Gheng

关于ncnn在海思3531dv200上的编译

1

## detail | 详细描述 | 詳細な説明 请问，该开发板编译器为aarch64-himix200-linux-g++，该使用哪个版本的tooltrains呢，求指教 > cd mkdir -p build cd build # Choose one cmake toolchain file depends on your target platform cmake -DCMAKE_TOOLCHAIN_FILE=../toolchains/hisiv300.toolchain.cmake .. cmake -DCMAKE_TOOLCHAIN_FILE=../toolchains/hisiv500.toolchain.cmake...

ncnn cpp

2

find_blob_index_by_name input.1 failed Try ex.input("data", in0); find_blob_index_by_name 792 failed Try

Ubuntu系统make错误，使用cmake编译cpp的ncnn工程-nanodet

1

In file included from /home/linkpi/ncnnprotest/zexample/nanodet-plus/main.cpp:6:0: /home/linkpi/ncnnprotest/zexample/nanodet-plus/nanodet.h:12:1: warning: ‘typedef’ was ignored in this declaration }; ^ /home/linkpi/ncnnprotest/zexample/nanodet-plus/nanodet.h:42:34: warning: non-static data member initializers only available with -std=c++11 or -std=gnu++11 int input_size[2] =...

P6 = slim.avg_pool2d(P5, kernel_size=[1, 1], stride=2, scope='build_P6') 这个kernel_size不应该是[2, 2]么？

I find that some issue on test, such as no NMS after detection? yes or no?,and the accuarcy is low, only use the dataset of yours "data.pk" with picture. If...

Message type "caffe.LayerParameter" has no field named "psroi_pooling_param". in test.....(demorfcn.py)

when I catkin_make the project, The problem about ros shows: `/usr/bin/ld: /lib/x86_64-linux-gnu/libicui18n.so.66: undefined reference to `log@GLIBC_2.29' /usr/bin/ld: /lib/x86_64-linux-gnu/libicui18n.so.66: undefined reference to `pow@GLIBC_2.29' collect2: error: ld returned 1 exit status make[2]:...

Train on my datasets, but performance is poor.

2

 thanks for your good job, i meet a question about train on my datasets (like kitti type). the checkpoints i saved can not detect anything, but your pretrain-model is...

how to eval the model?

1

thanks for your work, but I hard to know how to eval your trained model ? When I run the code "eval_auto" and "eval_active", it stuck. can you give a...