Closed chungjin closed 8 years ago
plus: When train the model, and in the forward functioon, the matlab will also crush.
Could you please tell how to solve this error? I am getting the same issue in the code.
this is a bug in batchnorm.cu, in the _use_global_term case, may be you should try to use my caffe version git@github.com:chungjin/caffe-R-FCN.git and I have successfully compiled it and make use in linux
Hi, how do you solve this problem? I have the exactly same issue. Thanks
website is here: https://github.com/chungjin/caffe-R-FCN
because i cannot train the model, so I tried to exam the caffe first. Because you don't implement the cpu code, so I cancel the cpu test. But each time I did the forward in BatchNormLayer, it will show _Check failed: data__ detail are shown as follows:
Hope you can help me to solve this issue