Closed jwnsu closed 7 years ago
Hi, can you successfully build the caffe-fast-rcnn of rbgirshick's py-fast-rcnn? I only mdified the eltwise layer and batchnorm layer in caffe-fast-rcnn.
Saw you only made 3 file changes. So I copied over a working caffe-fast-rcnn, then delta the changes. It's resolved now.
Model file from pan.baidu.com takes forever to download in US though, poor baidu service.
@jwnsu Add OneDrive download link in README.
Thanks! Got models in just seconds from onedrive.
Following @jwnsu 's way, I copied eltwise_layer. and batch_norm_layer. from caffe-fast-rcnn/src/caffe/layers/ to a working caffe-fast-rcnn. Now caffe compiles smoothly.
Env: Ubuntu 16.04 with latest CUDA 8.0 cuDNN 5.1
Copied Makefile.config from working py-faster-rcnn directory. make reports following error and stops:
CXX src/caffe/data_reader.cpp In file included from ./include/caffe/util/device_alternate.hpp:40:0, from ./include/caffe/common.hpp:19, from src/caffe/data_reader.cpp:6: ./include/caffe/util/cudnn.hpp: In function ‘void caffe::cudnn::createPoolingDesc(cudnnPoolingStruct, caffe::PoolingParameter_PoolMethod, cudnnPoolingMode_t, int, int, int, int, int, int)’: ./include/caffe/util/cudnn.hpp:127:41: error: too few arguments to function ‘cudnnStatus_t cudnnSetPooling2dDescriptor(cudnnPoolingDescriptor_t, cudnnPoolingMode_t, cudnnNanPropagation_t, int, int, int, int, int, int)’ pad_h, pad_w, stride_h, stride_w)); ^ ./include/caffe/util/cudnn.hpp:15:28: note: in definition of macro ‘CUDNN_CHECK’ cudnnStatus_t status = condition; \ ^ In file included from ./include/caffe/util/cudnn.hpp:5:0, from ./include/caffe/util/device_alternate.hpp:40, from ./include/caffe/common.hpp:19, from src/caffe/data_reader.cpp:6: /usr/local/cuda/include/cudnn.h:803:27: note: declared here cudnnStatus_t CUDNNWINAPI cudnnSetPooling2dDescriptor( ^ Makefile:563: recipe for target '.build_release/src/caffe/data_reader.o' failed make: [.build_release/src/caffe/data_reader.o] Error 1