Davide Abati

1) range of CIFAR-10 images: [0-1] 2) decoder activation function: None 3) range of the reconstruction: approximately [0-1]

I saw you are using 128x128 images. The number of downsampling in the model is tuned on 32x32 images. You might then have a huge linear layer before the bottleneck...

Keep in mind that, if I am right, the best way to lighten up the model would be to **add** downsample and upsample blocks.

Hi, I think you are not using the correct collate function in the `DataLoader`. One quick way to patch this is to reshape the input tensor before feeding the network:...

I used a single 12GB GPU. I'd say you could reduce the batch size and give it a try. D

Hi, If I remember correctly, we ran a training on moving MNIST as a safety check. We managed to forecast future frames successfully but we didn't measure any metric. Also...

Hi, Are you providing the link to a folder or to a archive? I don't think the former will work.

Hi! Thanks for interest. Here's the thing: I started from the weights provided in keras by [this gist](https://gist.github.com/albertomontesg/d8b21a179c1e6cca0480ebdf292c34d2). Such weights were ported from the [original caffe repo](https://github.com/facebook/C3D). In the keras...

@BarryBA I just ran the prediction script (without removing the mean) with PyTorch 0.4.0 and it provides me the correct results. Really weird.

Hi and thanks for interest. I can share with you the snippet to save parameters from keras (v=1.2.2) to file: ``` # save weights import os.path import numpy as np...