Code for our paper at ECCV 2020: Post-Training Piecewise Linear Quantization for Deep Neural Networks

Last update: Dec 15, 2022

Related tags

Deep Learning PWLQ

Overview

PWLQ

Updates

2020/07/16 - We are working on getting permission from our institution to release our source code. We will release it once we are granted approval.

2020/08/24 - Code released.

PyTorch Code for our paper at ECCV 2020 (oral presentation): Post-Training Piecewise Linear Quantization for Deep Neural Networks [Paper] [arXiv]

By Jun Fang, Ali Shafiee, Hamzah Abdel-Aziz, David Thorsley, Georgios Georgiadis, Joseph Hassoun

Approach

Performance

Requirements

The code was verified on Python-3.6+, PyTorch-1.2+.

Usage

Check PWLQ at quant/pwlq.py.

Run bash eval.sh to evaluate PWLQ on ImageNet. Results would be recorded at results/*.csv.

Results might be slightly different due to the randomness of calibration samples for activation ranges.

Citation

We appreciate it if you would please cite our paper:

@inproceedings{pwlq,
  title={Post-Training Piecewise Linear Quantization for Deep Neural Networks},
  author={Fang, Jun and Shafiee, Ali and Abdel-Aziz, Hamzah and Thorsley, David and Georgiadis, Georgios and Hassoun, Joseph},
  booktitle={ECCV},
  year={2020}
}

You might also like...

Code for paper ECCV 2020 paper: Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop.

Who Left the Dogs Out? Evaluation and demo code for our ECCV 2020 paper: Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization

29 Dec 28, 2022

Degree-Quant: Quantization-Aware Training for Graph Neural Networks.

Degree-Quant This repo provides a clean re-implementation of the code associated with the paper Degree-Quant: Quantization-Aware Training for Graph Ne

35 Oct 7, 2022

[CVPR'21] Locally Aware Piecewise Transformation Fields for 3D Human Mesh Registration

Locally Aware Piecewise Transformation Fields for 3D Human Mesh Registration This repository contains the implementation of our paper Locally Aware Pi

70 Dec 19, 2022

A framework that constructs deep neural networks, autoencoders, logistic regressors, and linear networks

A framework that constructs deep neural networks, autoencoders, logistic regressors, and linear networks without the use of any outside machine learning libraries - all from scratch.

2 Nov 14, 2022

QTool: A Low-bit Quantization Toolbox for Deep Neural Networks in Computer Vision

This project provides abundant choices of quantization strategies (such as the quantization algorithms, training schedules and empirical tricks) for quantizing the deep neural networks into low-bit counterparts.

51 Dec 10, 2022

IAST: Instance Adaptive Self-training for Unsupervised Domain Adaptation (ECCV 2020)

This repo is the official implementation of our paper "Instance Adaptive Self-training for Unsupervised Domain Adaptation". The purpose of this repo is to better communicate with you and respond to your questions. This repo is almost the same with Another-Version, and you can also refer to that version.

84 Dec 12, 2022

Code for the paper "Improving Vision-and-Language Navigation with Image-Text Pairs from the Web" (ECCV 2020)

Improving Vision-and-Language Navigation with Image-Text Pairs from the Web Arjun Majumdar, Ayush Shrivastava, Stefan Lee, Peter Anderson, Devi Parikh

44 Dec 14, 2022

Code for ECCV 2020 paper "Contacts and Human Dynamics from Monocular Video".

Contact and Human Dynamics from Monocular Video This is the official implementation for the ECCV 2020 spotlight paper by Davis Rempe, Leonidas J. Guib

207 Jan 5, 2023

PyTorch implementation of our Adam-NSCL algorithm from our CVPR2021 (oral) paper "Training Networks in Null Space for Continual Learning"

Adam-NSCL This is a PyTorch implementation of Adam-NSCL algorithm for continual learning from our CVPR2021 (oral) paper: Title: Training Networks in N

34 Dec 21, 2022

Comments

Training set

Hello, thanks for your paper and code. I read your code but I don't understand why you use training set when post training quantization does not require training set. please explain it if possible. Thank you.

opened by amkmpro 0
Activation Quantization Error
Hello, thanks for your paper and code. When I try to run activation quantization. I keep getting error that is

torch.clamp must be number not tensor

even if I set topk in QuantAct to one.

Thank you in adavnce
opened by rematchka 0
Inquiry about integer inference with bias correction

Hi, thank you for sharing the source code of your work, it's amazing. I would like to inquire about pure integer inference using bias correction as indicated in the paper. when using bias correction on INT 8 quantized weights, the weights change into floating-point values, therefore do you take round/floor/ ceil in order to do Pure INT computation on quantitated weights?

Thank you in advance

opened by rematchka 0
Support/code for other reported networks??

I got the following message when trying to run mobilenetV2:

RuntimeError: Please implement BatchNorm folding for mobilenet_v2 !!!

Will you release the code for other networks as well, please?

opened by Amin-Azar 0

Code for our paper at ECCV 2020: Post-Training Piecewise Linear Quantization for Deep Neural Networks

Related tags

Overview

PWLQ

Requirements

Usage

Citation

You might also like...

Code for paper ECCV 2020 paper: Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop.

Degree-Quant: Quantization-Aware Training for Graph Neural Networks.

[CVPR'21] Locally Aware Piecewise Transformation Fields for 3D Human Mesh Registration

A framework that constructs deep neural networks, autoencoders, logistic regressors, and linear networks

QTool: A Low-bit Quantization Toolbox for Deep Neural Networks in Computer Vision

IAST: Instance Adaptive Self-training for Unsupervised Domain Adaptation (ECCV 2020)

Code for the paper "Improving Vision-and-Language Navigation with Image-Text Pairs from the Web" (ECCV 2020)

Code for ECCV 2020 paper "Contacts and Human Dynamics from Monocular Video".

PyTorch implementation of our Adam-NSCL algorithm from our CVPR2021 (oral) paper "Training Networks in Null Space for Continual Learning"

Comments

Training set

Activation Quantization Error

Inquiry about integer inference with bias correction

Support/code for other reported networks??

Owner

git《Learning Pairwise Inter-Plane Relations for Piecewise Planar Reconstruction》(ECCV 2020) GitHub:

PyTorch implementation of the Deep SLDA method from our CVPRW-2020 paper "Lifelong Machine Learning with Deep Streaming Linear Discriminant Analysis"

Linear algebra python - Number of operations and problems in Linear Algebra and Numerical Linear Algebra

PyTorch code for our ECCV 2018 paper "Image Super-Resolution Using Very Deep Residual Channel Attention Networks"

PyTorch code for our ECCV 2020 paper "Single Image Super-Resolution via a Holistic Attention Network"

Code for the paper: Adversarial Training Against Location-Optimized Adversarial Patches. ECCV-W 2020.

DiffQ performs differentiable quantization using pseudo quantization noise. It can automatically tune the number of bits used per weight or group of weights, in order to achieve a given trade-off between model size and accuracy.

Quantization library for PyTorch. Support low-precision and mixed-precision quantization, with hardware implementation through TVM.

Nonuniform-to-Uniform Quantization: Towards Accurate Quantization via Generalized Straight-Through Estimation. In CVPR 2022.

With this package, you can generate mixed-integer linear programming (MIP) models of trained artificial neural networks (ANNs) using the rectified linear unit (ReLU) activation function