id'].isin(an_df['id'].compute())]


but this is very time consuming.

I want a solution that works as fast as dask. df has approximately 100 million rows.

Continue reading...