open-mmlab/mmrazor
A PyTorch-based toolbox for compressing deep learning models through pruning, quantization, knowledge distillation, and neural architecture search.
Not currently ranked — collecting fresh signals.
star history
This repository is part of the OpenMMLab ecosystem and provides a comprehensive model compression toolkit. It supports multiple compression techniques including network pruning, quantization, knowledge distillation, and neural architecture search (NAS). Built on PyTorch, it offers tools for compressing models across computer vision tasks like classification, detection, and segmentation, with support for methods such as AutoSlim and DARTS.
Frequently asked
- What is open-mmlab/mmrazor?
- A PyTorch-based toolbox for compressing deep learning models through pruning, quantization, knowledge distillation, and neural architecture search.
- Is mmrazor open source?
- Yes — open-mmlab/mmrazor is open source, released under the Apache-2.0 license.
- What language is mmrazor written in?
- open-mmlab/mmrazor is primarily written in Python.
- How popular is mmrazor?
- open-mmlab/mmrazor has 1.7k stars on GitHub.
- Where can I find mmrazor?
- open-mmlab/mmrazor is on GitHub at https://github.com/open-mmlab/mmrazor.