-
## Slimmed-down RCNP
Thanks to @uremes for realizing this!
We can consider a "slim" variant of RCNP that does not use the encoder representation of the context set, taking advantage of the fact …
-
Hi!
Thanks again for maintaining this valuable resource.
We have three new NeurIPS papers on (provably) robust learning on graphs
and wanted to ask if we could include them.
---
In this p…
-
if I want to add postion embedding in the VN-transformer? Will it disrupt the rotational equivariance property?
-
Hello! I'm very interested in your work. However, I've encountered an issue. I'm currently trying to evaluate the pre-trained VAE using mnist_eval.sh, but I've noticed that the repository doesn't cont…
-
In the example [code](https://github.com/QUVA-Lab/escnn/blob/master/examples/e2wrn.py#L135), there is a condition on the number of rotations for selecting the kernel size. What is the full intuition b…
-
Dear developers of shnetpack:
Hi. Thank you all so much for making the LAMMPS interface workable with shnetpack. I have three questions relating the LAMMPS interface:
1. Does the interface work …
-
- Write Methods and sketch Intro/Background
- Write sketch of abstract
-
We should add one!
-
I'm trying to train a image recognition network which should be invariant under SO(2). Here is a minimal example of my network that uses escnn's modules:
```
class EquivariantConvBN(enn.EquivariantM…
-
Hi Gabriele, thanks again for the great library!
I apologize in advance if the answer to this can be found in the docs/paper/thesis, I couldn't find it there.
My question is rather soft: what is…