Summing columns that are pyarrow StringScalar types....

[jayceslesar]

Hey all, I convert a large csv (~50ish gigs usually but will get bigger) to HDF5 and have Vaex churn through it for a lot of stats and analytics, and am wondering the best way to sum column that is of StringScalar type? How I do it now is
amount = int(np.sum(np.asarray(current_sig_vals.values, dtype=int))) , and I am wondering if Vaex has a solution to deal with the pyarrow StringScalars (they are all "0" or "1" by the way).

current_sig_vals is the Vaex expression that filters down my Vaex.dataframe to the data I want to look at.

Would I want to convert it to an df.sum([column]).astype("int8") or something similar?