Skip to content
No description, website, or topics provided.
C++ BitBake Makefile
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
recipes-arm Initial support for ARMNN May 24, 2019
recipes-devtools Bump tensorflow lite version to v1.14.0-rc0 Jun 7, 2019
recipes-support/libeigen Bump tensorflow lite version to v1.14.0-rc0 Jun 7, 2019
recipes-tensorflow-lite/tensorflow-lite Update May 10, 2019


This Yocto/OpenEmbedded machine-learning meta-layer provides support for machine learning libraries:

TensorFlow Lite:

The TensorFlow Lite ( recipe for the FullMetalUpdate project and is an evolution of the layer developped for the RZ/G1 family of System on Chips:

In order to add TensorFlow Lite support to your project, make sure tensorflow-lite is listed as a dependency to your recipe/package. Listing tensorflow-lite-staticdev and tensorflow-lite-dev in IMAGE_INSTALL variable could be beneficial when you just want to populate an SDK for developing an application based on TensorFlow Lite.

After the build is complete the static C++ TensorFlow Lite library (libtensorflow-lite.a) will be generated.

The library can be verified with the TensorFlow Lite image classification sample application named label_image which is included in the build (included by package tensorflow-lite-examples). The sample application is installed under /usr/bin/tensorflow-lite/examples/.

You can’t perform that action at this time.