PyPI page
Home page
Author:
Rishabh A. Patil
License:
MIT
Summary:
Topological Loss Engineering: differentiable, optimizer-free regularizers that embed 2024-2026 optimizer breakthroughs (Muon/XSAM/CWD/AdEMAMix/NTKMTL/SymNoise) directly into the loss.
Latest version:
0.1.0
Required dependencies:
torch
Optional dependencies:
matplotlib
|
numpy
|
pytest
Downloads last day:
0
Downloads last week:
6
Downloads last month:
42