memory & CPU eating about ydata-profiling

[JosPolfliet]

Have you tried using the pool_size keyword argument in the describe function? If you put that to 1, only 1 thread should be used. By default, there is one worker thread for every cpu.

Relevant code:

def describe(df, bins=10, correlation_overrides=None, pool_size=multiprocessing.cpu_count(), **kwargs):
pool = multiprocessing.Pool(pool_size)