-
- [ ] General forms using spherical harmonics
https://www.nature.com/articles/s41467-022-29939-5
https://docs.e3nn.org/en/latest/guide/convolution.html
- [x] Simpler equivariant layers
https://p…
-
[e3nn](https://www.connectedpapers.com/main/c7215ab4589ce3590910c597e86b1aba8e460d49/e3nn%3A-Euclidean-Neural-Networks/graph)
```mermaid
flowchart TD;
A[Freeman,1991\nThe design and use of stee…
-
In the section"Model Definition", IrrepsData is imported from e3nn-jax.
`from e3nn_jax import IrrepsData`
But this file is deleted from e3nn-jax, so I think it may need to be replaced by Irreps?
…
-
-
- [x] `EquivGraphConv` layer
- [ ] `EGNN` network and example
> Ref. [E(n) Equivariant Graph Neural Networks](https://arxiv.org/abs/2102.09844)
-
**General Interpretability:**
- [interpretable ml book](https://christophm.github.io/interpretable-ml-book/shapley.html), specifically sections on `learned features`, `Shapley values`, `Influential I…
-
Hi @awaelchli
So sorry for the last minute question! I have been closely following physics-informed / equivariant neural networks over the last few months, and I was very excited to see our overlap…
-
from https://github.com/jax-md/jax-md/blob/main/notebooks/equivariant_neural_networks.ipynb
```
def single_loss_fn(params, position, E_target, F_target):
l_nbrs = nbrs.update(position)
E, G …
-
I think it is interested to think of the projects using TFGA for quaternion transformation with nn.
-
### 🚀 The feature, motivation and pitch
A new equivariant model with SOTA on QM9+MD17. Could be a great addition to PyG!
* Paper: https://arxiv.org/abs/2204.05249
* Code: https://github.com/mir…