ssl_bad_gan
ssl_bad_gan copied to clipboard
kimiyoung
Good Semi-Supervised Learning That Requires a Bad GAN
which model you used for training on MNIST. simple CNN, restnet18 etc ?
If I understand correctly, the encoder is there to implement the Variational Inference term for enhancing entropy of the feature space. The latent z variable is represented by diagonal Gaussian....
Hi what's the updated url for http://kimi.ml.cmu.edu/mnist.True.3.best.pixel. This one is dead.
I cloned the repo and made the modifications to run with python 3.6. When running mnist_train.py, I get this error RuntimeError: The expanded size of the tensor (3) must match...
Can not open http://kimi.ml.cmu.edu/mnist.True.3.best.pixel HTTP ERROR 502 Could anybody send a copy to me? Thank you [email protected]
I run the code cifar_trainer.py . No changes to code, data and parameters. And the pytorch version is 0.1.12. But the result does not reach the 14.41 that you published...
Thank you for providing us with the code. I'm running the cifar_trainer.py The code, data, hyper parameter were taken as is and not modified. tail of cifar.FM+VI.run0(in the cifar10_log) looks...
Or alternatively is there another opensource code base you'd recommend for training the pixelCNN?
Line 136 in mnist_trainer.py, the operation (p_loss > 0).sum() may give unwanted result because p_loss > 0 returns a BytetTensor which can represent value from 0~255. However, the summation may...
HI! very thanksful for your code! i have a problem about run cifar_trainer.py when run in model.py line 165: " norm_weight = self.weight * (weight_scale[None,:,None,None] / torch.sqrt((self.weight ** 2).sum(3).sum(2).sum(0) +...
