R Devon Hjelm
R Devon Hjelm
Ah, so you want to train a classifier from scratch, for example?
Which original GAN do you speak of? The penalty that I use comes from Kevin Roth's paper, which works much better than WGAN-GP penalty in my experience. In Kevin's paper,...
`libboost-test-dev:amd64/xenial 1.58.0.1ubuntu1 uptodate` `libboost-all-dev` was already installed... (1.58.0.1ubuntu1) Removing and reinstalling `libboost-all-dev` sadly does not work. Could this be an `amd64` vs `i386` problem?
Can you fix the conflicts? I'm at a point of building the RNN demo for the new framework.
@dmitriy-serdyuk Can you rebase this to dev? I think there might be some overlapping in the branches
@dmitriy-serdyuk OK good. I'll fix the flake errors with dev and start merging.
It's implemented! iirc it should be runnable with `cortex mine --help`, but I've been working mostly on the next release of cortex so I don't know for sure.
So the implementation as-is is set up for image data and gaussian noise. So while there are no constraints about the dimensionality of either variables, the networks used are suitable...
I fixed an unrelated bug when running that script: https://github.com/rdevon/cortex/commit/1f566bc73e4ff9453a88a3268e5b1f7606ceeedc but I was unable to replicate your error. I noticed that you're using Theano-0.9.0dev0. How did you install Theano?
So you can run rbm_mnist but not rbm_ni? That's strange: the code is almost the same. I cannot reproduce your error, but I have a hunch what's going on. I've...