python 多进程pool.apply_async 在报错 AttributeError: Can't get attribute 'task' on __main__' (built-in)>

在用pycharm跑多进程的时候，不知道为什么总是报一个错误：

Process SpawnPoolWorker-21:
Traceback (most recent call last):
  File "D:\Program Files\Anaconda3\lib\multiprocessing\process.py", line 249, in _bootstrap
    self.run()
  File "D:\Program Files\Anaconda3\lib\multiprocessing\process.py", line 93, in run
    self._target(*self._args, **self._kwargs)
  File "D:\Program Files\Anaconda3\lib\multiprocessing\pool.py", line 108, in worker
    task = get()
  File "D:\Program Files\Anaconda3\lib\multiprocessing\queues.py", line 345, in get
    return ForkingPickler.loads(res)
AttributeError: Can't get attribute 'task' on <module '__main__' (built-in)>

定义的多进程任务函数及相关如下

import pandas as pd
import os
from multiprocessing import Queue, Pool

def task(path):
    # print(path)
    df = pd.read_csv(path, header=None)

    extreme = df[1][(df[27]) == 0].unique().tolist()  # 排除涨跌停的股票
    df = df[[x not in extreme for x in df[1]]]

    df.to_csv(path, header=None, index=None)
    print(path)

path = r'E:\Data\citics\201805'
p = Pool(20)
for file in os.listdir(path):
    print(file)
    filePath = os.path.join(path, file)
    p.apply_async(task, args=(filePath,))

p.close()
p.join()
print('end')

---

后来在网上搜到了相似的问题，见 https://blog.csdn.net/geekleee/article/details/77838288 ， 但是感觉解决方案有点繁琐，所以就自己摸索了下，最后发现如果将多进程代码放在 脚本主函数下： if _name_ == ‘_main_‘:
在运行整个脚本就不会报错！

import pandas as pd
import os
from multiprocessing import Queue, Pool

def task(path):
    # print(path)
    df = pd.read_csv(path, header=None)

    extreme = df[1][(df[27]) == 0].unique().tolist()  # 排除涨跌停的股票
    df = df[[x not in extreme for x in df[1]]]

    df.to_csv(path, header=None, index=None)
    print(path)

if __name__ == '__main__':

    path = r'E:\Data\citics\201805'
    p = Pool(20)
    for file in os.listdir(path):
        print(file)
        filePath = os.path.join(path, file)
        p.apply_async(task, args=(filePath,))

    p.close()
    p.join()
    print('end')