Python 中的多处理使用 Pool

Question

我想使用 Python 在数据帧上并行化一个函数。我看了教程，发现一些 code.I 已经根据我的需要进行了调整。当我执行地图功能时，程序冻结。代码似乎很可靠。请问是什么问题

import pandas as pd
import numpy as np
from multiprocessing import cpu_count, Pool

attributes1 = pd.read_csv('attributes1.csv')


def replace_data(data):
    for i in range(0, len(data.index)):
        temp = data.iloc[i, 1]
        temp = temp.replace('in.', 'inch')
        data.iloc[i, 1] = temp
    return data

num_partitions = 10 #number of partitions to split dataframe
num_cores = cpu_count() #number of cores on your machine

def parallelize_dataframe(df, func):
    df_split = np.array_split(df, num_partitions)
    pool = Pool(num_cores)
    df = pd.concat(pool.map(func, df_split))
    pool.close()
    pool.join()
    return df

df1 = parallelize_dataframe(attributes1, replace_data)

Answer 1

您可能正在使用 windows，要修复它您需要 freeze_support:

from multiprocessing cpu_count, Pool, freeze_support

...

if __name__ == '__main__':
    freeze_support()
    df1 = parallelize_dataframe(attributes1, replace_data)

Answer 2

这只是 Windows 用户的问题。首先，我创建了另一个 .py 文件，我们将其命名为 helpy.py，其中我有 replace_data 函数

def replace_data(data):
    for i in range(0, len(data.index)):
        temp = data.iloc[i, 1]
        temp = temp.replace('in.', 'inch')
        data.iloc[i, 1] = temp
    return data

然后我将我的函数导入到我的主 .py 文件中。

import pandas as pd
import numpy as np
from multiprocessing import cpu_count, Pool
from helpy import replace_data

attributes1 = pd.read_csv('attributes1.csv')


num_partitions = 10 #number of partitions to split dataframe
num_cores = cpu_count() #number of cores on your machine

    def parallelize_dataframe(df, func):
        df_split = np.array_split(df, num_partitions)
        pool = Pool(num_cores)
        df = pd.concat(pool.map(func, df_split))
        pool.close()
        pool.join()
        return df

    if __name__ == '__main__':
        df1 = parallelize_dataframe(attributes1, replace_data)

我也加了if __name__ == '__main__':现在程序运行流畅

Python 中的多处理使用 Pool

Multiprocessing in Python using Pool

python

dataframe

parallel-processing

spyder

pool