网站开发不让搜索引擎,广告设计找工作,平顶山网站建设,商标设计与制作在riscv编译安装飞桨PaddlePaddle参见#xff1a;
算能RISC-V通用云编译飞桨paddlepaddleopenKylin留档_在riscv下进行paddlelite源码编译-CSDN博客
安装好飞桨#xff0c;就可以用飞桨进行推理了。刚开始计划用ONNX推理#xff0c;但是在算能云没有装上#xff0c;所以最…在riscv编译安装飞桨PaddlePaddle参见
算能RISC-V通用云编译飞桨paddlepaddleopenKylin留档_在riscv下进行paddlelite源码编译-CSDN博客
安装好飞桨就可以用飞桨进行推理了。刚开始计划用ONNX推理但是在算能云没有装上所以最终是用的飞桨推理。但是还是用到了Paddle2ONNX里的代码。 下载Paddle2ONNX源代码
git clone https://github.com/PaddlePaddle/Paddle2ONNX ocr文字识别
下面是使用ONNX进行推理的示例我没有弄成希望有成功的朋友交流下经验。
安装需要的库
pip install shapely pyclipper
如果在riscv系统无法编译安装shapely那么可以先安装包
apt install libgeos-dev
下载相关模型文件 cd Paddle2ONNX/model_zoo/ocr# 下载det模型
wget https://bj.bcebos.com/paddle2onnx/model_zoo/ch_PP-OCRv2_det_infer.onnx# 下载rec模型
wget https://bj.bcebos.com/paddle2onnx/model_zoo/ch_PP-OCRv2_rec_infer.onnx# 下载cls模型
wget https://bj.bcebos.com/paddle2onnx/model_zoo/ch_ppocr_mobile_v2.0_cls_infer.onnx
进行识别
python3 infer.py \
--det_model_dir./ch_PP-OCRv2_det_infer.onnx \
--rec_model_dir./ch_PP-OCRv2_rec_infer.onnx \
--cls_model_dir./ch_ppocr_mobile_v2.0_cls_infer.onnx \
--image_path./images/lite_demo.png
使用飞桨推理进行文字识别
这个推理成功了。
下载和解压模型
wget -nc -P ./inference https://bj.bcebos.com/paddle2onnx/model_zoo/ch_PP-OCRv2_det_infer.tar
cd ./inference tar xf ch_PP-OCRv2_det_infer.tar cd ..wget -nc -P ./inference https://bj.bcebos.com/paddle2onnx/model_zoo/ch_PP-OCRv2_rec_infer.tar
cd ./inference tar xf ch_PP-OCRv2_rec_infer.tar cd ..wget -nc -P ./inference https://bj.bcebos.com/paddle2onnx/model_zoo/ch_ppocr_mobile_v2.0_cls_infer.tar
cd ./inference tar xf ch_ppocr_mobile_v2.0_cls_infer.tar cd ..进行文字识别
python3 infer.py \
--cls_model_dir./inference/ch_ppocr_mobile_v2.0_cls_infer \
--rec_model_dir./inference/ch_PP-OCRv2_rec_infer \
--det_model_dir./inference/ch_PP-OCRv2_det_infer \
--image_path./images/lite_demo.png \
--use_paddle_predictTrue
推理识别结果
输出可紧致头发磷层从而达到0.99444813即时持久改善头发光泽的效果给干燥的头0.99019814发足够的滋养0.997668花费了0.457335秒
效果还不错 调试
ocr读取文件的时候报错
Python 3.11.0rc2 (main, Sep 4 2023, 07:22:49) [GCC 9.3.0] on linux Type help, copyright, credits or license for more information. import cv2 Traceback (most recent call last): File stdin, line 1, in module File /root/py311/lib/python3.11/site-packages/cv2/__init__.py, line 8, in module from .cv2 import * ImportError: libtiff.so.6: cannot open shared object file: No such file or directory
安装libtiff
pip install pylibtiff
结果还是报错
找到libtiff.so.5文件在/lib/riscv64-linux-gnu做了个链接。 ln -s libtiff.so.5 libtiff.so.6
报错ImportError: libgtk-x11-2.0.so.0
SyntaxError: invalid syntax (py311) root863c89a419ec:~/github/Paddle2ONNX/model_zoo/ocr# python -c import cv2 Traceback (most recent call last): File string, line 1, in module File /root/py311/lib/python3.11/site-packages/cv2/__init__.py, line 8, in module from .cv2 import * ImportError: libtiff.so.6: cannot open shared object file: No such file or directory (py311) root863c89a419ec:~/github/Paddle2ONNX/model_zoo/ocr# python -c import cv2 Traceback (most recent call last): File string, line 1, in module File /root/py311/lib/python3.11/site-packages/cv2/__init__.py, line 8, in module from .cv2 import * ImportError: libgtk-x11-2.0.so.0: cannot open shared object file: No such file or directory
安装包apt-get install libgtk2.0-dev
ok了
识别是numpy.int报错 File /root/github/Paddle2ONNX/model_zoo/ocr/utils/predict_det.py, line 210, in box_score_fast xmin np.clip(np.floor(box[:, 0].min()).astype(np.int), 0, w - 1)
File /usr/local/lib/python3.8/dist-packages/numpy/__init__.py, line 305, in __getattr__ raise AttributeError(__former_attrs__[attr])
AttributeError: module numpy has no attribute int. np.int was a deprecated alias for the builtin int. To avoid this error in existing code, use int by itself. Doing this will not modify any behavior and is safe. When replacing np.int, you may wish to use e.g. np.int64 or np.int32 to specify the precision. If you wish to review your current use, check the release note link for additional information. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at:
进入/root/github/Paddle2ONNX/model_zoo/ocr/utils/predict_det.py源码把np.int都改成np.int32
