ax-samples
ax-samples copied to clipboard
Samples code for world class Artificial Intelligence SoCs for computer vision applications.
送板子嘛?
想要适配N多模型
BEV算法部署
有没有关于BEV算法的部署教程
Hi, I have build AX650A and running ax_yolov5s using below command. ./ax_yolov5s -m yolov5s.axmodel -i cat.jpg on CPU. and i got below error. -------------------------------------- model file : /home/sagarkhimani/Downloads/AXERA/650/yolov5s.joint image file...
make Consolidate compiler generated dependencies of target ax_imgproc [ 1%] Building CXX object examples/ax650/CMakeFiles/ax_imgproc.dir/ax_imgproc_steps.cc.o In file included from /opt/rh/devtoolset-7/root/usr/include/c++/7/bits/stl_iterator_base_types.h:67, from /opt/rh/devtoolset-7/root/usr/include/c++/7/numeric:61, from /work/home/ysuanap86/project/ax-samples/examples/ax650/ax_imgproc_steps.cc:23: /opt/rh/devtoolset-7/root/usr/include/c++/7/type_traits:347:39: error: ‘__float128’ was not declared in...
Hi, I created one python file and placed on AXERA AX620A device and i'm importing opencv in python file and want to access device camera (os04a10) using cv2.VideoCapture() and want...
Hello @ltdat1201, @shitouren1994 , we are trying to convert Arcface ONNX to joint format to use it on AX620 chip. The conversion for both w600k_R50 & w600k_mbf were successful, however,...
[ 1%] Building CXX object examples/ax650/CMakeFiles/ax_model_info.dir/ax_model_info.cc.o In file included from /home/wzhpc/wjw/github_file/AXERA_DEMO/ax-samples/examples/ax650/ax_model_info.cc:27:0: /home/wzhpc/wjw/github_file/AXERA_DEMO/ax-samples/examples/ax650/middleware/io.hpp:29:10: fatal error: ax_engine_api.h: No such file or directory #include ^~~~~~~~~~~~~~~~~ compilation terminated. make[2]: *** [examples/ax650/CMakeFiles/ax_model_info.dir/build.make:63: examples/ax650/CMakeFiles/ax_model_info.dir/ax_model_info.cc.o] Error 1...
从 https://dl.sipeed.com/shareURL/MaixIV/M4N-Dock 下载的1.45.0版本SDK的文档中, [00 - AX SDK 使用说明.pdf](https://github.com/user-attachments/files/17246750/00.-.AX.SDK.pdf) 00 - AX SDK 使用说明的第4.4节,如何调整分区大小 ,这部分给出的方法中,  这一段代码,在ax650_common.h里已经没有了,所以无法修改EMMC启动的各分区大小 请求解决
按照自定义数据集进行了修改,并按照子图的切法进行了操作,为什么训练精度在85%左右,但是在板子上跑的时候就非常的低,低到10%左右,甚至根本检测不出来?有人遇到这种问题吗?是什么原因导致的呢?