paddleocr遇到的一些坑

问题1:not find model file path inference/ch_PP-OCRv3_det_infer/model

解决方法:看看自己下载的文件名是不是带了develop,如果是,重新去下一个release版本的即可。

问题2:No module named ‘Polygon‘

解决方法

pip uninstall Polygon3

pip install Polygon3 -i https://pypi.tuna.tsinghua.edu.cn/simple

 问题3:Microsoft Visual C++ 14.0 is required. Get it with “Microsoft Visual C++ Build Tools

解决方法:下载visual C++ 14.0    https://zhuanlan.zhihu.com/p/165008313

问题4:ModuleNotFoundError: No module named 'lanms'

解决方法

conda install -f pip

pip install lanms-nova

问题5:The third-party dynamic library (cudnn64_8.dll) that Paddle depends on is not configured correctly.

解决方法:下载对应版本的cuda和cudann

问题6:Could not locate zlibwapi.dll. Please make sure it is in your library path

解决方法http://t.csdn.cn/AJkUR

猜你喜欢

转载自blog.csdn.net/dal1223/article/details/131060945