GOWIN GoAI 2.0 Machine Learning Interface User Guide

: June 6, 2024
: GOWIN

Table of Contents

GoAI 2.0 Quick Start Guide
Prerequisite Software
How to Deploy Examples
Retraining Models
Replacing Trained Model
GoAI Information
Support and Feedback
Revision History
References
Read User Manual Online (PDF format)
Download This Manual (PDF format)

GoAI 2.0 Quick Start Guide

Prerequisite Software

ARM Keil v5
Gowin IDE or Gowin Programmer

How to Deploy Examples

Pull the following GoAI Github Repository. https://github.com/gowinsemi/GoAI.git
Navigate to the GoAI Board you are using and the example you wish to deploy under ‘\SDK\win\’.
Navigate to the `*\bin\goai_run.bat’ batch file and open in any text editor.
Update the file and program paths based on your machine.
a) ‘set tflite_model_file=’ ← flatbuffers (*.tflite) file and path
b) ‘set gowin_programmer_path=’ ← Gowin programmer path
c) ‘set keil_path=’ ← Keil v5 IDE path
Once steps 1-4 are performed they do not need to be repeated. Plug in and turn on GoAI 2.0 board.
Open a command window by typing ‘cmd’ in the Windows search bar.
Navigate to directory containing the batch file. Type goai_run.bat and enter to run the batch file. Model and design will be compiled and loaded to the board automatically.

Retraining Models

Gowin’s public GoAI demos are based on the following TinyML book by Pete Warden and Daniel Situnayake. Detailed instructions on how to train models in Tensorflow and TFLu are available in this book. https://www.oreilly.com/library/view/tinyml/9781492052036/

Replacing Trained Model

Models are trained, quantized and optimized into a Tensorflow Flatbuffers (.tflite) file. This file contains all model details as well as weight and bias training coefficients. The Flatbuffers file can be replaced with an updated and retrained model you wish to deploy under `\data\model\*`

GoAI Information

For more GoAI informationm, please refer to https://www.gowinsemi.com/en/market/featured_detail/11/ [email protected]

Support and Feedback

Gowin Semiconductor provides customers with comprehensive technical support. If you have any questions, comments, or suggestions, please feel free to contact us directly using the information presented below.
Website: www.gowinsemi.com
E-mail: [email protected]

Revision History

GOWIN-GoAI-2.0-Machine-Learning-Interface- fig1

Guangdong Gowin Semiconductor Corporation. All Rights Reserved. No part of this document may be reproduced or transmitted in any form or by any denotes, electronic, mechanical, photocopying, recording or otherwise, without the prior written consent of GOWINSEMI

Disclaimer
GOWINSEMI®, LittleBee®, Arora, and the GOWINSEMI logos are trademarks of GOWINSEMI and are registered in China, the U.S. Patent and Trademark Office, and other countries. All other words and logos identified as trademarks or service marks are the property of their respective holders, as described at www.gowinsemi.com.cn. GOWINSEMI assumes no liability and provides no warranty (either expressed or implied) and is not responsible for any damage incurred to your hardware, software, data, or property resulting from usage of the materials or intellectual property except as outlined in the GOWINSEMI Terms and Conditions of Sale. All information in this document should be treated as preliminary. GOWINSEMI may make changes to this document at any time without prior notice. Anyone relying on this documentation should contact GOWINSEMI for the current documentation and errata.