Caffe-ShuffleNet
Caffe-ShuffleNet copied to clipboard
Why is temp_blob_ necessary?
I noticed that the current shuffle operation use temp_blob_
as the intermediate variable for memory copy. Why can't we just set top_data as the "output" argument of Resize_gpu
?
This is the first version. You can do that
when i running there is always has problem as follows: [math_functions.cpp:91] Check failed: error == cudaSuccess (11 vs. 0) invalid argument.
I think this is caused by caffe_copy(). Could you help me solve this problem?
paste more error information