RuntimeError: DataLoader worker (pid(s) 94080) exited unexpectedly

[jiandandan001]

Thank you for sharing the code.

I run the deblocking code and have the following problem. Could you know the reason? Thanks.

File "G:\Anaconda\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.

Traceback (most recent call last): File "G:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 779, in _try_get_data data = self._data_queue.get(timeout=timeout) File "G:\Anaconda\lib\queue.py", line 178, in get raise Empty _queue.Empty

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "test.py", line 53, in for data in test_loader: File "G:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 363, in next data = self._next_data() File "G:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 974, in _next_data idx, data = self._get_data() File "G:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 931, in _get_data success, data = self._try_get_data() File "G:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 792, in _try_get_data raise RuntimeError('DataLoader worker (pid(s) {}) exited unexpectedly'.format(pids_str)) RuntimeError: DataLoader worker (pid(s) 94080) exited unexpectedly

[jiandandan001]

I re-run this code successfully with the Linux system. However, i have another question. For image deblocking, I use the official model provided by you. I found that the results at QF =10 are not very good, in terms of bot h subjective and objective qualities. The results at QF = 20/30/40 are reasonable. Could you help me check the model, or could you share me the deblocked images of classic5 and LIVE1? Thanks.

[jimyug]

Hi! @jiandandan001, We've checked again and are sure the shared model is fine, with which I've again successfully run the codes on the shared dataset (Google Cloud Disk) and got the same results as the ones proposed in the paper (29.36 dB at q=10).