huziling

> @kurapan > I trained it for about 500 epochs , and got about 0.65 hmean，not reached 0.8 as paper show。I think you can raise the score map threshold ,...

> > > > @Ocelot7777 when I use one image as training set, I come across the same problem, but I will not come across the same problem when I...

> > > > > > @Ocelot7777 when I use one image as training set, I come across the same problem, but I will not come across the same problem...

> @huziling 0.7hmean also, and have you found the reason? 我并没找到问题所在，但我怀疑可能是数据读取，模型初始化，以及pytorch和tensor flow的损失计算区别之类的原因。由于时间问题我并没有进一步寻找原因

> > > @huziling @jiangmiaotong > did you change anything in code to get to 0.7 hmean? i can't remember all details cause it's long time ago. i may change...

i uninstall opencv 4.xx and try opencv 3.4.5 and it still not work

i also try a opecv 3.1.0 and still not work . May you give me a adaptor.so. Thanks

> > > I met this too! Make sure your opencv in /usr/include/ dir or modify the Makefile to config your opencv. @ @huziling my opencv is opencv-python, and i...

Thanks for your help!!! I will try install opencv with source code

hi! 我最近下载了opencv的源码 并且make了 我下载的位置在 /mnt/3T/xxx/opencv 其中有 /mnt/3T/xxx/opencv/include 和 /mnt/3T/xxx/opencv/build/lib/ 这是是否意味着我应该修改Makefile 为： CXXFLAGS = -I include -std=c++11 -O3 -I `/mnt/3T/xxx/opencv/include` DEPS = lanms.h $(shell find include -xtype f) CXX_SOURCES = adaptor.cpp...