https://github.com/alibaba/tinyneuralnetwork
TinyNeuralNetwork is an efficient and easy-to-use deep learning model compression framework.
https://github.com/alibaba/tinyneuralnetwork
deep-learning deep-neural-networks model-compression model-converter post-training-quantization pruning pytorch quantization-aware-training
Last synced: 3 months ago
JSON representation
TinyNeuralNetwork is an efficient and easy-to-use deep learning model compression framework.
- Host: GitHub
- URL: https://github.com/alibaba/tinyneuralnetwork
- Owner: alibaba
- License: mit
- Created: 2021-11-02T06:56:18.000Z (about 4 years ago)
- Default Branch: main
- Last Pushed: 2025-08-20T08:00:21.000Z (5 months ago)
- Last Synced: 2025-08-20T09:34:01.567Z (5 months ago)
- Topics: deep-learning, deep-neural-networks, model-compression, model-converter, post-training-quantization, pruning, pytorch, quantization-aware-training
- Language: Python
- Homepage:
- Size: 25.1 MB
- Stars: 842
- Watchers: 20
- Forks: 127
- Open Issues: 34
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
# TinyNeuralNetwork
[简体中文](README_zh-CN.md)
TinyNeuralNetwork is an efficient and easy-to-use deep learning model compression framework, which contains features like neural architecture search, pruning, quantization, model conversion and etc. It has been utilized for the deployment on devices such as Tmall Genie, Haier TV, Youku video, face recognition check-in machine, and etc, which equips over 10 million IoT devices with AI capability.
## Installation
Python >= 3.9, PyTorch >= 1.10
```shell
# Install the TinyNeuralNetwork framework
git clone https://github.com/alibaba/TinyNeuralNetwork.git
cd TinyNeuralNetwork
python setup.py install
# Alternatively, you may try the one-liner
pip install git+https://github.com/alibaba/TinyNeuralNetwork.git
```
Or you could build with docker
```shell
sudo docker build -t tinynn:pytorch1.9.0-cuda11.1 .
```
## Contributing
We appreciate your help for improving our framework. More details are listed [here](CONTRIBUTING.md).
## Basic modules
+ Computational graph capture: The Graph Tracer in TinyNeuralNetwork captures connectivity of PyTorch operators, which automates pruning and model quantization. It also supports code generation from PyTorch models to equivalent model description files (e.g. models.py).
+ Dependency resolving: Modifying an operator often causes mismatch in subgraph, i.e. mismatch with other dependent operators. The Graph Modifier in TinyNeuralNetwork handles the mismatchs automatically within and between subgraphs to automate the computational graph modification.
+ Pruner: OneShot (L1, L2, FPGM), ADMM, NetAdapt, Gradual, End2End and other pruning algorithms have been implemented and will be opened gradually.
+ Quantization-aware training: TinyNeuralNetwork uses PyTorch's QAT as the backend (we also support simulated bfloat16 training) and optimizes its usability with automating the fusion of operators and quantization of computational graphs (the official implementation requires manual implementation by the user, which is a huge workload).
+ Model conversion: TinyNeuralNetwork supports conversion of floating-point and quantized PyTorch models to TFLite models for end-to-end deployment.
## Project architecture
+ [examples](examples): Provides examples of each module
+ [models](models): Provides pre-trained models for getting quickstart
+ [tests](tests): Unit tests
+ [tinynn](tinynn): Code for model compression
+ [graph](tinynn/graph) : Foundation for computational graph capture, resolving, quantization, code generation, mask management, and etc
+ [prune](tinynn/prune) : Pruning algorithms
+ [converter](tinynn/converter) : Model converter
+ [util](tinynn/util): Utility classes
## RoadMap
+ Nov. 2021: A new pruner with adaptive sparsity
+ Dec. 2021: Model compression for Transformers
## Citation
If you find this project useful in your research, please consider cite:
```
@misc{tinynn,
title={TinyNeuralNetwork: An efficient deep learning model compression framework},
author={Ding, Huanghao and Pu, Jiachen and Hu, Conggang},
howpublished = {\url{https://github.com/alibaba/TinyNeuralNetwork}},
year={2021}
}
```
## Frequently Asked Questions
Because of the high complexity and frequent updates of PyTorch, we cannot ensure that all cases are covered through automated testing.
When you encounter problems You can check out the [FAQ](docs/FAQ.md), or join the Q&A group in DingTalk via the QR Code below.
