vq-vae-2-pytorch
vq-vae-2-pytorch copied to clipboard
rosinality
Implementation of Generating Diverse High-Fidelity Images with VQ-VAE-2 in PyTorch
Hi, I am training the PixelSNAIL network with 128x128 image sizes and currently having the runtime error below: ``` background = self.background[:, :, :height, :].expand(batch, 2, height, width) RuntimeError: The...
I noticed in the `ResBlock` class of vqvae.py that you put a ReLU activation at the start of the residual stack, and no ReLU at the end: ```python class ResBlock(nn.Module):...
@rosinality Hi. I am looking for a more compact representation, so I tried to reduce the latent map size by half, i.e top one now corresponds to 16x16 and the...
@rosinality I am wondering if we could generate the results in section 4.4. We should create sequential context from latent frames, so we need a scheme to process latent frames...
@rosinality I have a doubt related to the image format used for training the model. As FFHQ is a collection of high dimensional PNG images, does training with LFW dataset,...
Dear Rosinality, Thank you for sharing with us this nice implement. I find the checkpoint for vq-vqe in the mani page. Would you mind share with us the checkpoint for...
How to get the dataset that was used to train the examples? What's it's format? (Like the folder hierarchy etc, do they use JPEG or PNG, what are the images...
@rosinality Is it that there is single codebook for all latent levels, or is it separate for each level ? Also what will be the affect of reducing the number...
I am using 4x Nvidia V100 and I am not able to get a batch size larger than 32 for the hyperparameters of this paper for training on the top...
Metadata
Owner
Metadata
Implementation of Generating Diverse High-Fidelity Images with VQ-VAE-2 in PyTorch