-
Hi, thanks for the code!
I have a question about pixel-cnn/model.py, line38. It's appending ones to the tensor x.
But I'm wondering how this is used -- it seems that after the "convolution(x_pad)"…
-
## 一言でいうと
RNNを使った自己回帰モデルの(hidden stateの)途中結果をキャッシュしておくことで計算の無駄な繰り返しを無くしWavenetを21倍、PixelCNN++を183倍早くした。
### 論文リンク
https://arxiv.org/abs/1704.06001
### 著者/所属機関
Prajit Ramachandran, Tom Le Pain…
-
Let's use this issue to talk about what we want to see in the first "clean" notebook.
EiffL updated
2 years ago
-
Is it possible to update the example for vqvae to also include how to use it on raw audio and/or video data?
Thank you
-
First of all, thank you very much for your help. When I ran the mnist_trainer.py you provided, there was a problem beyond the memory of the video, at which point my GPU was 8G. The problem is still on…
-
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 codes. I have also changed the loss to discretized mixtures…
-
This *may* not be the best framework for it, but it would be a good exercise to see how it looks compared to competing frameworks, e.g. Keras, Edward.
-
From here: https://groups.google.com/a/tensorflow.org/g/tfprobability/c/_9EuU3FnErY/m/cpNQENy6AQAJ
> 1) Re: saving models: I think it'd be both advantageous and straightforward for us to implement …
-
Hi @ibab,
I'm a bit late to the wavenet paper implementation party, but I'm reading the paper and your code and trying to understand where dilated convolution kernels are present. Your ASCII diagram…
-
Hi, thanks for the great work!
I just noticed that your paper is actually a concurrent work with LDM (exactly the same conference publication!), just wondering what's the main difference between th…