Reproducibility issue with CuDNN backend

[songyouwei]

mt-dnn/data_utils/utils.py

Line 25 in 25c3bd1

def set_environment(seed, set_cuda=False):

When running on the CuDNN backend, there might be reproducibility issue.
According to pytorch document, two further options must be set:

torch.backends.cudnn.deterministic = True
torch.backends.cudnn.benchmark = False

[namisan]

Thanks pointing it out!!
Does this work on pytorch version 0.4x?
We will test it soon.

[songyouwei]

I haven't tested it on pytorch version 0.4x.