Go binding for MXNet C predict API. This is used by the MXNet agent in MLModelScope to perform model inference in Go.

Download and install go-mxnet:

go get -v github.com/rai-project/go-mxnet

The binding requires MXNet and other Go packages.

The MXNet C library is expected to be under /opt/mxnet.

To install MXNet on your system, you can follow the MXNet documentation, or refer to our scripts or the LIBRARY INSTALLATION section in the dockefiles. OpenBLAS is used in our default build.

• The default blas is OpenBLAS. The default OpenBLAS path for macOS is /usr/local/opt/openblas if installed throught homebrew (openblas is keg-only, which means it was not symlinked into /usr/local, because macOS provides BLAS and LAPACK in the Accelerate framework).

• The default mxnet installation path is /opt/mxnet for linux, darwin and ppc64le without powerai; /opt/DL/mxnet for ppc64le with powerai.

• The default CUDA path is /usr/local/cuda

See lib.go for details.

If you get an error about not being able to write to /opt then perform the following

sudo mkdir -p /opt/mxnet
sudo chown -R `whoami` /opt/mxnet

If you are using MXNet docker images or other libary paths, change CGO_CFLAGS, CGO_CXXFLAGS and CGO_LDFLAGS enviroment variables. Refer to Using cgo with the go command.

For example,

    export CGO_CFLAGS="${CGO_CFLAGS} -I/tmp/mxnet/include"
    export CGO_CXXFLAGS="${CGO_CXXFLAGS} -I/tmp/mxnet/include"
    export CGO_LDFLAGS="${CGO_LDFLAGS} -L/tmp/mxnet/lib"

You can install the dependency through go get.

cd $GOPATH/src/github.com/rai-project/tensorflow
go get -u -v ./...

Or use Dep.

dep ensure -v

This installs the dependency in vendor/.

Configure the linker environmental variables since the MXNet C library is under a non-system directory. Place the following in either your ~/.bashrc or ~/.zshrc file

Linux

export LIBRARY_PATH=$LIBRARY_PATH:/opt/mxnet/lib
export LD_LIBRARY_PATH=/opt/mxnet/lib:$DYLD_LIBRARY_PATH

macOS

export LIBRARY_PATH=$LIBRARY_PATH:/opt/mxnet/lib
export DYLD_LIBRARY_PATH=/opt/mxnet/lib:$DYLD_LIBRARY_PATH

Run go build in to check the dependences installation and library paths set-up. On linux, the default is to use GPU, if you don't have a GPU, do go build -tags nogpu instead of go build.

Note : The CGO interface passes go pointers to the C API. This is an error by the CGO runtime. Disable the error by placing

export GODEBUG=cgocheck=0

in your ~/.bashrc or ~/.zshrc file and then run either source ~/.bashrc or source ~/.zshrc

Examples of using the Go MXNet binding to do model inference are under examples.

This example shows how to use the MLModelScope tracer to profile the inference.

Refer to Set up the external services to start the tracer.

Then run the example by

  cd example/batch_mlmodelscope
  go build
  ./batch

Now you can go to localhost:16686 to look at the trace of that inference.

This example shows how to use nvprof to profile the inference. You need GPU and CUDA to run this example.

  cd example/batch_nvprof
  go build
  nvprof --profile-from-start off ./batch_nvprof

Refer to Profiler User's Guide for using nvprof.