daybreak-u / solov2_ncnn Goto Github PK
View Code? Open in Web Editor NEWthe C++ version of solov2 with ncnn
the C++ version of solov2 with ncnn
Can you provide solov2_head.py
您好,我在make的时候出现了下面的错误,请问您知道什么原因吗?
solov2.cpp:(.text+0x6c1): undefined reference to `ncnn::create_layer(char const*)'
/usr/bin/ld: solov2.cpp:(.text+0x6d7): undefined reference to `ncnn::ParamDict::ParamDict()'
/usr/bin/ld: solov2.cpp:(.text+0x6f1): undefined reference to `ncnn::ParamDict::set(int, int)'
/usr/bin/ld: solov2.cpp:(.text+0x70a): undefined reference to `ncnn::ParamDict::set(int, int)'
hi!
Have you tried tensorrt to transform the model into TRT file and then predict
感谢这么好的进展!如何才能训练自己的数据集呢?
大佬 能不能把你改过的solov2_head.py 发一下 我修改的总报错 谢谢
What copyright license are you releasing this under?
A natural choice would be to use the same as the library you build on. ncnn is under the BSD 3-clause license (excluding some dependencies it ships with that are also under permissive licenses):
https://github.com/Tencent/ncnn/blob/master/LICENSE.txt
Thanks!
I can not download the models because Baidu wants a valid account. Something you can not get in most countries. You have only a limited list of overseas telephone numbers.
Can you provide us an alternative download route?
hello there,
i was trying this code to train my custom dataset with this code :
https://github.com/aim-uofa/AdelaiDet
the problem appears when i was trying to convert the model to onnx then ncnn
it would be very grateful if u can show me the progress or some enlightenment
hope for your response
请问大佬,我用自制单分类数据集完成训练后,按照教程转换onnx模型. 出现以下错误:
File "/SOLO/mmdet/models/anchor_heads/solov2_head.py", line 139, in forward
new_feats[0] = torch.cat([new_feats[0], p3_input], dim = 1)
TypeError: 'tuple' object does not support item assignment
求教是怎么回事
Error happen in make, I am using opencv 4.2.0, Ubuntu 20.04
[ 25%] Linking CXX executable solov2_fast
/usr/bin/ld: CMakeFiles/solov2_fast.dir/src/solov2_fast.cpp.o: in function `ins_decode(ncnn::Mat const&, ncnn::Mat const&, std::vector<int, std::allocator<int> >&, std::map<int, int, std::less<int>, std::allocator<std::pair<int const, int> > >&, ncnn::Mat*, int, ncnn::Option&)':
solov2_fast.cpp:(.text+0x7a7): undefined reference to `ncnn::create_layer(char const*)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x7bd): undefined reference to `ncnn::ParamDict::ParamDict()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x7d7): undefined reference to `ncnn::ParamDict::set(int, int)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x7f0): undefined reference to `ncnn::ParamDict::set(int, int)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x813): undefined reference to `ncnn::ParamDict::set(int, int)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x9f5): undefined reference to `ncnn::ModelBinFromMatArray::ModelBinFromMatArray(ncnn::Mat const*)'
/usr/bin/ld: solov2_fast.cpp:(.text+0xa1a): undefined reference to `ncnn::ModelBinFromMatArray::~ModelBinFromMatArray()'
/usr/bin/ld: solov2_fast.cpp:(.text+0xb1b): undefined reference to `ncnn::ParamDict::~ParamDict()'
/usr/bin/ld: solov2_fast.cpp:(.text+0xb59): undefined reference to `ncnn::ModelBinFromMatArray::~ModelBinFromMatArray()'
/usr/bin/ld: solov2_fast.cpp:(.text+0xbba): undefined reference to `ncnn::ParamDict::~ParamDict()'
/usr/bin/ld: CMakeFiles/solov2_fast.dir/src/solov2_fast.cpp.o: in function `generate_res(ncnn::Mat&, ncnn::Mat&, std::map<int, int, std::less<int>, std::allocator<std::pair<int const, int> > >&, std::vector<std::vector<Object, std::allocator<Object> >, std::allocator<std::vector<Object, std::allocator<Object> > > >&, float, float, int, int, int, float, int, int)':
solov2_fast.cpp:(.text+0xf5d): undefined reference to `cv::Mat::Mat(int, int, int)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x10f2): undefined reference to `cv::Mat::Mat()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x11b8): undefined reference to `cv::Mat::Mat()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x127e): undefined reference to `cv::Mat::Mat(int, int, int)'
/usr/bin/ld: CMakeFiles/solov2_fast.dir/src/solov2_fast.cpp.o: in function `detect_solov2(cv::Mat const&, std::vector<Object, std::allocator<Object> >&)':
solov2_fast.cpp:(.text+0x1674): undefined reference to `ncnn::Net::Net()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x169c): undefined reference to `ncnn::Net::load_param(char const*)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x16af): undefined reference to `ncnn::Net::load_model(char const*)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x1863): undefined reference to `ncnn::Mat::from_pixels_resize(unsigned char const*, int, int, int, int, int, ncnn::Allocator*)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x18db): undefined reference to `ncnn::Option::Option()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x1966): undefined reference to `ncnn::copy_make_border(ncnn::Mat const&, ncnn::Mat&, int, int, int, int, int, float, ncnn::Option const&)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x1a35): undefined reference to `ncnn::Mat::substract_mean_normalize(float const*, float const*)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x20ac): undefined reference to `ncnn::Net::create_extractor() const'
/usr/bin/ld: solov2_fast.cpp:(.text+0x20c9): undefined reference to `ncnn::Extractor::input(char const*, ncnn::Mat const&)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x20e6): undefined reference to `ncnn::Extractor::input(char const*, ncnn::Mat const&)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x2103): undefined reference to `ncnn::Extractor::input(char const*, ncnn::Mat const&)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x2120): undefined reference to `ncnn::Extractor::input(char const*, ncnn::Mat const&)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x21e7): undefined reference to `ncnn::Extractor::extract(char const*, ncnn::Mat&, int)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x2209): undefined reference to `ncnn::Extractor::extract(char const*, ncnn::Mat&, int)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x222b): undefined reference to `ncnn::Extractor::extract(char const*, ncnn::Mat&, int)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x224d): undefined reference to `ncnn::Extractor::extract(char const*, ncnn::Mat&, int)'
/usr/bin/ld: solov2_fast.cpp:(.text+0x226f): undefined reference to `ncnn::Extractor::extract(char const*, ncnn::Mat&, int)'
/usr/bin/ld: CMakeFiles/solov2_fast.dir/src/solov2_fast.cpp.o:solov2_fast.cpp:(.text+0x2291): more undefined references to `ncnn::Extractor::extract(char const*, ncnn::Mat&, int)' follow
/usr/bin/ld: CMakeFiles/solov2_fast.dir/src/solov2_fast.cpp.o: in function `detect_solov2(cv::Mat const&, std::vector<Object, std::allocator<Object> >&)':
solov2_fast.cpp:(.text+0x2356): undefined reference to `ncnn::Option::Option()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x2c7c): undefined reference to `ncnn::Extractor::~Extractor()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x2cd3): undefined reference to `ncnn::Net::~Net()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x2ed4): undefined reference to `ncnn::Extractor::~Extractor()'
/usr/bin/ld: solov2_fast.cpp:(.text+0x2f46): undefined reference to `ncnn::Net::~Net()'
/usr/bin/ld: CMakeFiles/solov2_fast.dir/src/solov2_fast.cpp.o: in function `Object::Object()':
solov2_fast.cpp:(.text._ZN6ObjectC2Ev[_ZN6ObjectC5Ev]+0x1c): undefined reference to `cv::Mat::Mat()'
collect2: error: ld returned 1 exit status
make[2]: *** [CMakeFiles/solov2_fast.dir/build.make:98: solov2_fast] Error 1
make[1]: *** [CMakeFiles/Makefile2:78: CMakeFiles/solov2_fast.dir/all] Error 2
make: *** [Makefile:84: all] Error 2
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.