Open DhanasekarChellamuthu opened 5 years ago
me to,have you solved this problem?
me to,do you have some methods to solved this problem?
In Icdar.py, modify to: enqueuer = GeneratorEnqueuer(generator(**kwargs), use_multiprocessing=False) enqueuer.start(max_queue_size=1, workers=1)
I changed the trainset path and remaining things mentioned in readme. when i am try to run multigpu_train code it is throwing an following error:
AttributeError: Can't pickle local object 'GeneratorEnqueuer.start..data_generator_task'