Differentiable stochastic-computing primitives for PyTorch: train neural networks natively SC-aware.
-
Updated
Jun 14, 2026 - Python
Differentiable stochastic-computing primitives for PyTorch: train neural networks natively SC-aware.
A PyTorch framework bridging the device-to-algorithm gap for Edge AI. MagNet (Magnetic + Neural Network) simulates SOT-MTJ hardware constraints (quantization, read noise) to enable Quantization-Aware Training (QAT) for spintronic neural networks.
Add a description, image, and links to the hardware-aware-training topic page so that developers can more easily learn about it.
To associate your repository with the hardware-aware-training topic, visit your repo's landing page and select "manage topics."