A package for installing the (good old) SyntaxNet (Parsey McParseface) dependency parser.
$ cd build
$ wget https://repo.anaconda.com/archive/Anaconda2-5.1.0-Linux-x86_64.sh
$ cd ..
$ bash build/Anaconda2-5.1.0-Linux-x86_64.sh
optional: add anaconda2/bin to PATH
optional: update
$ conda update conda
$ conda update anaconda
$ cd build
$ wget https://github.com/bazelbuild/bazel/releases/download/0.3.1/bazel-0.3.1-installer-linux-x86_64.sh
$ chmod +x bazel-0.3.1-installer-linux-x86_64.sh
$ ./bazel-0.3.1-installer-linux-x86_64.sh --user
$ cd ..
add $HOME/bin to PATH
$ cd tensorflow-0.10.0
$ bash ./configure
If choose GPU support and CUDA is not installed at /usr/local, edit file third_party/gpus/crosstool/CROSSTOOL.tpl:
# Include directory for cuda headers.
# cxx_builtin_include_directory: "/usr/local/cuda%{cuda_version}/include"
cxx_builtin_include_directory: "/path/to/cuda/include"
Now build the python package:
-
with GPU:
$ bazel build -c opt --config=cuda //tensorflow/tools/pip_package:build_pip_package -
without GPU:
$ bazel build -c opt //tensorflow/tools/pip_package:build_pip_package
Install the python package:
$ mkdir -p ../build/tensorflow_pkg
$ bazel-bin/tensorflow/tools/pip_package/build_pip_package ../build/tensorflow_pkg
$ cd ..
$ pip install --ignore-installed --upgrade build/tensorflow_pkg/tensorflow-0.10.0-py2-none-any.whl
$ cd syntaxnet-goodold-repo
$ bazel build -c opt //syntaxnet:parser_ops.so
$ cp -r ../syntaxnet ~
$ cp bazel-bin/syntaxnet/parser_ops.so ~/syntaxnet
$ cp -r syntaxnet/models/parsey_mcparseface ~/syntaxnet
$ ./parsey_mcparseface_context.sh ~/syntaxnet
$ cp ../bin/* ~/bin
$ echo "John loves Mary." | parse -v