How to eliminate this warning？

[zhousaYolo]

WRN MultiProcessRunner does support Windows. However, Windows requires more strict picklability on processes, which may lead of failure on some of the code.

[610265158]

yes,

windows cannot run these code, for multiprocess reasons

[zhousaYolo]

Thanks, but I don't use mulprocess, and then I can train.
ds = DataFromGenerator(self.generator)
if cfg.DATA.mutiscale and self.training_flag:
ds = MutiScaleBatcher(ds, self.num_gpu * self.batch_size, scale_range=cfg.DATA.scales,
input_size=(cfg.DATA.hin, cfg.DATA.win))
else:
ds = MutiScaleBatcher(ds, self.num_gpu * self.batch_size, input_size=(cfg.DATA.hin, cfg.DATA.win))
# ds = MultiProcessPrefetchData(ds, self.prefetch_size, self.process_num)
# ds.reset_state()
ds = ds.get_data()
return ds

[610265158]

I thought if you can run the code, then just ignore the warning