Context

we want to

da.apply(lambda x : x, num_worker=4 )

This will be different from doing

for doc in da:
    doc = f(doc)

map(f,a)

Is that we will leverage multiprocessing. (Need to be extra careful and add good test to check that nothing is going wrong)

• benchmarking

We need to choose between Map or Apply but we need only of them.