I’m trying to detect the orange note with my limelight.
First, I’m building a model in Teachable Machine and exporting it as a Tensorflow Lite model, then I’m trying to upload both the tflite file and the label txt file into the Limelight Dashboard, but when I upload the label, it just crashes and the camera freezes.
Does anyone have any idea of what’s going wrong ?
Any help is appreaciated.
We are having the same issue. I am not 100% sure but it seems like the Teachable Machine creates a classifier tflite file. It seems it won’t work as a detector in Limelight. Limelight flashes green lights and freezes on screen.
We loaded our tflite file to the classifier pipeline type and it doesn’t crash. I don’t think we can use the classifier type to identify notes on field. So not sure if Teachable machine will work.
Were you able to get your limelight to work after you uploaded the tflite files to the detector? Ours just keeps on freezing. I don’t see a way to delete the tflite or label files. I am thinking we have to re-image the Limelight.
I will post back if we have any luck this weekend. If anyone else has tips that would be great!!
When I uploaded the tflite file to the detector, the camera freezed and the software just crashed, also I cannot re-enter, I have to refresh it and reboot the limelight dashboard. I think we have to wait until Limelight releases the trained network (like the one for cones and cubes last year) but for the note. Thanks for the help!
we had the same issue, the way we got around it was uploading the labels.txt file first, then the model
Team 3847 on their Day 8 post has a tflight file that works on limelight detector pipeline for the note game piece. We tried this out today and it worked. I did reach out to Limelight. They said they should have one ready soon.
Spectrum 3847 | Build Blog 2024 - FIRST / Open Alliance - Chief Delphi
1 Like
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.