parallel_apply.py

parallel_apply.py

from functools import partial
import pandas as pd
import numpy as np

def apply(df, func):
    return df.apply(func, axis=1)

def parallel_apply(df, func, n_cores):
    df_split = np.array_split(df, n_cores)
    pool = Pool(n_cores)
    series = pd.concat(pool.map(partial(apply, func=func), df_split))
    pool.close()
    pool.join()
    return series