Please HELP on Maix Yolo KModel generation tutorial

Hello, Is there resource sharing how to convert self-trained tiny-yolo-v2 keras h5 model or darknet .weight file to K-Model? Please help! Thanks in advance!

Train, Convert, Run MobileNet on Sipeed MaixPy and MaixDuino ! (mobilenet)
I tried to follow this example to do model conversion. Unfortunately, I am stuck in the final step (./tflite2kmodel.sh) with the following error message:
Fatal: Layer TensorflowReshape is not supported

If a similar tutorial could be provided for yolo, like the above-mentioned mobilenet one, it should be great. I appreciate very much!

Or introducing the detailed model structure in this yolo-v2 face demo (http://blog.sipeed.com/p/675.html) will be also very helpful. How the yolo face Kmodel generated from pre-trained yolo model?

1 Like