FastONN -- Python based open-source GPU implementation for Operational Neural Networks

Abstract

Operational Neural Networks (ONNs) have recently been proposed as a special class of artificial neural networks for grid structured data. They enable heterogenous non-linear operations to generalize the widely adopted convolution-based neuron model. This work introduces a fast GPU-enabled library for training operational neural networks, FastONN, which is based on a novel vectorized formulation of the operational neurons. Leveraging on automatic reverse-mode differentiation for backpropagation, FastONN enables increased flexibility with the incorporation of new operator sets and customized gradient flows. Additionally, bundled auxiliary modules offer interfaces for performance tracking and checkpointing across different data partitions and customized metrics.

References in corpus (2)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• cuDNN: Efficient Primitives for Deep Learning

Cited by in corpus (7)

• Robust Peak Detection for Holter ECGs by Self-Organized Operational Neural Networks
• Robust R-Peak Detection in Low-Quality Holter ECGs using 1D Convolutional Neural Network
• Self-Organized Variational Autoencoders (Self-VAE) for Learned Image Compression
• Convolutional versus Self-Organized Operational Neural Networks for Real-World Blind Image Denoising
• Image denoising by Super Neurons: Why go deep?
• Improved Active Fire Detection using Operational U-Nets
• Early Bearing Fault Diagnosis of Rotating Machinery by 1D Self-Organized Operational Neural Networks