LEGACY

Quickstart

• Refer to Installing TensorFlow for Java
• Javadoc

Nightly builds

Releases built from release branches are available on Maven Central. Additionally, every day binaries are built from the master branch on GitHub:

• JAR
• Source JAR
• JNI:
  • Linux CPU-only
  • Linux GPU
  • MacOS
  • Windows: (No nightly builds available yet)

Building from source

If the quickstart instructions above do not work out, the TensorFlow Java and native libraries will need to be built from source.

1. Install bazel

2. Setup the environment to build TensorFlow from source code (Linux or macOS). If you'd like to skip reading those details and do not care about GPU support, try the following:

   shCopy

   # On Linux
   sudo apt-get install python swig python-numpy
   
   # On Mac OS X with homebrew
   brew install swig
   

3. Configure (e.g., enable GPU support) and build:

   shCopy

   ./configure
   bazel build --config opt \
     //tensorflow/java:tensorflow \
     //tensorflow/java:libtensorflow_jni
   

The command above will produce two files in the bazel-bin/tensorflow/java directory:

• An archive of Java classes: libtensorflow.jar
• A native library: libtensorflow_jni.so on Linux, libtensorflow_jni.dylib on OS X, or tensorflow_jni.dll on Windows.

To compile Java code that uses the TensorFlow Java API, include libtensorflow.jar in the classpath. For example:

javac -cp bazel-bin/tensorflow/java/libtensorflow.jar ...

To execute the compiled program, include libtensorflow.jar in the classpath and the native library in the library path. For example:

java -cp bazel-bin/tensorflow/java/libtensorflow.jar \
  -Djava.library.path=bazel-bin/tensorflow/java \
  ...

Installation on Windows requires the more experimental bazel on Windows. Details are omitted here, but find inspiration in the script used for building the release archive: tensorflow/tools/ci_build/windows/libtensorflow_cpu.sh.

Bazel

If your project uses bazel for builds, add a dependency on //tensorflow/java:tensorflow to the java_binary or java_library rule. For example:

bazel run -c opt //tensorflow/java/src/main/java/org/tensorflow/examples:label_image