[Documentation] [TitleIndex] [WordIndex

  Show EOL distros: 

Package Summary

Catkin-friendly C++ bindings for the tensorflow package.

Package Summary

Catkin-friendly C++ bindings for the tensorflow package.

Package Summary

Catkin-friendly C++ bindings for the tensorflow package.

A Catkin-friendly package for utilizing the C++ API of Tensorflow.

Get Tensorflow C++ API into ROS as easy as

find_package(catkin REQUIRED COMPONENTS
  ... your other packages ...
  tensorflow_ros_cpp
)

See the usage example at https://github.com/tradr-project/tensorflow_ros_test .

Starting with ROS Kinetic, the supported Ubuntu versions started to use C++11 ABI for building system packages. This makes problems when using the pip-installed Tensorflow (as described further). See section C++ ABI difference problems for hints on how to overcome this limitation.

Note: If you only intend to use Tensorflow from Python nodes, you don't need this package. Just add <run_depend>python-tensorflow-pip</run_depend> to your package.xml and that's it.

Multiple Tensorflow installation options

This package allows its users to "provide" Tensorflow in several ways, and the developer doesn't have to care about which option the user chooses:

A comprehensive list of compatible options for various OSes and ROS distros is available in the package README.

Detailed configuration options are available for each of the options.

Note for rosdep users

If you're managing dependencies via rosdep, it is likely that you do not want it to try to install the optional dependencies (currently python-tensorflow-pip and tensorflow_catkin). In such case, add the following to the rosdep call:

rosdep install ... --skip-keys=tensorflow_catkin --skip-keys=python-tensorflow-pip

C++ ABI difference problems

Ubuntu systems starting with 16.04 (Xenial) are using the new C++11 ABI for all system libraries. Until pip bug 1707002 gets resolved (if ever!), the pip-distributed Tensorflow is built againts an older C++ ABI, which is incompatible with the C++11 ABI. This means linking to the Tensorflow library will fail on such systems and there's no way around it.

The only "workaround" is to wrap all tensorflow code into a library that exposes its functions using C API (so no std::strings, std::vectors or Eigen types) and which does not link to any system library with a C++ API. Such library can then be built separately with -D_GLIBCXX_USE_CXX11_ABI=0. Then you can freely link the rest of your code using ROS, Eigen and so on to this library.

An example of this approach can be found at kinetic-devel branch of tensorflow_ros_test.

A library created this way is "backwards compatible" with older systems, so it can be used everywhere. However, separating all Tensorflow code can be tiresome, so we only suggest it as a last resort option. This is why searching for the pip-installed Tensorflow is disabled by default on systems using the new C++11 ABI (you can force it using -DFORCE_TF_PIP_SEARCH=ON).


2022-05-28 13:04