Skip to content
프로젝트

awai54st/PYNQ-Classification

GitHub에서 보기

awai54st/PYNQ-Classification has 624 GitHub stars and 221 forks. GitHot measured -2 stars in 24 hours. Last checked 2026-11-20 14:57 UTC. Last push 2018-05-15.

이 수치를 측정하는 방법

측정된 스타 증가

in 24h
-2
in 7 days
기록 수집 중
in 15 days
기록 수집 중
in 1 month
기록 수집 중

서로 겹치는 기간의 순증가량이며 일별 합계가 아닙니다. 측정값이 없으면 비워 둡니다. 스타는 관심도를 나타내며 품질을 보장하지 않습니다.

소개

Python on Zynq FPGA for Convolutional Neural Networks

머신러닝

README.md

Original

PROJECT NAME:

PYNQ Classification - Python on Zynq FPGA for Convolutional Neural Networks (Alpha Release)

BRIEF DESCRIPTION:

This repository presents a fast prototyping framework, which is an Open Source framework designed to enable fast deployment of embedded Convolutional Neural Network (CNN) applications on PYNQ platforms.

CITATION:

If you make use of this code, please acknowledge us by citing our article:

@inproceedings{Wang_FCCM18,
    author={E. Wang and J. J. Davis and P. Y. K. Cheung},
    booktitle={IEEE Symposium on Field-programmable Custom Computing Machines (FCCM)},
    title={{A PYNQ-based Framework for Rapid CNN Prototyping}},
    year={2018}
}

INSTRUCTIONS TO BUILD AND TEST THE PROJECT:

Repository Organisation

The project demo accepts pre-trained CNN models in either Caffe or Theano syntax, hence the step 1 and 2 introduces how to install Caffe and Theano (with Lasagne) on PYNQ. Step 3 explains how to download and run the demos for LeNet and CIFAR-10 (Caffe "quick" version) models.

For a quick overview on the project please watch my video tutorial.

1. PYNQ SD Card Image

We have prepared a SD card image with pre-installed Caffe and Theano dependencies. A SD card with at least 16GB is needed. The static IP for the PYNQ Jupyter Notebook is 192.168.2.99

Download Link (on Baidu Drive)

Download Link (on Google Drive)

If you wish to setup Caffe and Theano dependencies on your own, please see MANUAL_INSTAL.md for instructions. (NOT RECOMMENDED since multiple issues have been reported)

2. Vivado Project Setup - How to implement more CNN models?

Please go to hw/README.md for guide on regenerating the Vivado and Vivado HLS projects.

References

1. BNN-PYNQ

If you find BNN-PYNQ useful, please cite the FINN paper:

@inproceedings{finn,
author = {Umuroglu, Yaman and Fraser, Nicholas J. and Gambardella, Giulio and Blott, Michaela and Leong, Philip and Jahre, Magnus and Vissers, Kees},
title = {FINN: A Framework for Fast, Scalable Binarized Neural Network Inference},
booktitle = {Proceedings of the 2017 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays},
series = {FPGA '17},
year = {2017},
pages = {65--74},
publisher = {ACM}
}

2. Caffe

Caffe is released under the BSD 2-Clause license. The BAIR/BVLC reference models are released for unrestricted use.

Please cite Caffe in your publications if it helps your research:

@article{jia2014caffe,
  Author = {Jia, Yangqing and Shelhamer, Evan and Donahue, Jeff and Karayev, Sergey and Long, Jonathan and Girshick, Ross and Guadarrama, Sergio and Darrell, Trevor},
  Journal = {arXiv preprint arXiv:1408.5093},
  Title = {Caffe: Convolutional Architecture for Fast Feature Embedding},
  Year = {2014}
}

3. Theano

@ARTICLE{2016arXiv160502688short,
   author = {{Theano Development Team}},
    title = "{Theano: A {Python} framework for fast computation of mathematical expressions}",
  journal = {arXiv e-prints},
   volume = {abs/1605.02688},
 primaryClass = "cs.SC",
 keywords = {Computer Science - Symbolic Computation, Computer Science - Learning, Computer Science - Mathematical Software},
     year = 2016,
    month = may,
      url = {http://arxiv.org/abs/1605.02688},
}

4. Lasagne

@misc{lasagne,
  author       = {Sander Dieleman and
                  Jan Schlüter and
                  Colin Raffel and
                  Eben Olson and
                  Søren Kaae Sønderby and
                  Daniel Nouri and
                  others},
  title        = {Lasagne: First release.},
  month        = aug,
  year         = 2015,
  doi          = {10.5281/zenodo.27878},
  url          = {http://dx.doi.org/10.5281/zenodo.27878}
}