The below code seems to have some issues. The aim would be to append each result of new_df() to some list, e.g. out.

import pandas as pd
import random
import time
from multiprocessing import Pool

def new_df(rows=10000):  # proxy for complex dataframe
    temp = pd.DataFrame({'a': [''.join(chr(random.randint(65,122)) for _ in range(200))
                               for _ in range(rows)]})
    temp['b'] = temp['a'].str.lower()
    temp['c'] = temp['a'].str.upper()
    return temp

pool = Pool(4)
start = time.time()
out = pool.map(new_df, [9999,10000,10001,10002])
print(f"{time.time() - now} sec")

Issues - VisualStudioCode

    raise RuntimeError('''
RuntimeError:
        An attempt has been made to start a new process before the
        current process has finished its bootstrapping phase.

        This probably means that you are not using fork to start your
        child processes and you have forgotten to use the proper idiom
        in the main module:

            if __name__ == '__main__':
                freeze_support()
                ...

        The "freeze_support()" line can be omitted if the program
        is not going to be frozen to produce an executable.

Traceback

Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "C:\Mambaforge\lib\multiprocessing\spawn.py", line 116, in spawn_main
    exitcode = _main(fd, parent_sentinel)
  File "C:\Mambaforge\lib\multiprocessing\spawn.py", line 125, in _main
    prepare(preparation_data)
  File "C:\Mambaforge\lib\multiprocessing\spawn.py", line 236, in prepare
    _fixup_main_from_path(data['init_main_from_path'])
  File "C:\Mambaforge\lib\multiprocessing\spawn.py", line 287, in _fixup_main_from_path
    main_content = runpy.run_path(main_path,
  File "C:\Mambaforge\lib\runpy.py", line 268, in run_path
    return _run_module_code(code, init_globals, run_name,
  File "C:\Mambaforge\lib\runpy.py", line 97, in _run_module_code
    _run_code(code, mod_globals, init_globals,
  File "C:\Mambaforge\lib\runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "c:\Users\XXX\untitled0.py", line 13, in <module>
    pool = Pool(4)
  File "C:\Mambaforge\lib\multiprocessing\context.py", line 119, in Pool
    return Pool(processes, initializer, initargs, maxtasksperchild,
  File "C:\Mambaforge\lib\multiprocessing\pool.py", line 212, in __init__
    self._repopulate_pool()
  File "C:\Mambaforge\lib\multiprocessing\pool.py", line 303, in _repopulate_pool
    return self._repopulate_pool_static(self._ctx, self.Process,
  File "C:\Mambaforge\lib\multiprocessing\pool.py", line 326, in _repopulate_pool_static
    w.start()
  File "C:\Mambaforge\lib\multiprocessing\process.py", line 121, in start
    self._popen = self._Popen(self)
  File "C:\Mambaforge\lib\multiprocessing\context.py", line 327, in _Popen
    return Popen(process_obj)
  File "C:\Mambaforge\lib\multiprocessing\popen_spawn_win32.py", line 45, in __init__
    prep_data = spawn.get_preparation_data(process_obj._name)
  File "C:\Mambaforge\lib\multiprocessing\spawn.py", line 154, in get_preparation_data
    _check_not_importing_main()
  File "C:\Mambaforge\lib\multiprocessing\spawn.py", line 134, in _check_not_importing_main

CodePudding user response：

Code reconstructed to utilise the main module idiom:

import pandas as pd
import random
import time
from multiprocessing import Pool

def new_df(rows=10000):
    temp = pd.DataFrame({'a': [''.join(chr(random.randint(65,122)) for _ in range(200))
                               for _ in range(rows)]})
    temp['b'] = temp['a'].str.lower()
    temp['c'] = temp['a'].str.upper()
    return temp

def main():
    start = time.perf_counter()
    with Pool(4) as pool:
        pool.map(new_df, [9999, 10000, 10001, 10002])
    print(f"{time.perf_counter() - start:.2f}s")

if __name__ == '__main__':
    main()

Output:

1.24s