Zak Jost

DGL team: can you please describe the impact of this bug? Has this always been a problem? If not, which versions of DGL/PyTorch are impacted?

Yes, thank you. I'm wondering which versions of DGL/PyTorch are affected by this bug.

It looks like the core neural net operations are implemented from scratch (e.g. convolution). Why not use [built-in PyTorch c++ support](https://pytorch.org/tutorials/advanced/cpp_export.html)?

It's a good question and the answer will depend on if the optimizations they've made are well-aligned with this use-case. For example, I've heard that instantiating tensors is relatively slow...

You should set your PYTHONPATH environment variable to be equal to the directory that contains your cloned version of this repo. For instance, if you cloned this repo into a...

This question is better for Stack Overflow, as it isn't specific to this code, it is a general Python question. To make sure your changes took effect, you can `echo...

I fixed this problem by making several updates to this code and one to prettytensor to accommodate the new TensorFlow API. You can get those changes in [my fork](https://github.com/zjost/InfoGAN) of...

@metatl here's what you need to change in prettytensor: https://github.com/google/prettytensor/pull/57/files This is linked to in my fork's ReadMe.

I forked this repo and made the changes to use tensorflow 1.3.0. You can find it [here](https://github.com/zjost/InfoGAN). But [this diff](https://github.com/zjost/InfoGAN/commit/858faf6f011492f73b7092066ae4800bfed1e0f5) shows the changes. I also needed to change a part...

I'm here because ellipsis are used in [StableDiffusion code](https://github.com/CompVis/stable-diffusion/blob/main/scripts/img2img.py#L234), and I had a hard time understanding since I couldn't find in the docs.