alibaba/lightweight-neural-architecture-search
A collection of training-free neural architecture search methods for designing efficient vision models on CPU within limited time budgets.
Not currently ranked — collecting fresh signals.
star history
This repository implements zero-cost NAS techniques including DeepMAD and Zen-NAS, which evaluate architecture quality without full training by analyzing entropy or gradient signals. It supports joint quantization and architecture search for mixed-precision models. Applications include object detection (MAE-DET) and video action recognition, all built on PyTorch.
Frequently asked
- What is alibaba/lightweight-neural-architecture-search?
- A collection of training-free neural architecture search methods for designing efficient vision models on CPU within limited time budgets.
- Is lightweight-neural-architecture-search open source?
- Yes — alibaba/lightweight-neural-architecture-search is open source, released under the Apache-2.0 license.
- What language is lightweight-neural-architecture-search written in?
- alibaba/lightweight-neural-architecture-search is primarily written in Python.
- How popular is lightweight-neural-architecture-search?
- alibaba/lightweight-neural-architecture-search has 447 stars on GitHub.
- Where can I find lightweight-neural-architecture-search?
- alibaba/lightweight-neural-architecture-search is on GitHub at https://github.com/alibaba/lightweight-neural-architecture-search.