README.rst
1.. _tensorflow_hello_world:
2
3TensorFlow Lite Micro Hello World sample
4########################################
5
6Overview
7********
8
9This sample TensorFlow application replicates a sine wave and
10demonstrates the absolute basics of using TensorFlow Lite Micro.
11
12The model included with the sample is trained to replicate a
13sine function and generates x values to print alongside the
14y values predicted by the model. The x values iterate from 0 to
15an approximation of 2π.
16
17The sample also includes a full end-to-end workflow of training
18a model and converting it for use with TensorFlow Lite Micro for
19running inference on a microcontroller.
20
21.. Note::
22 This README and sample have been modified from
23 `the TensorFlow Hello World sample for Zephyr`_.
24
25.. _the TensorFlow Hello World sample for Zephyr:
26 https://github.com/tensorflow/tflite-micro/tree/main/tensorflow/lite/micro/examples/hello_world
27
28Building and Running
29********************
30
31This sample should work on most boards since it does not rely
32on any sensors.
33
34This application can be built and executed on QEMU as follows:
35
36.. zephyr-app-commands::
37 :zephyr-app: samples/modules/tensorflow/hello_world
38 :host-os: unix
39 :board: qemu_x86
40 :goals: run
41 :compact:
42
43Exit QEMU by pressing :kbd:`CTRL+A` :kbd:`x`.
44
45Sample Output
46=============
47
48.. code-block:: console
49
50 ...
51
52 x_value: 1.0995567*2^1, y_value: 1.6951603*2^-1
53
54 x_value: 1.2566366*2^1, y_value: 1.1527088*2^-1
55
56 x_value: 1.4137159*2^1, y_value: 1.1527088*2^-2
57
58 x_value: 1.5707957*2^1, y_value: -1.0849024*2^-6
59
60 x_value: 1.7278753*2^1, y_value: -1.0509993*2^-2
61
62 ...
63
64The modified sample prints 50 generated-x-and-predicted-y pairs.
65
66Modifying Sample for Your Own Project
67*************************************
68
69It is recommended that you copy and modify one of the two TensorFlow
70samples when creating your own TensorFlow project. To build with
71TensorFlow, you must enable the below Kconfig options in your :file:`prj.conf`.
72
73:file:`prj.conf`:
74
75.. code-block:: console
76
77 CONFIG_CPLUSPLUS=y
78 CONFIG_NEWLIB_LIBC=y
79 CONFIG_TENSORFLOW_LITE_MICRO=y
80
81Training
82********
83Follow the instructions in the :file:`train/` directory to train your
84own model for use in the sample.
85
