Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Embedded Linux GUI App with Tensorflow Lite - QT C++ or Kivy Python

Status
Not open for further replies.

sakibnaz

Full Member level 3
Joined
Jul 6, 2008
Messages
162
Helped
6
Reputation
12
Reaction score
6
Trophy points
1,298
Activity points
2,755
Hi All.

I am working on an Project which have ARM Quad-Core Board with Touch LCD. I need to make a nice GUI modern looking App. However, my project also needs Tensorflow Lite for some ML application.

I am little bit confused about choosing the proper development path either C++ or Python for the GUI App.

Normally QT C++ looks great for the modern looking GUI elements. But not sure if C++ App will be good to deploy the TFLite ML Model?

Or, for my application best will be Python GUI App (i.e. KiVy based) as with Python its easy to load the TFLite Model.

However, the TFLite Mode I'll train to a desktop PC with Python.

Looking forward your suggestion.

Thank you.

Regards.
 

C++ and qt should be fine for deploying your tensor flow lite. There is a lot of examples using tensor flow lite with c++. I guess converting the model to tensor flow to tensor flow lite maybe the tedious part. I could be wrong with tensor flow lite but my company always had issue converting to OpenVino or tensorrt like freezing the model, padding, layers not supported, different framework version just name a few things we have encountered.
 

Thanks eziggurat. Yes I chooses Qt C++.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top