jianhayes/NESTQUANT
NestQuant: Post-Training Integer-Nesting Quantization for On-Device DNN [IEEE TMC 2025]
This project helps machine learning engineers and researchers optimize their deep neural networks for deployment on resource-constrained mobile and edge devices. It takes an existing, trained neural network model (like a CNN or Vision Transformer) and applies a technique called integer-nesting quantization. The output is a more efficient, smaller model that maintains accuracy while requiring less memory and computational power for inference. This is ideal for those working on AI for embedded systems or mobile applications.
No commits in the last 6 months.
Use this if you need to reduce the size and computational demands of your deep learning models so they can run efficiently on devices with limited resources, such as smartphones, IoT devices, or specialized embedded hardware.
Not ideal if your models are already sufficiently small or if you are deploying them on powerful cloud-based GPUs where resource constraints are not a primary concern.
Stars
9
Forks
—
Language
Python
License
—
Category
Last pushed
Oct 10, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/jianhayes/NESTQUANT"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
open-mmlab/mmengine
OpenMMLab Foundational Library for Training Deep Learning Models
Xilinx/brevitas
Brevitas: neural network quantization in PyTorch
fastmachinelearning/qonnx
QONNX: Arbitrary-Precision Quantized Neural Networks in ONNX
google/qkeras
QKeras: a quantization deep learning library for Tensorflow Keras
tensorflow/model-optimization
A toolkit to optimize ML models for deployment for Keras and TensorFlow, including quantization...