fab-jul

So let's say you have have the following distribution ``` P(100) = p1 P(200) = p2 P(300) = p3 ``` Now, you would transform this to a CDF that is:...

Hi, sorry to hear that. I assume this is because we rely on TF1 and I do not know if there is away to upgrade CUDA for TF1 :(

Hi Yifei, thanks for your interest in HiFiC. I will look into this this week. could you provide some details? It is my understanding that HiFiC does in theory support...

Anyone still has the checkpoint? @huanzhang12 @anishathalye

ups, overlooked! thanks!

@june1819 1. We encode with the CDF, and we encode channel per channel. The CDF and PDF are calculatable from each other, and in the paper we use the simpler...

Awesome! I‘m currently busy with the next CVPR but will have a look as soon as possible.

Hi, thanks for the effort. Have been busy but will look into this as soon as I have time!

Looks good! I suppose this will break support for 1.1 though? Is there anything else that might break by upgrading to a torch minimum of 1.4? @wangkaiwan Are you successfully...

Thanks for the info. With „using this pckage“, do you mean torchac or the broader training pipelines? I‘m a bit worried that some things in my networks silently break if...