search

eric612 / MobileNet-YOLO

A caffe implementation of MobileNet-YOLO detection network
Other
865 stars 442 forks source link

make error #175

Open EvaneSnow opened 5 years ago

EvaneSnow commented 5 years ago

@eric612 when I make the project, and it has two error:

/home/snow/MobileNet-YOLO-master/src/caffe/util/io.cpp: In function ‘bool caffe::ReadJSONToAnnotatedDatum(const string&, int, int, caffe::AnnotatedDatum*)’: /home/snow/MobileNet-YOLO-master/src/caffe/util/io.cpp:525:17: warning: unused variable ‘anno’ [-Wunused-variable] Annotation* anno = NULL; ^ [ 83%] Linking CXX shared library ../../lib/libcaffe.so /usr/bin/ld: /usr/local/lib/libgflags.a(gflags.cc.o): relocation R_X86_64_32 against.rodata.str1.1' can not be used when making a shared object; recompile with -fPIC /usr/local/lib/libgflags.a: 无法添加符号: 错误的值 collect2: error: ld returned 1 exit status src/caffe/CMakeFiles/caffe.dir/build.make:4122: recipe for target 'lib/libcaffe.so.1.0.0' failed make[2]: [lib/libcaffe.so.1.0.0] Error 1 CMakeFiles/Makefile2:304: recipe for target 'src/caffe/CMakeFiles/caffe.dir/all' failed make[1]: [src/caffe/CMakeFiles/caffe.dir/all] Error 2 Makefile:127: recipe for target 'all' failed make: *** [all] Error 2 ` My opencv is 3.3.1 and 3.4.5; thanks!

daixia commented 5 years ago

hi ,Have you solved the problem，I met the same problem。Can you help me?? thx

EvaneSnow commented 5 years ago

@daixia what's your opencv version? opencv need to be 3.3.1

daixia commented 5 years ago

Must be 3.3.1？3.4.1 is it OK？

EvaneSnow commented 5 years ago

I don't know, when My opencv is 3.4.5, the make is error, so i uninstall opencv3.4.5,and install opencv3.3.1

daixia commented 5 years ago

ok ,thank you,

eric612 commented 5 years ago

In docker build , I use opencv 4.x version and ubuntu18 , and it work fine https://hub.docker.com/r/eric612/mobilenet-yolo

EvaneSnow commented 5 years ago

you can try it

EvaneSnow commented 5 years ago

@eric612 你好，我有一些问题，请问你有时间吗？谢谢

daixia commented 5 years ago

thank you so much

EvaneSnow commented 5 years ago

@eric612 我想用scripts 中的train_pelee_ssd_bdd100k.sh脚本训练pelee网络，下面是我训练记录： 我使用了bdd100k数据集中的1000张图片资料（1000张原图.jpg，1000张drivable_maps图片.png，还有1000个jpg对应的xml文件）。 使用create_list.sh脚本生成train.txt文件，然后用create_data.sh脚本生成lmbd数据（在这一步中，出现了一些问题，.prototxt文件只能使用labelmap_detection.prototxt，而不能使用labelmap.prototxt文件，否则会出错）。 在训练中，train.txt中的.prototxt文件只能使用labelmap.prototxt文件，可是训练出来的结果却没有目标检测，只有分割，请问我需要怎么修改，谢谢