Open domarps opened 6 years ago
I've not used Docker at all (well, I've not successfully used Docker at all), but I can try to point you in the right direction. I was running this with python 2.7.12, Theano 0.7 or 0.8, and Lasagne 0.2 back in 2016, on GTX 980s and my laptop GPU for inference (GT730m).
At the end of the day though, I'm not sure you'll be able to run it on the V100s, what with Theano being deprecated and this code being 2+ years old at this point.
If you just want to use the interface, it should be trivial to plug in a PyTorch/Chainer/TF pre-trained GAN, just implement the appropriate methods in API.py.
Hi,
I am trying to reproduce the code on a V100 instance and I ran into the following issues when I ran
python NPE.py
Do you have any recommendations on how we can reproduce your experimental setup in the form of a Dockerfile?