TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi
TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi copied to clipboard
Custom model from Azure Custom Vision
trafficstars
Hi, I would like to ask about compatibility these example with other tflite, exported at the Azure Custom Vision.
To train my custom dataset, I used MS Azure Custom Vision. That is pretty easy to do labelin, train and export Leave the model's correctness, I cannot use it with your detection examples.
It returns
IndexError: list index out of range
at
classes = interpreter.get_tensor(output_details[classes_idx]['index'])[0]
I tried TF2 index order for boxes_idx, classes_idx and scores_idx.