search

hpi-xnor / BMXNet

(New version is out: https://github.com/hpi-xnor/BMXNet-v2) BMXNet: An Open-Source Binary Neural Network Implementation Based on MXNet
Apache License 2.0
349 stars 95 forks source link

Cannot Build with USE_DIST_KVSTORE=ON #19

Closed ehsanmok closed 5 years ago

ehsanmok commented 6 years ago

Environment info

Operating System: Ubuntu 16.04 gcc, g++ 5.4.0 protoc 3.5.0

Built protobuf from the source: https://github.com/google/protobuf/tree/master/src and have all the required env variables:

PROTOBUF_INCLUDE_DIR="/usr/include/" PROTOBUF_LIBRARY="/usr/lib/libprotobuf.so" PROTOBUF_LIBRARY_DEBUG="/usr/lib/libprotobuf.so" PROTOBUF_LITE_LIBRARY="/usr/lib/libprotobuf-lite.so"
PROTOBUF_LITE_LIBRARY_DEBUG="/usr/lib/libprotobuf-lite.so"
PROTOBUF_PROTOC_EXECUTABLE="/usr/bin/protoc" PROTOBUF_PROTOC_LIBRARY="/usr/lib/libprotoc.so" PROTOBUF_PROTOC_LIBRARY_DEBUG="/usr/lib/libprotoc.so"

Error Message:

You have called ADD_LIBRARY for library mxnet without any source files. This typically indicates a problem with your CMakeLists.txt file

CMake Error at CMakeLists.txt:471 (target_link_libraries): The "debug" argument must be followed by a library.

Minimum reproducible example

USE_CUDA=0, USE_OPENCV=1 and USE_DIST_KVSTORE=1 And when I remove that line in CMakeList.txt the tests fails! Any idea what might be causing the problem?

This was also reported here with no solution!

ehsanmok commented 6 years ago

I could build mxnet==0.12.1 from source with USE_CUDA=0 USE_CUDNN=0 and USE_DIST_KVSTORE=1 though. I figured out my main CMakefileList.txt was changed and I cloned a new one. Also had some proxy issue for building ps-lite/make/dep.mk which I resolved. I know this issue might not be very related to bmxnet specific feature, but I want to know if you are able to build distributed bmxnet. Thanks!

yanghaojin commented 6 years ago

Hi ehsanmok, the current BMXNet version is based mxnet==0.1.0 and CUDA 8.0. We are working on the upgrade to mxnet1.0, will release it soon, stay tuned. HJ

ehsanmok commented 6 years ago

Hi @yanghaojin Do you know of any time estimate when v1.0 will be released? The current version is too much of headache.

yanghaojin commented 6 years ago

Hi, I hope the upgrade could be finished in the next week. We are working on the model testing. After the upgrade we can take advantages of cudnn to train a BNN, so this could largely improve the training speed.

ehsanmok commented 6 years ago

@yanghaojin that's great to hear! looking forward to it.