fab-jul
fab-jul
Thanks for the info. I think the most sensible path forward is to package this as a separate repo, where we can change requirements without breaking any code here! I...
I will also try to disentagle the CUDA part more, as it is only really relevant for this paper I think
FYI, I released this as a standalone library that does not need CUDA here: https://github.com/fab-jul/torchac
That’s not gonna work, the model is trained for 8bit and thus produces probability distributions for that. You would need to: - change the model to predict a probability for...
Depends on the file! What's `006239`?
`AT_CHECK` is deprecated, use `TORCH_CHECK`.
Probably easier to just create an environment with PyTorch 1.1, or is this not possible?
Arxiv is now available [https://arxiv.org/abs/1811.12817v3](https://arxiv.org/abs/1811.12817v3). Merged feat/v3 into master. Please let me know here if there are issues with this release.
I think one contains values at the symbols. e.g. if your quantization grid is: ``` levels = [0.2, 0.5, 0.7] ``` and your symbols are ``` sym = [0, 0,...
hi, I think this should work. can you post the exact content of the base.cf file? and the command you run, as well as the error. Thanks