The hand landmarker example script for Raspberry Pi does not parse float inputs correctly:
$ python3 detect.py --minHandPresenceConfidence 0.33
Traceback (most recent call last):
File "/home/adiasg/Projects/mediapipe-samples/examples/hand_landmarker/raspberry_pi/detect.py", line 224, in <module>
main()
File "/home/adiasg/Projects/mediapipe-samples/examples/hand_landmarker/raspberry_pi/detect.py", line 218, in main
run(args.model, int(args.numHands), args.minHandDetectionConfidence,
File "/home/adiasg/Projects/mediapipe-samples/examples/hand_landmarker/raspberry_pi/detect.py", line 92, in run
detector = vision.HandLandmarker.create_from_options(options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/adiasg/Projects/mediapipe-samples/venv/lib/python3.11/site-packages/mediapipe/tasks/python/vision/hand_landmarker.py", line 365, in create_from_options
task_info.generate_graph_config(
File "/home/adiasg/Projects/mediapipe-samples/venv/lib/python3.11/site-packages/mediapipe/tasks/python/core/task_info.py", line 83, in generate_graph_config
task_options_proto = self.task_options.to_pb2()
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/adiasg/Projects/mediapipe-samples/venv/lib/python3.11/site-packages/mediapipe/tasks/pyt
hon/vision/hand_landmarker.py", line 273, in to_pb2
hand_landmarker_options_proto.hand_landmarks_detector_graph_options.min_detection_confidence = (
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: must be real number, not str
The hand landmarker example script for Raspberry Pi does not parse float inputs correctly: