Face Detection GPU Models Errors

[Data science]

Hi all,

I need help to implement the face detection full-range and short-range GPU models on my windows. I have downloaded the FaceDetectionFullRangeGpu and FaceDetectionShortRangeGpu modules on my windows laptop and saved them inside the mediapipe/modules/face_detection location and modified the face_detection.py accordingly but it gives this errors:

--------------------------------------------------------------------------- RuntimeError Traceback (most recent call last) Input In [3], in <cell line: 8>() 5 mp_face_detection = mp.solutions.face_detection 7 #set up the face detection function by selecting the full-range model, for the short-range model use 'model_selection=0' ----> 8 face_detector = mp_face_detection.FaceDetection(model_selection=1, min_detection_confidence = 0.6) File D:\MediaPipe\mediapipe\lib\site-packages\mediapipe\python\solutions\face_detection.py:89, in FaceDetection.__init__(self, min_detection_confidence, model_selection) 86 binary_graph_path = _FULL_RANGE_GRAPH_FILE_PATH if model_selection == 1 else _SHORT_RANGE_GRAPH_FILE_PATH 87 subgraph_name = 'facedetectionfullrangecommon' if model_selection == 1 else 'facedetectionshortrangecommon' ---> 89 super().__init__( 90 binary_graph_path=binary_graph_path, 91 calculator_params={ 92 subgraph_name + '__TensorsToDetectionsCalculator.min_score_thresh': 93 min_detection_confidence, 94 }, 95 outputs=['detections']) File D:\MediaPipe\mediapipe\lib\site-packages\mediapipe\python\solution_base.py:253, in SolutionBase.__init__(self, binary_graph_path, graph_config, calculator_params, side_inputs, outputs, stream_type_hints) 251 validated_graph = validated_graph_config.ValidatedGraphConfig() 252 if binary_graph_path: --> 253 validated_graph.initialize( 254 binary_graph_path=os.path.join(root_path, binary_graph_path)) 255 else: 256 validated_graph.initialize(graph_config=graph_config) RuntimeError: Failed to parse the binary graph: D:\MediaPipe\mediapipe\lib\site-packages\mediapipe/modules/face_detection/face_detection_full_range_gpu.pbtxt

I have attached the face_detection.py, MediaPipe.ipynb, GPU models files to this message. Could you please help me with this issue? Thank you!