STT-tensorflow/tensorflow/lite/experimental/examples/unity/TensorFlowLitePlugin
Austin Anderson 61c6c84964 Migrate TensorFlow Lite out of tensorflow/contrib
This change moves //tensorflow/contrib/lite to //tensorflow/lite in preparation
for TensorFlow 2.0's deprecation of contrib/. If you refer to TF Lite build
targets or headers, you will need to update them manually. If you use TF Lite
from the TensorFlow python package, "tf.contrib.lite" now points to "tf.lite".
Please update your imports as soon as possible.

For more details, see https://groups.google.com/a/tensorflow.org/forum/#!topic/tflite/iIIXOTOFvwQ

@angersson and @aselle are conducting this migration. Please contact them if
you have any further questions.

PiperOrigin-RevId: 219536476
2018-10-31 14:20:28 -07:00
..
Assets Migrate TensorFlow Lite out of tensorflow/contrib 2018-10-31 14:20:28 -07:00
ProjectSettings Migrate TensorFlow Lite out of tensorflow/contrib 2018-10-31 14:20:28 -07:00
UnityPackageManager Migrate TensorFlow Lite out of tensorflow/contrib 2018-10-31 14:20:28 -07:00
.gitignore Migrate TensorFlow Lite out of tensorflow/contrib 2018-10-31 14:20:28 -07:00
README.md Migrate TensorFlow Lite out of tensorflow/contrib 2018-10-31 14:20:28 -07:00

TF Lite Experimental Unity Plugin

This directory contains an experimental sample Unity (2017) Plugin, based on the experimental TF Lite C API. The sample demonstrates running inference within Unity by way of a C# Interpreter wrapper.

Note that the native TF Lite plugin(s) must be built before using the Unity Plugin, and placed in Assets/TensorFlowLite/SDK/Plugins/. For the editor (note that this has only been tested on Linux; the syntax may differ on Mac/Windows):

bazel build -c opt --cxxopt=--std=c++11 \
  //tensorflow/lite/experimental/c:libtensorflowlite_c.so

and for Android:

bazel build -c opt --cxxopt=--std=c++11 \
  --crosstool_top=//external:android/crosstool \
  --host_crosstool_top=@bazel_tools//tools/cpp:toolchain \
  --cpu=armeabi-v7a \
  //tensorflow/lite/experimental/c:libtensorflowlite_c.so

If you encounter issues with native plugin discovery on Mac ("Darwin") platforms, try renaming libtensorflowlite_c.so to tensorflowlite_c.bundle. Similarly, on Windows you'll likely need to rename libtensorflowlite_c.so to tensorflowlite_c.dll.