Published July 27, 2024
| Version 0.5.2
Software
Open
e3nn/e3nn: 2024-07-26
Authors/Creators
- Mario Geiger
- Tess Smidt1
- Alby M.2
- Benjamin Kurt Miller3
- Wouter Boomsma
- Bradley Dice4
- Kostiantyn Lapchevskyi
- Michał Tyszkiewicz
- Martin Uhrin
- Mit Kotak
- Simon Batzner
- Dylan Madisetti
- Saurav Maheshkar5
- Jes Frellsen6
- Nuri Jung7
- Colin Kohler8
- Richard Barnes
- Sophia Sanborn9
- jkh
- Mingjian Wen10
- Boris Fomitchev11
- Alex Morehead12
- Josh Rackers
- Marcel Rød13
- Michael Bailey
- Michael Brocidiacono
- 1. @atomicarchitects
- 2. @mir-group @ Harvard University
- 3. University of Amsterdam
- 4. @NVIDIA @rapidsai
- 5. @lightly-ai
- 6. Technical University of Denmark (DTU)
- 7. @seoklab
- 8. Northeastern University
- 9. Science Corporation
- 10. University of Houston
- 11. NVIDIA
- 12. University of Missouri
- 13. Stanford
Description
[0.5.2] - 2024-07-26
Added
o3.experimental.FullTensorProductv2 | ElementwiseTensorProductv2for compatibility withtorch.compile(..., fulgraph=True)- enable pip caching in CI
- Optional scalar bias term in _batchnorm.py
Changed
- refactor to use pyproject.toml for packaging
- refactor gh community files
- move pylint, coverage and flake8 configuration to pyproject.toml
Fixed
- Fix TorchScript warning "doesn't support instance-level annotations" (#437)
Files
e3nn/e3nn-0.5.2.zip
Files
(499.7 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:b04453431f93d80a852916c21a17cacd
|
499.7 kB | Preview Download |
Additional details
Related works
- Is supplement to
- Software: https://github.com/e3nn/e3nn/tree/0.5.2 (URL)