Get column value after searching for row in dask

Question

I have a pandas dataframe that I converted to a dask dataframe using the from_pandas function of dask. It has 3 columns namely col1, col2 and col3.

Now I am searching for a specific row using daskdf[(daskdf.col1 == v1) & (daskdf.col2 == v2)] where v1 and v2 are values that I am searching for. But when I try to fetch the value of col3 using daskdf[(daskdf.col1 == v1) & (daskdf.col2 == v2)]['col3'] it gives me a dask series structure and not the column value.

In pandas I could do pandasdf[(pandasdf.col1 == v1) & (pandasdf.col2 == v2)]['col3'].tolist(). How do I fetch the value of col3 here ?

Answer 1

Running .compute will convert a dask object into pandas object, so the following will give the required result:

result = daskdf[(daskdf.col1 == v1) & (daskdf.col2 == v2)]['col3'].compute().tolist()