Skip to main content

omnibias.binary.torch

0.1.0a2AlphaPython >=3.10Apache-2.0

Binary and multilevel quantization with explicit hard-forward and surrogate-gradient contracts.

import omnibias.binary.torch
Module contract & mathematical conventions

Torch backend for omnibias-binary.

Shared formulas do not imply universal floating-point identity. Verify dtype, shape, derivative-order support, and the stated numerical assumptions for the path you use.