NanoEdge AI Library for extrapolation (E)

1. What is NanoEdge AI Library for extrapolation?

NanoEdge™ AI Library is an Artificial Intelligence (AI) static library for embedded C software running on Arm^® Cortex^® microcontrollers.

When embedded on microcontrollers, it gives them the ability to easily "predict" target value from sensor patterns, by themselves, without the need for the user to have additional skills in Mathematics, Machine Learning, or data science.

The NanoEdge AI static library for extrapolation is the code that contains an AI model (for example, as a bundle of signal treatment, Machine Learning model, and optimally tuned hyperparameters) designed to estimate / predict a target value from a sensor pattern. The predicted values (target values) used to fit the model are defined by the user in NanoEdge AI Studio (NanoEdgeAIStudio), also called the Studio, and are used during the training process of the AI model.

2. Install / Getting started

The main functions available via the library are:

init()	run first before the extrapolation process
extrapolation()	run a regression process for extrapolation, to predict the response (inference)

2.1. Static library

• In NanoEdge AI Studio, after obtaining a library, click Compile (on the "Deployment" screen, which follows the "Benchmark" and "Emulator" screens)
• Open the .zip file obtained
• Select and copy the static library libneai.a
• Link this static library to your project code

2.2. NanoEdge AI Library functions

Most NanoEdge AI function return the status of the library in the following enum, neai_state:

enum neai_state {
    NEAI_OK = 0,
    NEAI_ERROR = 1,
    NEAI_NOT_INITIALIZED = 2,
    NEAI_INVALID_PARAM = 3,
    NEAI_NOT_SUPPORTED = 4,
    NEAI_LEARNING_DONE = 5,
    NEAI_LEARNING_IN_PROGRESS = 6
};

Here are the possible statuses:

NEAI_OK: library working as expected

NEAI_ERROR: internal error with the library

NEAI_NOT_INITIALIZED: learn or detect functions were called without running the init function first; initialize your library.

NEAI_INVALID_PARAM: neai function was called with one or more incorrect or missing parameters.

NEAI_NOT_SUPPORTED: board not supported

NEAI_LEARNING_DONE: minimum number of learning iterations reached

NEAI_LEARNING_IN_PROGRESS: fail-safe to prevent insufficient number of learning iterations; run more iterations.

Information

Although it is present in all NanoEdge AI header files, the NEAI_NOT_ENOUGH_CALL_TO_LEARNING status is only relevant for anomaly detection libraries.

2.2.1. Initialization

enum neai_state neai_extrapolation_init(void);

Initialization must be called at the beginning.

• Input:

None

• Output:

the neai_state enum (NEAI_OK == 0, in case of success).

For more details on the output of the initialization function, refer to the header file NanoEdgeAI.h provided in the .zip file containing the static NanoEdge AI Library, or in the code examples below.

2.2.2. Extrapolation

enum neai_state neai_extrapolation(float input_buffer[], float *extrapolated_value);

This function returns the target extrapolated value.

• Input:

float input_buffer[], the length of the buffer is DATA_INPUT_USER * AXIS_NUMBER.

• Output:

The predicted value (floating point).

The neai_state enum (NEAI_OK == 0, in case of success).

Information

The float extrapolated_value variable must be defined prior to calling the extrapolation function, and pointed to using &extrapolated_value when passed as an argument (see code example below).

Warning

NanoEdge AI Library uses float data types instead of int. If you are using intdata types, convert (cast) them into float.

2.3. Example "Hello World!"

Header files:

NanoEdgeAI.h and knowledge.h (provided in the .zip file that you download by clicking Compile in NanoEdge AI Studio (on the "Deploy" screen after obtaining a library)

Example of NanoEdge AI Library header file:

This snippet is provided AS IS, and by taking it, you agree to be bound to the license terms that can be found here for the component: Application.

                                                                                             
/* =============
Copyright (c) 2026, STMicroelectronics

All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted 
provided that the following conditions are met:

* Redistributions of source code must retain the above copyright notice, this list of conditions 
  and the following disclaimer.

* Redistributions in binary form must reproduce the above copyright notice, this list of
  conditions and the following disclaimer in the documentation and/or other materials provided 
  with the distribution.

* Neither the name of the copyright holders nor the names of its contributors may be used to 
  endorse or promote products derived from this software without specific prior written 
  permission.

*THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR 
 IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY 
 AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER / 
 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 
 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR 
 SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 
 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR 
 OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 
 POSSIBILITY OF SUCH DAMAGE.*
*/


#ifndef NANOEDGEAI_H
#define NANOEDGEAI_H

#include <stdint.h>

/* NEAI ID */
#define NEAI_ID "None"

/* Input signal configuration */
#define NEAI_INPUT_SIGNAL_LENGTH 512
#define NEAI_INPUT_AXIS_NUMBER 1

/* NEAI State Enum */
enum neai_state {
	NEAI_OK = 0,
	NEAI_ERROR = 1,
	NEAI_NOT_INITIALIZED = 2,
	NEAI_INVALID_PARAM = 3,
	NEAI_NOT_SUPPORTED = 4,
	NEAI_LEARNING_DONE = 5,
	NEAI_LEARNING_IN_PROGRESS = 6
};


#ifdef __cplusplus
extern "C" {
#endif

/**
 * @brief  Must be called at the beginning to initialize the regression model by loading the
 *         pretrained model.
 * @return NEAI_OK on success, error code otherwise.
 */
enum neai_state neai_extrapolation_init(void);

/**
 * @brief  Perform regression on a new input sample by returning the predicted output values.
 * @param  in         [in]   Pointer to the input signal array
 *                           (size NEAI_INPUT_SIGNAL_LENGTH * NEAI_INPUT_AXIS_NUMBER).
 * @param  prediction [out]  Pointer to the predicted output values.
 * @return NEAI_OK on success.
 *         Error code otherwise.
 */
enum neai_state neai_extrapolation(float *in, float *prediction);

/* ===== Common getter functions ===== */
/**
 * @brief  Get the NEAI identifier.
 * @return Pointer to a string containing the NEAI ID.
 */
char* neai_get_id(void);

/**
 * @brief  Get the input signal size (number of samples per axis).
 * @return Input signal size.
 */
int neai_get_input_signal_size(void);

/**
 * @brief  Get the number of input axes/channels.
 * @return Number of input axes.
 */
int neai_get_axis_number(void);


#ifdef __cplusplus
}
#endif

#endif /* NANOEDGEAI_H */


/* =============
Declarations to add to your main program to use the NanoEdge AI library.
You may copy-paste them directly or rename variables as needed.
WARNING: Respect the structures, types, and buffer sizes; only variable names may be changed.

enum neai_state state;   // Captures return states from NEAI functions
float input_signal[NEAI_INPUT_SIGNAL_LENGTH * NEAI_INPUT_AXIS_NUMBER];   // Input signal buffer
float prediction;  // Predicted value by regression function
============= */

Important

The input buffer collected from your sensor is already declared in the header file NanoEdgeAI.h, float input_user_buffer[DATA_INPUT_USER * AXIS_NUMBER];, feel free to comment this line, and then rename and declare your input buffer in the main program. The knowledge buffer knowledge is declared in the header file knowledge.h.

Main program: main.c

This program must be completed by the user (depending on the applications or the desired features).

This snippet is provided AS IS, and by taking it, you agree to be bound to the license terms that can be found here for the component: Application.

/**
  **************************************************************************
  * Demo: NanoEdge AI process to include in main program body
  *
  * @note  This program must be completed and customized by the user
  **************************************************************************
  */

/* Includes --------------------------------------------------------------------*/
#include "NanoEdgeAI.h"
/* Private define --------------------------------------------------------------*/
/* Private variables defined by user -------------------------------------------*/
float input_signal[NEAI_INPUT_SIGNAL_LENGTH * NEAI_INPUT_AXIS_NUMBER]; // Buffer of input values
/* Private function prototypes defined by user ---------------------------------*/
/*
 * @brief Collect data process
 *
 * This function is defined by user, depends on applications and sensors
 *
 * @param input_signal: [in, out] buffer of sample values
 * @retval None
 * @note   If NEAI_INPUT_AXIS_NUMBER = 3 (cf NanoEdgeAI.h), the buffer must be
 *         ordered as follow:
 *         [x0 y0 z0 x1 y1 z1 ... xn yn zn], where xi, yi and zi
 *         are the values for x, y and z axes, n is equal to
 *         NEAI_INPUT_SIGNAL_LENGTH (cf NanoEdgeAI.h)
 */
void fill_buffer(float *input_signal)
{
	/* USER BEGIN */
	/* USER END */
}
/* ----------------------------------------------------------------------------*/
int main(void)
{
	/* Initialization -----------------------------------------------------------*/
	enum neai_state error_code = neai_extrapolation_init();
	if (error_code != NEAI_OK) {
		/* Check the returned error code (cf NanoEdgeAI.h). */
	}
	/* Extrapolation ------------------------------------------------------------*/
	float prediction = 0.0;
	while (1) {
		fill_buffer(input_signal);
		neai_extrapolation(input_signal, &prediction);
		/* USER BEGIN */
    /*
    * e.g.: Trigger functions depending on predicted value, prediction
    * (blink LED, ring alarm, etc.).
    */
		/* USER END */
	}
}

3. Resources

Documentation
All NanoEdge AI Studio documentation is available here.

Tutorials
Step-by-step tutorials to use NanoEdge AI Studio to build a smart device from A to Z.