Stuck in 'p.join()'

[RickWangww]

Program stuck in p.join()(77 line, in 'main.py'). More specifically, it will stuck when use 'model' in 'test.py' and 'train.py', such as 'model.load_state_dict(shared_model.state_dict())'( line 35, in 'test.py').
Actually, I tried to run the test process and train processes separately instead of using multiprocessing and it worked. But, obviously, the multiprocessing is a better choice. Has anyone encountered and solved this problem?

[hudsontek]

I came across the same issue. I guess you are also using Linux? Seems to be an inherent bug of pytorch. I have already filed a report(see here). Hope they can get it fixed soon.