huawei-noah/AdderNet
A PyTorch implementation of AdderNet, a CVPR 2020 paper that replaces expensive multiplications with additions in convolutional neural networks for efficient inference.
Not currently ranked — collecting fresh signals.
star history
The repository implements a deep learning approach that trades multiplication operations for additions in convolutional neural networks, significantly reducing computational cost. It provides training code for CIFAR and ImageNet classification benchmarks using ResNet and VGG architectures with the L1-norm distance metric, achieving competitive accuracy without any multiplication operations in convolution layers.
Frequently asked
- What is huawei-noah/AdderNet?
- A PyTorch implementation of AdderNet, a CVPR 2020 paper that replaces expensive multiplications with additions in convolutional neural networks for efficient inference.
- Is AdderNet open source?
- Yes — huawei-noah/AdderNet is open source, released under the BSD-3-Clause license.
- What language is AdderNet written in?
- huawei-noah/AdderNet is primarily written in Python.
- How popular is AdderNet?
- huawei-noah/AdderNet has 968 stars on GitHub.
- Where can I find AdderNet?
- huawei-noah/AdderNet is on GitHub at https://github.com/huawei-noah/AdderNet.