TFLite for Microcontrollers on RT686

cancel
Showing results for 
Search instead for 
Did you mean: 

TFLite for Microcontrollers on RT686

154 Views
Contributor I

Hi,

I attempting to build and run the TFLite for Microcontrollers Benchmarkson the i.MX RT600 Evaluation Kit, however I'm getting stuck at trying to deploy the built application onto the board. Using gdb to load & run does not work, neither did JTAG.

For reference, I was able to successfully build and deploy the Hello World adaptation – Create a eIQ (TensorFlow Lite library) demo for i.MX RT6xx – using the project zip file provided (imported into the IDE).

I also tried replacing the target binary in the project above with the binary built using Make but that was not successful.

 

1. Are there any instructions on how to flash binaries built outside MCUXpresso IDE onto the board?

2. Any other alternative ways that I may have overlooked using the IDE itself?

3. Is there a way to make MCUXpresso IDE print out the deployment commands ran behind the scenes? (perhaps I could replicate these on the command line)

Thanks

Labels (1)
0 Kudos
1 Reply

79 Views
NXP TechSupport
NXP TechSupport

Hi  Gino Keeling,

Thank you for your interest in NXP Semiconductor products and
for the opportunity to serve you.
Before answering your question, I was wondering if you can clarify your inquiry.
According to the sentence: 'I also tried replacing the target binary in the project above with the binary built using Make but that was not successful', I guess you use the newly trained model to replace the model in the Create a eIQ (TensorFlow Lite library) demo for i.MX RT6xx , and later, you try to debug the updated demo project, is it that?

Looking forward to your reply.

Have a great day,
TIC

 

-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!

 

- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------

0 Kudos