Published June 21, 2021
| Version 0.3.3
Software
Open
e3nn/e3nn: 2021-06-21
Creators
- 1. @atomicarchitects
- 2. MIR group @ Harvard University
- 3. University of Amsterdam
- 4. University of Michigan
- 5. Harvard
- 6. Technical University of Denmark (DTU)
- 7. Redwood Center for Theoretical Neuroscience
Description
[0.3.3] - 2021-06-21 Changed
FullyConnectedNetis now atorch.nn.Sequential
BatchNormwas not equivariant for pseudo-scalars
biasesargument too3.Linearnn.models.v2106:MessagePassingtakes a sequence of irrepsnn.models.v2106:Convolutioninpired from Batch Normalization Biases Residual Blocks Towards the Identity Function in Deep Networks
Files
e3nn/e3nn-0.3.3.zip
Files
(736.7 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:a0b413ab7451b344ba14233f3799e8fe
|
736.7 kB | Preview Download |
Additional details
Related works
- Is supplement to
- https://github.com/e3nn/e3nn/tree/0.3.3 (URL)