使用PaddleOCR，读取数据很慢，但是用PaddleOCRSharp就很快

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem

系统环境/System Environment：win10

代码： 'ocr = PaddleOCR(use_angle_cls=True, use_gpu=True,lang="ch") start_time = time.time() result = ocr.ocr(path) end_time = time.time() execution_time = end_time - start_time print("Execution time:", execution_time, "seconds")' 执行结果 '[2024/04/30 14:36:13] ppocr DEBUG: dt_boxes num : 52, elapse : 0.46878957748413086 [2024/04/30 14:36:14] ppocr DEBUG: cls num : 52, elapse : 0.9789741039276123 [2024/04/30 14:36:29] ppocr DEBUG: rec_res num : 52, elapse : 14.643944025039673 Execution time: 16.11536145210266 seconds'

发现执行 rec_res 这里速度特别的慢

然后是PaddleOCRSharp 输出是平均时间: 0:00:03.898628

都是同一台机器，不知道为什么，难道是我依赖太多的问题?求大佬解答这是我的依赖 'C:\Users\Administrator>pip list Package Version

aiohttp 3.8.6 aiosignal 1.3.1 altgraph 0.17.4 annotated-types 0.5.0 anyio 3.7.1 astor 0.8.1 async-timeout 4.0.3 asynctest 0.13.0 attrdict 2.0.1 attrdict3 2.0.2 attrs 23.2.0 auto-py-to-exe 2.43.3 Babel 2.14.0 bce-python-sdk 0.9.7 beautifulsoup4 4.12.3 bottle 0.12.25 bottle-websocket 0.2.9 cachetools 5.3.3 certifi 2024.2.2 cffi 1.15.1 charset-normalizer 3.3.2 ci-info 0.3.0 click 8.1.7 colorama 0.4.6 coloredlogs 15.0.1 colorlog 6.8.2 configobj 5.0.8 configparser 5.3.0 cssselect 1.2.0 cssutils 2.7.1 cycler 0.11.0 Cython 3.0.10 datasets 2.13.2 decorator 5.1.1 dill 0.3.4 Eel 0.16.0 et-xmlfile 1.1.0 etelemetry 0.3.1 exceptiongroup 1.2.1 fastapi 0.103.2 filelock 3.12.2 fire 0.6.0 fitz 0.0.1.dev2 Flask 2.2.5 flask-babel 3.1.0 Flask-Cors 4.0.0 flatbuffers 24.3.25 fonttools 4.38.0 frozenlist 1.3.3 fsspec 2023.1.0 future 1.0.0 gevent 22.10.2 gevent-websocket 0.10.1 greenlet 3.0.3 h11 0.14.0 httpcore 0.17.3 httplib2 0.22.0 httpx 0.24.1 huggingface-hub 0.16.4 humanfriendly 10.0 idna 3.7 imageio 2.31.2 imgaug 0.4.0 importlib-metadata 4.13.0 isodate 0.6.1 itsdangerous 2.1.2 jieba 0.42.1 Jinja2 3.1.3 joblib 1.3.2 kiwisolver 1.4.5 lmdb 1.4.1 looseversion 1.3.0 lxml 5.2.1 markdown-it-py 2.2.0 MarkupSafe 2.1.5 matplotlib 3.5.3 mdurl 0.1.2 mpmath 1.3.0 multidict 6.0.5 multiprocess 0.70.12.2 networkx 2.6.3 nibabel 4.0.2 nipype 1.8.6 numpy 1.21.6 onnx 1.14.1 onnxruntime 1.14.1 opencv-contrib-python 4.6.0.66 opencv-python 4.6.0.66 opencv-python-headless 4.9.0.80 openpyxl 3.1.2 opt-einsum 3.3.0 packaging 24.0 paddle-bfloat 0.1.7 paddle2onnx 1.0.6 paddlefsl 1.1.0 paddlenlp 2.6.1 paddleocr 2.7.0.2 paddlepaddle 2.5.2 pandas 1.3.5 pdf2docx 0.5.8 pefile 2023.2.7 Pillow 8.3.2 pip 24.0 premailer 3.10.0 protobuf 3.20.2 prov 2.0.0 psutil 5.9.8 pyarrow 12.0.1 pyclipper 1.3.0.post5 pycparser 2.21 pycryptodome 3.20.0 pydantic 2.5.3 pydantic_core 2.14.6 pydot 2.0.0 Pygments 2.17.2 pyinstaller 5.13.2 pyinstaller-hooks-contrib 2024.5 PyMuPDF 1.20.2 pypandoc 1.13 pyparsing 3.1.2 pyreadline 2.1 python-dateutil 2.9.0.post0 python-docx 1.1.0 pytz 2024.1 PyWavelets 1.3.0 pywin32-ctypes 0.2.2 pyxnat 1.6.2 PyYAML 6.0.1 rapidfuzz 3.4.0 rarfile 4.2 rdflib 6.3.2 requests 2.31.0 rich 13.7.1 safetensors 0.4.3 scikit-image 0.19.3 scikit-learn 1.0.2 scipy 1.7.3 sentencepiece 0.2.0 seqeval 1.2.2 setuptools 68.0.0 shapely 2.0.4 shellingham 1.5.4 simplejson 3.19.2 six 1.16.0 sniffio 1.3.1 soupsieve 2.4.1 starlette 0.27.0 sympy 1.10.1 termcolor 2.3.0 threadpoolctl 3.1.0 tifffile 2021.11.2 tqdm 4.66.2 traits 6.3.2 typer 0.12.3 typing_extensions 4.7.1 urllib3 1.25.11 uvicorn 0.22.0 visualdl 2.5.3 Werkzeug 2.2.3 whichcraft 0.6.1 xxhash 3.4.1 yacs 0.1.8 yarl 1.9.4 zipp 3.15.0 zope.event 5.0 zope.interface 6.3'

PaddlePaddle / PaddleOCR

使用PaddleOCR，读取数据很慢，但是用PaddleOCRSharp就很快 #12031