Show HN: CNET – Complex-valued neural networks from scratch in C++/CUDA
Hacker News · 23 Oct 2026, 08:05 AEST
CNet is a C++/CUDA framework for building and researching deep complex valued neural networks. It also optimizes complex valued functions using gradient descent with Wirtinger derivatives.
Key facts
- CNet is a C++/CUDA framework for building and researching deep complex valued neural networks, as well as for optimization of complex valued functions using gradient descent with Wirtinger derivatives.
- Every built-in layer ships both a CPU reference and a CUDA kernel.
- The fields where these can be implemented most readily are: Signal Communications; Healthcare (both medical image and ECG); Deep Fake Detections; and Acoustic Analysis for Industrial Maintenance and expansion.
Read aloud
CNet is a C++/CUDA framework for building and researching deep complex valued neural networks. It also optimizes complex valued functions using gradient descent with Wirtinger derivatives. CNet is a C++/CUDA framework for building and researching deep complex valued neural networks, as well as for optimization of complex valued functions using gradient descent with Wirtinger derivatives. Every built-in layer ships both a CPU reference and a CUDA kernel. The fields where these can be implemented most readily are: Signal Communications; Healthcare both medical image and ECG; Deep Fake Detections; and Acoustic Analysis for Industrial Maintenance and expansion.
Sources in this cluster
- Show HN: CNET – Complex-valued neural networks from scratch in C++/CUDA
Hacker News · 23 Oct 2026, 08:05 AEST