problem run with train.py

[ahworld22]

Warning: multi_tensor_applier fused unscale kernel is unavailable, possibly because apex was installed without --cuda_ext --cpp_ext. Using Python fallback. Original ImportError was: ModuleNotFoundError("No module named 'amp_C'")
Epoch 0/119

Traceback (most recent call last):
File "", line 1, in
File "D:\Anaconda\envs\university\lib\multiprocessing\spawn.py", line 116, in spawn_main
exitcode = _main(fd, parent_sentinel)
File "D:\Anaconda\envs\university\lib\multiprocessing\spawn.py", line 125, in _main
prepare(preparation_data)
File "D:\Anaconda\envs\university\lib\multiprocessing\spawn.py", line 236, in prepare
_fixup_main_from_path(data['init_main_from_path'])
File "D:\Anaconda\envs\university\lib\multiprocessing\spawn.py", line 287, in _fixup_main_from_path
main_content = runpy.run_path(main_path,
File "D:\Anaconda\envs\university\lib\runpy.py", line 265, in run_path
return _run_module_code(code, init_globals, run_name,
File "D:\Anaconda\envs\university\lib\runpy.py", line 97, in _run_module_code
_run_code(code, mod_globals, init_globals,
File "D:\Anaconda\envs\university\lib\runpy.py", line 87, in _run_code
exec(code, run_globals)
File "C:\Users\hujia\Desktop\University1652-Baseline-master\University1652-Baseline-master\train.py", line 529, in
model = train_model(model, model_test, criterion, optimizer_ft, exp_lr_scheduler,
File "C:\Users\hujia\Desktop\University1652-Baseline-master\University1652-Baseline-master\train.py", line 237, in train_model
for data, data2, data3, data4 in zip(dataloaders['satellite'], dataloaders['street'], dataloaders['drone'],
File "D:\Anaconda\envs\university\lib\site-packages\torch\utils\data\dataloader.py", line 438, in iter
return self._get_iterator()
File "D:\Anaconda\envs\university\lib\site-packages\torch\utils\data\dataloader.py", line 384, in _get_iterator
return _MultiProcessingDataLoaderIter(self)
File "D:\Anaconda\envs\university\lib\site-packages\torch\utils\data\dataloader.py", line 1048, in init
w.start()
File "D:\Anaconda\envs\university\lib\multiprocessing\process.py", line 121, in start
self._popen = self._Popen(self)
File "D:\Anaconda\envs\university\lib\multiprocessing\context.py", line 224, in _Popen
return _default_context.get_context().Process._Popen(process_obj)
File "D:\Anaconda\envs\university\lib\multiprocessing\context.py", line 327, in _Popen
return Popen(process_obj)
File "D:\Anaconda\envs\university\lib\multiprocessing\popen_spawn_win32.py", line 45, in init
prep_data = spawn.get_preparation_data(process_obj._name)
File "D:\Anaconda\envs\university\lib\multiprocessing\spawn.py", line 154, in get_preparation_data
_check_not_importing_main()
File "D:\Anaconda\envs\university\lib\multiprocessing\spawn.py", line 134, in _check_not_importing_main
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.

[layumi]

Hi @ahworld22
Please reinstall apex with --cuda_ext --cpp_ext.
Or skip apex installation.

[ahworld22]

Hi @ahworld22 Please reinstall apex with --cuda_ext --cpp_ext. Or skip apex installation.
Yea!But when i run as python setup.py install ,it done.When i run as python setup.py install --cuda_ext --cpp_ext,it always fails and says it cann't find the csrc/flatten_unflatten.cpp which is correctly in my files.

[ahworld22]

It always says:
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.

I think it's the problem with windows,it need the "main" to run multiprocess

[ahworld22]

I use Windows to run the project ,but it seems something wrong.Because the train.py doesn't have 'main',which cann't do on Windows with multiprocess. At 2024-09-12 00:56:58, "Zhedong Zheng" ***@***.***> wrote: Hi @ahworld22 Please reinstall apex with --cuda_ext --cpp_ext. Or skip apex installation. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>