1. What is NanoEdge AI Library for 1-class classification (outlier detection)?
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 "identify" outliers 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 1-class classification 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 identify if a sensor pattern is inside the class or outside the class (outlier/anomaly). The class under consideration is 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:
outlier_init()
|
run first before the outlier detection process |
outlier()
|
run an outlier detection iteration to identify if the pattern is inside or outside the class (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 expectedNEAI_ERROR: internal error with the libraryNEAI_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 supportedNEAI_LEARNING_DONE: minimum number of learning iterations reachedNEAI_LEARNING_IN_PROGRESS: fail-safe to prevent insufficient number of learning iterations; run more iterations.
2.2.1. Initialization
enum neai_state neai_outlier_init(void);
Initialization must be called at the beginning.
- Input:
- None
- Output:
- the
neai_stateenum (NEAI_OK == 0, in case of success).
- the
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. 1-class classification
enum neai_state neai_outlier(float *in, uint8_t *is_outlier);
This function returns the result of outlier detection, whether or not the detected buffer is an outlier.
- Input:
float *in, the length of the buffer isNEAI_INPUT_SIGNAL_LENGTH * NEAI_INPUT_AXIS_NUMBER.
- Output:
1if the input pattern is detected as an outlier (outside the class considered), and0if it is not an outlier (it belongs to the class).- The
neai_stateenum (NEAI_OK == 0, in case of success).
2.3. Example "Hello World!"
Header files:
NanoEdgeAI.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:
/* =============
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 outlier detection model by loading
* the pretrained model.
* @return NEAI_OK on success, error code otherwise.
*/
enum neai_state neai_outlier_init(void);
/**
* @brief Perform outlier detection on a new input sample by returning whether it is an outlier
* or not.
* @param in [in] Pointer to the input signal array
* (size NEAI_INPUT_SIGNAL_LENGTH * NEAI_INPUT_AXIS_NUMBER).
* @param is_outlier [out] Pointer to the outlier flag (1 = outlier, 0 = normal).
* @return NEAI_OK on success.
* Error code otherwise.
*/
enum neai_state neai_outlier(float *in, uint8_t *is_outlier);
/* ===== 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
uint8_t is_outlier; // Outlier flag returned by outlier detection function (1 = outlier, 0 = normal)
============= */
Main program: main.c
This program must be completed by the user (depending on the applications or the desired features).
/**
**************************************************************************
* 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_outlier_init();
if (error_code != NEAI_OK) {
/* Check the returned error code (cf NanoEdgeAI.h). */
}
/* Outlier ----------------------------------------------------------------*/
uint8_t is_outlier = 0;
while (1) {
fill_buffer(input_signal);
neai_outlier(input_signal, &is_outlier);
/* USER BEGIN */
/*
* e.g.: Trigger functions depending on is_outlier flag, 1 = outlier, 0 = inlier
* (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.