From wio terminal, I learned about the tensorflow lite library. Thanks for that.
This library is what I want to get. Therefore wio terminal is so important for me.
I would like to use Wio terminal as the small tensorflow test platform.
Is there any further articles w.r.t. this tensorflow lite library implementation to wio terminal?
Already I enrolled the coursera lecture, which describes only Sparfun board.
I would like to your making history of the implementation of the tensorflow lite library to Wio terminal and future release plan.
Thanks in advance.
We have been working with edgeimpulse to launch a more convenient and useful embedded deep learning solution.
https://github.com/edgeimpulse @youngil_sohn
I am studyng and trying to apply tensorflow lite library.
I found that there are several C++ class definitions.
Do you know how to compile this with C-only mode?
Second question : What is edge impulse? is this a candidate for tensorflow lite?
is this library use class (C++) or not?
C language is not suitable for deep learning. Please use the c++
Thanks for response. I am familiar with C++ and classes. I have been using python and MFC since 2002. However the reason why I ask is that I want to (must) use the real C-based Micom now. I would like to test Wio Terminal as a test-platform for that micom. the Specification of the Micom is similar. I don’t think that we cannot implement ANN using C-libraries. Recently our engineers made c-implementation of the ANN models. it consists of just multiplying and adding, and just a simple non-linear function like relu. so there is no reason not to implement of ANN using C-library. Anyway. I looked into the general library functions of the tensorflow lite. Without the support of the google.com it seems to be impossible to convert ANN models to my Micom.
I understand your opinion so. Thanks for your comments.