Current collected state Current
Observed since 20 Oct, 19:00 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
Adam-atan2 Fused Optimizer
Drop-in replacement of torch.optim.AdamW.
foreach, fused argument, as the optimizer is already fusedamsgrad, maximize, capturable, differentiable argument yetpip install adam-atan2
from adam_atan2 import AdamATan2
# All supported arguments are listed below
optim = AdamATan2(model.parameters(),
lr=1e-3,
weight_decay=0.1,
betas=(0.9, 0.95)
)
We tested the consistency against reference AdamW-atan2 PyTorch implementation. To run tests, clone this repository, run pytest:
pip install -e .
pytest
Download counts and popularity will appear here as the collector receives them.
| Version | Published | Direct dependencies | Open version |
|---|---|---|---|
| 0.0.3 | 2 Apr 2025 | 0 |
This describes our collected data; it is not a security assessment.
Observed since 20 Oct, 19:00 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
package:e00d1062cfc51d92408394a4b1381d0cc7c2d282156c82a8bd6819fc7196d853