Edge AI

How to check the classification accuracy of image classification models on STM32H7 ?    Ritesh           

Hi,I see the ST Edge AI Suite case studies People counting with a ranging sensor. Is there any demo project for that ?Regards

'ai_layer_conv2d' {aka 'struct ai_layer_conv2d_'} has no member named 'in_ch_format''ai_layer_conv2d' {aka 'struct ai_layer_conv2d_'} has no member named 'out_ch_format' Also i am trying to run a object detection model in which i want to glows LEDs based on the position of the object on the screen based on the coordinates of bounding box. Can anyone help in that.

Hi everyone,I'm currently working on a project that involves implementing Edge AI using the X-CUBE-AI software package within STM32CubeIDE. After adding the AI model, which requires 44 bytes of RAM and generating the necessary code, We encountered a problem: one of the peripherals, specifically I2C, starts hanging after a while when interfacing with a sensor to obtain readings.Here’s a summary of the steps we've taken so far:Isolated the Issue: To determine if the hanging issue was related to the AI initialization, we commented out the AI functions from the main file. However, the I2C communication still hangs when waiting for a response from the sensor after some time.Sensor Exemption: We tried excluding the specific sensor to see if other sensors would work correctly. Unfortunately, the other sensors are also exhibiting the same hanging behavior.Peripherals in Use: Our project involves multiple active peripherals, including I2C1-3 and UART4-5.Additional InformationWorkspace Vers

X-CUDE-AI 9.1.0 Win11Analysis & Validating on Desktop, Passed.Validating ONNX model on NUCLEO F767zi (automated), Failed.Since this automated validation process cannot be intervened or debugged, it seems I can't do anything. Why this appears?Error here:[AI:network] Building project[AI:network] Generating project[AI:network] Temporary project loaded[AI:network] Temporary AI configuration done[AI:network] Generating temporary project code[AI:network] Building project[AI:network] D:/STM32CubeIDE/STM32CubeIDE_1.13.2/STM32CubeIDE/plugins/com.st.stm32cube.ide.mcu.externaltools.gnu-tools-for-stm32.12.3.rel1.win32_1.0.200.202406191623/tools/bin/arm-none-eabi-gcc -c -mcpu=cortex-m7 -mfpu=fpv5-d16 -mfloat-abi=hard -DDEBUG -DUSE_HAL_DRIVER -DSTM32F767xx -D -IDrivers/CMSIS/Device/ST/STM32F7xx/Include -IDrivers/CMSIS/Include -IDrivers/STM32F7xx_HAL_Driver/Inc/Legacy -IDrivers/STM32F7xx_HAL_Driver/Inc -IInc -IMiddlewares/ST/AI/Inc -O3 -Wall -fdata-sections -ffunction-sections Drivers/STM32F7xx_H

Dear Support As I am trying to get familiar with the NAnoEdge AI Sudio, I tried the followed steps to setup a demo project:1. New project -> select the EVB(NUCLEO-F401RE) and Sensor(Accelerometer - 3 Axes)2. Regular Signals : Create a datalogger -> Select the EVB(F401RE + X-NUCLEO-IKS02A1: 3 axis, 25 Hz, range-8, samples per axis - 16) -> generate Data logger.3. Regular Signals: Add signal-> From File -> Choose the neai_datalog.bin which is generated in STEP#2 -> continue but with followed capture's error: problem detected on line(s) 5.6.7.8,xxx. Could you please guide how to go to next step for benchmark? RegardsXiao Kang

i need some assistance with modelzoo . so first of all . i converted a dataset to darknet using converter.py than i trained it using train mode i had a 2.6mo  .h5 file  then i quatized it to 600kb .tflite file. but when i tried to deploy it i had an error saying just exception "wrong"  from compile.py file. so is there any assistances about this error and second of all i realized that despite modelzoo say that he can only deploy tflite models . there are some tflite models that he cannot deploy . ( i tried 4 different tflite models 1 was deployed and 3 did not) so is there any parameters or configurations i should know about  in builiding my model(conditions to be respected) so my model can be compatible with modelzoo

私はNANOEDGE AI STUDIO内のData Logger Generatorを参照しました。現在ここに私が所有するB-L475E-IOT01A2（開発ボード）がありません。そこでB-L475E-IOT01A2で温度・湿度・加速度のデータ・ログを取得し、NANOEDGE AI STUDIOにて機械学習を実行したいと思っています。なお温度は内部温度センサを使おうと思います。そこでB-L475E-IOT01A2で使えるおすすめの湿度センサ・加速度センサを教えてください。また、内部温度センサ・湿度センサ・加速度センサのデータを取得するサンプルプログラムがありましたら教えてください。Translated to English by Community team with DeepL.com (free version)I have referenced the Data Logger Generator in the NANOEDGE AI STUDIO.Currently, I do not have my B-L475E-IOT01A2 (development board) here.So I used B-L475E-IOT01A2 to get data logs of temperature, humidity, and acceleration,I would like to use B-L475E-IOT01A2 to acquire temperature, humidity, and acceleration data/logs, and execute machine learning with NANOEDGE AI STUDIO.For temperature, I would like to use an internal temperature sensor.So, please let me know the recommended humidity sensor and acceleration sensor that can be used with B-L475E-IOT01A2.Please let me know.Also, please let me know if there is a sample program to

Hello, everyone! I'm learning to deploy my target detection model using cubeai. My model is trained using pytorch and I found that cubeai doesn't support .pth file conversion so I'm trying to convert my model to onnx format. Since the size of my model did not match the requirements of the development board, I performed a quantization operation. Noting that cubeai officially recommends quantization in onnx format, I did so. However, when I analyze the model with cubeai, the following error occurs (the model before quantization can be analyzed normally). Can anyone help me to see what's going on? 

Hi, I was working with NEAI and saw these constraints mentioned in the wiki:General considerations for input file format:.txt / .csv filesNumerical values only, no headersUniform separators (single space, tab, comma, or semicolon)Decimal values formatted with a period (.)Fewer than 16,384 values per lineConsistent number of numerical values on each lineMinimum of 20 lines per sensor axisFewer than ~100,000 total linesFile size less than ~1 GbitWhat is the reason for these constraints and are these strictly enforced? Is there any issue if I don't follow these?

The free online platform (formerly named STM32Cube.AI Developer Cloud) now supports MPUs and smart sensors in addition to MCUs. It also supports AI hardware acceleration (NPU) on compatible devices.   Leveraging the ST Edge AI Core technology, the tool allows you to bring your own model or select one from our model zoo, optimize it, and run benchmarks remotely on real ST boards to identify the ideal platform for your project.   This tool is also part of the recently introduced ST Edge AI Suite.   Check it out here!  

I am trying to integrate the a neural network machine learning model into the stm32h755 chip. I downloaded the stmCubeMx, installed X-Cube-Ai, selected the core, added the network using tflite file and then generated the code. I can't find the ai_model_data.c, ai_model_data.h, ai_model.h and some other important files. Could you please provide a detailed steps by step process to integrate the Neural Network in an ALREADY created/ EXISTING project?

I'm a senior student majored in AI before and recently bought an STM32H743VIT6 to experiment with Edge AI. I haven't had any experience with embedded systems before. My ports seem to be missing, but CubeProgrammer successfully found my device through USB and allowed me to flash it.How strange it is! My classmate mentioned that I need try an ST-Link, but I'm not entirely sure.It may be a very simple question but I don't know anything about Stm32H743VIT6 and how to start.By the way , there are lots of requires about models on stm32.Most of the ONNX models I used to developed during my four years of university even can't be analysed ,made me very upset because of these requires.I almost lost my confidence in embedded systems and AI. :(

Hi, I am using STM32F746G Discovery Board to deploy an object detection model to detect person in an image. The model size of the object detection model is less than 1Mb. The implementation would be like to make an inference in the microcontroller like access the jpg images in the sd card (since i dont have a camera module) and display it in the LCD together with its prediction like bounding boxes on the image in case there is a person in it.   I used the STM32Cube MX and enabled XCUBE AI then upload the pre trained model which is the ssd mobile net v2 but I am having trouble making the configuration for displaying image, accessing files in sd card, decoding the jpg images from the sd card in STM32Cube MX land I also don't know how to use the XCUBE AI application template like the methods ai_run(), acquire_and_process_data(ai_i8* data[]) and others.   

I am testing a tflite model on the ISPU, using the following hardware:NUCLEO-F401RE + X-NUCLEO-IKS4A1I downloaded the x-cube-ispu repo and imported the template_stedgeai example to STM32CubeIDE. Finally i generate the network library with:  stedgeai generate --target ispu --no-workspace --no-report -m /path/to/model.tflite --output .   Everything works perfectly fine with networks under 3000 kB. If the network has over 3000 parameters (quantized to int8) I get:ispu_stedgeai.elf section `.rodata' will not fit in region `ROM'region `ROM' overflowed by X bytesThe 3000 sounds way too small as the ISPU has 32kB of ROM. Is there something I need to do in order to enable bigger models sizes or am I doing something wrong?I attached the output of the "stedgeai generate" and my main.c for context. The output from the file, gave: "region `ROM' overflowed by 2596 bytes" when building in STM32CubeIDE. 

Hello there,I'm using stm32h747i-DISCO board, and running object detection demo from model-zoo. Now i want to add Ethernet functionality with the demo. I have some questions for the same, as below...Q1) there is only one linker script in object detection project : STM32H747XIHx_CM7.ld. It's for RAM or FLASH ? how it's running with one .ld file ?Q2) I have enabled the Ethernet in individual project, in which CUBEmx showing the warning.In linker script "RAM_D1" is there at 0x24000000.In object detection project, At 0x24000000 address "AXIRAM" is there...and some Image_Buffers are stored in the same.Then, how can i make space for Ethernet (LwIP + mqtt)?which memory should be choose for the same ?Please guide us for the same, And please share any guide for memory utilisation for AI (model-zoo), how to decide which buffer should be placed in which memory.Thanks and regards Ishan 

Hi,1) I Am currently exploring the capabilities of NanoEdge for a project involving multiclass classification, and I am eager to learn more about the specific machine learning algorithms utilized within the NanoEdge framework. I feel as though I am using a black box to generate the model, and I would like to know if it is possible to make optimizations during the learning step.2) After generating the model, I used the time.h library to compute the inference time of the model in STM32CubeIDE by creating a timer before and after calling the neai_classification function. However, the result is always 0, even though the model produces the correct classification results. This may be because the inference time is very small. Are there other solutions to accurately measure the inference time? Thank you very much for your help 

I am lost in the maze of documentation for AI features at STM. I have access right now to:STEVAL-MKBOXPRO (SensorTile.box PRO)Custom board with LSM6DSV16XI want to:gather IMU data to train the modeltrain the IMU device decision tree classifier (3 class)deploy it ultimately on the custom board, but the MKBOXPRO is also acceptable as a target for nowMy background is in data science, so the 'AI' part I understand. My embedded programming and hardware design knowledge is decent but still beginner level compared to most people here.So far:Unico-GUI - doesn't work on the mkboxpro?MEMS studio doesn't have a data collection feature like Unico? And I'm having connection issues, although I suspect I'll figure that out soon enoughApps for mkboxpro on phone don't seem to look like they do in the documentation, perhaps due to it not being in Pro modeThere's yet another tool called STEdge AI Core which I haven't explored yet but seems to overlap with the othersThe simplest path I see right now

 1. Is there a ucf file or setup description for touch or swipe ? 2. The power consumption seems very high for touch and swipe. Using the adapter board and Unico, with just the accelerometer and qvar running at 200hz low power mode, its about 13uA. I can see the qvar data plot working correctly but without the configuration of the mlc or fsm for touch. However, when touch is configured using the "Set Touch Detection Configuration' button, its running at about 52uA for touch and about 36uA for swipe. Does the mlc and fsm really use that much power for touch? Or are there some issues with the ST dev board in this scenario? Would setting the mlc odr lower help ? (but I'd need a way to reconfigure the mlc config, hence the need for the ucf or description)

Dear STM,is there a way to avoid the installation of a new version of NanoEdge AI when a new version is available ?Or maybe to pospone the installation ?When a new version is available I got a dialog Win like the one below, if I click on Exit then the app closes itself, if I click on Download then the application downloads the new version and it tries to install the new version. Unfortunately, based on specific company policies, I haven't the Admin priviledeges to install every new version when it becomes available.  thanks   Max 

INTERNAL ERROR: object of type 'NoneType' has no len()

I bought a STM32MP257F-EV1 board, but on the board UM, I can not find any description about AI/NN, I checked the STM32MP257F data sheet, it does support the AI/NN, so I am confused, if I can deploy the AI model on the EVB? 

I'm working on a mobile app that involves image processing and I'm considering using an STM32 microcontroller. Does anyone have experience with optimizing these for image editing tasks? I've been exploring photo editing apps like Meitu and would love to hear how similar capabilities can be achieved with STM32. Any insights would be helpful!

Hello,I am trying to run an Edge Impulse machine learning model for audio detection using STM's IMP34DT05TR microphone and the STM32G473VET6 microcontroller.I have imported the Edge Impulse C++ library, but I am encountering issues during inference such as 'Failed to run DSP process' and 'Failed to run classifier.'Can someone help with the steps required to successfully run the ML model, including interfacing the microphone and troubleshooting these errors?Any help is much appreciated!Thank you!

Hi, I have trained an .onnx model, using only conv1d, conv1dtrans and relu layers. But when I put it in the ST Edge AI Developer Cloud, the it has TOOL ERROR: tuple index out of range. The model structure is shown below, and please find the .onnx file in the attachment. Jinchen.