yunren
yunren
Thanks for your incredible work. I have a question about the shared networks . Why the RPN stage does not use deformable ConvNets in your prototxt?
目前电脑配置的显卡是2080ti,好像只能装cuda10以上的版本,目前我装的cuda10.2 但是按照步骤编译总是error,以前在vs2015+cuda9.0的环境下编译成功过。 想求助大佬,能否抽空指导一下,vs2015+cuda10.2 应该如何修改啊? 小弟先谢过了
您好,我把对应层PSROIAlign层对应的.hpp/cpp/cu文件加入到我的caffe工程里面,proto也修改了,可是一编译就出现如下错误: error: no instance of function template "caffe::PSROIAlignForward" matches the agument list 错误信息定位到 psroi_align_layer.cu的154行1列。 求助,该怎么解决,谢谢
Hi, thanks for your sharing spirit. Do you have merged OHEM into Faster RCNN for the official MATLAB version? Could you teach me how to modify the codes?
RT
大佬,您好,我按照你给的说明,并参考 happynear 的Caffe-windows的第三方库进行相应的配置,脱离NuGet,在vs2015 cuda10.2环境下编译成功。 单gpu VGG16训练没问题,但是多GPU训练时就出现]错误,麻烦您有空能够指导一下,谢谢! 
Would you mind to share your trained Xception.caffemodel ?