⚡️An Easy-to-use and Fast Deep Learning Model Deployment Toolkit for ☁️Cloud 📱Mobile and 📹Edge. Including Image, Video, Text and Audio 20+ main stream scenarios and 150+ SOTA models with end-to-end optimization, multi-platform and multi-framework support.
【模型跑不通】
ocr多语言检测openvino和onnx两个引擎在cpu上报错
模型文件：
https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_infer.tar
使用测试代码
https://github.com/PaddlePaddle/FastDeploy/blob/develop/examples/vision/ocr/PP-OCR/cpu-gpu/python/infer.py
调用参数：
--det_model Multilingual_PP-OCRv3_det_infer --cls_model ch_ppocr_mobile_v2.0_cls_infer --rec_model ch_PP-OCRv3_rec_infer --rec_label_file ppocr_keys_v1.txt --image 12.jpg --device cpu --backend openvino
报错：
Traceback (most recent call last):
File "/usr/lib/python3.8/runpy.py", line 194, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/usr/lib/python3.8/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "/root/.vscode-server/extensions/ms-python.python-2023.16.0/pythonFiles/lib/python/debugpy/adapter/../../debugpy/launcher/../../debugpy/main.py", line 39, in
cli.main()
File "/root/.vscode-server/extensions/ms-python.python-2023.16.0/pythonFiles/lib/python/debugpy/adapter/../../debugpy/launcher/../../debugpy/../debugpy/server/cli.py", line 430, in main
run()
File "/root/.vscode-server/extensions/ms-python.python-2023.16.0/pythonFiles/lib/python/debugpy/adapter/../../debugpy/launcher/../../debugpy/../debugpy/server/cli.py", line 284, in run_file
runpy.run_path(target, run_name="main")
File "/root/.vscode-server/extensions/ms-python.python-2023.16.0/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 321, in run_path
return _run_module_code(code, init_globals, run_name,
File "/root/.vscode-server/extensions/ms-python.python-2023.16.0/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 135, in _run_module_code
_run_code(code, mod_globals, init_globals,
File "/root/.vscode-server/extensions/ms-python.python-2023.16.0/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 124, in _run_code
exec(code, run_globals)
File "/FastDeploy/examples/vision/ocr/PP-OCR/cpu-gpu/python/infer.py", line 211, in
result = ppocrv3.predict(im)
File "/usr/local/lib/python3.8/dist-packages/fastdeploy/vision/ocr/ppocr/init.py", line 966, in predict
return self.system.predict(input_image)
RuntimeError: Can't set input blob with name: x, because model input (shape=[?,3,960,960]) and blob (shape=(1.3.960.672)) are incompatible
rainyfly
commented
5 months ago
温馨提示：根据社区不完全统计，按照模板提问，可以加快回复和解决问题的速度
环境
问题日志及出现问题的操作流程