zephyr

History

Jamie McCrae ec7044437e treewide: Disable automatic argparse argument shortening Disables allowing the python argparse library from automatically shortening command line arguments, this prevents issues whereby a new command is added and code that wrongly uses the shortened command of an existing argument which is the same as the new command being added will silently change script behaviour. Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>		2023-01-26 20:12:36 +09:00
..
boards	samples: modules: tflite-micro: magic_wand: add missing CONFIG_I2C=y	2022-08-05 12:55:51 +02:00
renode
src	includes: prefer <zephyr/kernel.h> over <zephyr/zephyr.h>	2022-09-05 16:31:47 +02:00
train	treewide: Disable automatic argparse argument shortening	2023-01-26 20:12:36 +09:00
CMakeLists.txt
prj.conf	treewide: Update deprecated CONFIG_LIB_CPLUSPLUS usages	2023-01-13 17:42:55 -05:00
README.rst	treewide: Use CONFIG_CPP instead of CONFIG_CPLUSPLUS	2023-01-13 17:42:55 -05:00
sample.yaml	samples: add module requirement into samples	2022-04-19 09:38:55 -04:00

README.rst

.. _tensorflow_magic_wand:

TensorFlow Lite Micro Magic Wand sample
#######################################

Overview
********

This sample application shows how to use TensorFlow Lite Micro
to run a 20 kilobyte neural network model that recognizes gestures
from an accelerometer.

.. Note::
    This README and sample have been modified from
    `the TensorFlow Magic Wand sample for Zephyr`_ and
    `the Antmicro tutorial on Renode emulation for TensorFlow`_.

.. _the TensorFlow Magic Wand sample for Zephyr:
    https://github.com/tensorflow/tflite-micro/tree/main/tensorflow/lite/micro/examples/magic_wand

.. _the Antmicro tutorial on Renode emulation for TensorFlow:
    https://github.com/antmicro/litex-vexriscv-tensorflow-lite-demo

Building and Running
********************

The application can be built for the :ref:`litex-vexriscv` for
emulation in Renode as follows:

.. zephyr-app-commands::
   :zephyr-app: samples/tensorflow/magic_wand
   :host-os: unix
   :board: litex_vexriscv
   :goals: build
   :compact:

Once the application is built, `download and install Renode 1.12 or higher as a package`_
following the instructions in the `Renode GitHub README`_ and
start the emulator:

.. code-block:: console

    renode -e "set zephyr_elf @./build/zephyr/zephyr.elf; s @./samples/modules/tflite-micro/magic_wand/renode/litex-vexriscv-tflite.resc"

.. _download and install Renode 1.12 or higher as a package:
    https://github.com/renode/renode/releases/

.. _Renode GitHub README:
    https://github.com/renode/renode/blob/master/README.rst

Sample Output
=============

The Renode-emulated LiteX/VexRiscv board is fed data that the
application recognizes as a series of alternating ring and slope
gestures.

.. code-block:: console

    Got accelerometer, label: accel-0

    RING:
              *
           *     *
         *         *
        *           *
         *         *
           *     *
              *

    SLOPE:
            *
           *
          *
         *
        *
       *
      *
     * * * * * * * *

    RING:
              *
           *     *
         *         *
        *           *
         *         *
           *     *
              *

    SLOPE:
            *
           *
          *
         *
        *
       *
      *
     * * * * * * * *

Modifying Sample for Your Own Project
*************************************

It is recommended that you copy and modify one of the two TensorFlow
samples when creating your own TensorFlow project. To build with
TensorFlow, you must enable the below Kconfig options in your :file:`prj.conf`:

.. code-block:: kconfig

    CONFIG_CPP=y
    CONFIG_NEWLIB_LIBC=y
    CONFIG_TENSORFLOW_LITE_MICRO=y

Training
********
Follow the instructions in the :file:`train/` directory to train your
own model for use in the sample.