I want to run Mediapipe hand gesture recognition in “streamlit“, i used Mediapipe for Python version, but i got this error :

RuntimeError: Unable to open file at F:pyprogrampythonlibsite-packages/F:pyprogramhandgestureFrontendgesture_recognizer.task, errno=22

Traceback:

File "F:pyprogrampythonlibsite-packagesstreamlitruntimescriptrunnerexec_code.py", line 88, in exec_func_with_error_handling
    result = func()
File "F:pyprogrampythonlibsite-packagesstreamlitruntimescriptrunnerscript_runner.py", line 579, in code_to_exec
    exec(code, module.__dict__)
File "F:pyprogramhandgestureFrontendpages10_test.py", line 29, in <module>
    with GestureRecognizer.create_from_options(options) as recognizer:
File "F:pyprogrampythonlibsite-packagesmediapipetaskspythonvisiongesture_recognizer.py", line 340, in create_from_options
    return cls(
File "F:pyprogrampythonlibsite-packagesmediapipetaskspythonvisioncorebase_vision_task_api.py", line 70, in __init__
    self._runner = _TaskRunner.create(graph_config, packet_callback)

• when i run Mediapipe’s python code (without streamlit) it works well.

• I have tested all suggested ways, like changing version of libraries(mediapipe & streamlit) , or changing “gesture_recognizer.task” path , … .

I don’t understand why is it automatically adding “F:pyprogrampythonlibsite-packages” path to the “gesture_recognizer.task” ABSOLUTE path as it is mentioned :

model_path = ‘/absolute/path/to/gesture_recognizer.task’

folder structure :

Frontend
│  
│   gesture_recognizer.task  
└── pages 
    └── 10_test.py

code :

BaseOptions = mp.tasks.BaseOptions
GestureRecognizer = mp.tasks.vision.GestureRecognizer
GestureRecognizerOptions = mp.tasks.vision.GestureRecognizerOptions
GestureRecognizerResult = mp.tasks.vision.GestureRecognizerResult
VisionRunningMode = mp.tasks.vision.RunningMode


options = GestureRecognizerOptions(
 base_options=BaseOptions(model_asset_path='F:/pyprogram/handgesture/Frontend/gesture_recognizer.task'),
    running_mode=VisionRunningMode.LIVE_STREAM,
    result_callback=print_result)

with GestureRecognizer.create_from_options(options) as recognizer:
.
.