Edge AI

We are using the Nano Edge studio to run the STWIN DTMF Classifier on the STEVAL-STWINCSV1 on a Linux system. We followed the instructions from: https://wiki.st.com/stm32mcu/wiki/AI   Entitled:  How to create a dual-tone multi-frequency classifier using NanoEdge AI Studio When we run and capture 100 lines of samples, we move to the "Import Signal" and 99% of the data is misaligned and classed as corrupted. We have a space as our separator, but we need a new line to separate each line and this causes corruption. We also noted that on occasion we have, as an example, 0.0.000. We have tried modifying the code in the classifier example: for( uint16_t class_cpt = 0; class_cpt < CLASS_NUMBER; class_cpt++ ){printf( " %.3f", output_class_buffer[ class_cpt ] );}printf( "\n" ); But no matter how we format the code the date is always classed as corrupt. We are unable to use the button to "Open file location" as it does nothing. 1) Could you please tell us where we would find the outp

As stated, I have problem analyzing the ONNX model ported from a pytorch model. The code of the model is simple as follows: class LSTM(nn.Module): def __init__(self, input_size, hidden_size, num_layers, num_output=1): super(LSTM, self).__init__() self.num_layers = num_layers self.input_size = input_size self.hidden_size = hidden_size self.num_output = num_output self.lstm = nn.LSTM(input_size=input_size, hidden_size=hidden_size, num_layers=num_layers, batch_first=True) self.fc = nn.Linear(hidden_size, num_output) def forward(self, x, device='cuda'): ula, (h_out, _) = self.lstm(x) out = self.fc(h_out[-1]) return out  Netron vis:I will really appreciate any advice or workarounds. Thank you!

Hi ,I go into the hardfault handler when trying to run the ai_network_run which I call inside my run_inference function which is inside my ADC ISR , then it calls ai_platform_network_process , that is where we get thrown into the HardFault Handler . here is my code in case anyone has improvements please don’t hesitate to reply .Note : I tried all the previously mentioned suggestions like clocking CRC and increasing stack size , I increased both stack size and heap size to 0x8000 .   /* USER CODE BEGIN Header */ /** ****************************************************************************** * @file : main.c * @brief : Main program body ****************************************************************************** * @attention * * Copyright (c) 2024 STMicroelectronics. * All rights reserved. * * This software is licensed under terms that can be found in the LICENSE file * in the root directory of this software component. * If no LICENSE file comes with this

 I built a simple FCNN model based on onnx(regression, 4 input and 4 output), including three hidden layers with 64, 32, 16 nodes, and successfully "validate on target" in STM32 through X-CUBE-AI.Now I want to send(receive) data directly to(from) the STM32 via the serial port, instead of clicking the button "validate on target". However, I don't know how the data from the npy file was sent during the "validate on target" process. Is there any documentation to explain the process?

When I used V4.3.1 before, there was a "5 Emulator" step, but it doesn't exist in V4.6.0.Is Emulator no longer available?

Dear community members,Here’s your chance to win some prizes from ST until the end of 2024! You can join the challenge by sharing your ISPU neural network implementation. We’ll provide instructions and hints during the ISPU webinar held on October 29, 2024. https://content.st.com/iot-applications-with-neural-networks-in-mems-sensors.html Your proposal must relate to embedding a neural network on an ST MEMS sensor with ISPU.The ISPU is an ultralow power, computationally efficient, high-performance programmable core that can execute signal processing and AI algorithms in the edge. It is available in the LSM6DSO16IS and ISM330IS sensors from STMicroelectronics. Simply create a comment in this thread. You can also attach files to document your solution (.log, .mp4, .docx, .jpg, .png, .xlsx, .7z, .hex, .c, .cpp, .zip). Use the “kudos” button on ideas or implementations from other members to indicate that you like them. Everybody can vote on implementations and ideas, and the most voted on

I am trying to use MEMS Studio to configure the Machine Learning Core (MLC) of the ISM330DHCX sensor on both the B-U585I-IOT02A and a custom board, but after installing MEMS Studio and connecting the board, I realized that it is not recognized by the software. How can I use MEMS Studio with a board that is not on the supported list? If this is not possible, what is an alternative way to utilize the MLC functionality of the ISM330DHCX? *This post has been translated from Korean to comply with the ST Community guidelines.

I was able to run the Serial emulator in NANOEDGE AI STUDIO Validation, so as the next step, I run [COMPILE LIBRARY] in Deployment and don't know where to put the libneai.a file generated in Project Explore.Also, please let me know if any other settings are required.The evaluation board is B-L475E-IOT01A2.

Hi everyone,I am working on implementing an image classification model on an STM32 board using the X-CUBE-AI extension. I am relatively new to embedded software development; my background is primarily in AI and deep learning. While I have managed to generate the necessary .c and .h files for my model using the X-CUBE-AI extension, I am having trouble providing input to the model from an image stored on an SD card. As far as understand from created cods I have modify acquire_and_process_data and post_process functions that have write down below. Also I have provide output of my code. In main function just  MX_X_CUBE_AI_Process function works in while for AI proces.Here’s a brief summary of what I’ve done so far:Model Conversion: Successfully converted my image classification model using the X-CUBE-AI tool and generated the necessary files.SD Card Integration: Implemented functions to read BMP image files from the SD card and store them in a memory buffer. These functions

The STM32N6 has been officially launched yesterday at the STM32 Summit!      The STM32N6 is based on the Arm® Cortex®-M55 running at 800 MHz. It is the first STM32 MCU to embed the ST Neural-ART accelerator™, an in-house developed neural processing unit (NPU) engineered for power-efficient edge AI applications.   A dedicated computer vision pipeline with a MIPI CSI-2 interface and image signal processing (ISP) ensures compatibility with a wide range of cameras. The STM32N6 also features an H264 hardware encoder and the NeoChrom™ Accelerator for graphics, making it suitable for feature-rich products.   4.2 Mbytes of contiguous embedded RAM are also offered, ideal for neural networks or graphics applications, complemented by high-speed external memory interfaces (hexa-SPI, OCTOSPI, FMC).   The full AI ecosystem is out as well! ST Edge AI Core v2.0: introducing the STM32N6 with Neural Art, and also the Stellar P ST Edge AI Developer Cloud sup

Hi,We are trying a very simple deep neural network model with 20 inputs and 3 outputs and want to generate inferences in real-time. We are following the instructions from the document "Getting started with X-CUBE-AI Expansion Package for Artificial Intelligence (AI)" numbered UM2526 with X Cube AI 7.1.0 with STM32CubeMX.Has anyone encountered the same error?Thanks in advance.

I am currently using the STM32 Edge AI Developer Cloud to convert my ONNX model, which I built using PyTorch.The ONNX model is attached for your reference.I would greatly appreciate your guidance on this matter, as it is blocking my research.Thank you for your assistance.  

Dear Sir/Madam,I have a trained CNN model converted to .TFlite file. The file has been analysed and validated on PC and Target. All are correct. Then an application template was developed. When I tried to run initiate CNN model with generated function " static int ai_boostrap(ai_handle *act_addr)", there is issue in this layer:ai_bool cnn1_configure_activations(ai_network* net_ctx, constai_network_params* params) (Can not go deep anymore).  Always return false.  From ai_cnn1_init() returned an error type 48 and error code 19 as shown in the screenshot attached.Would you please help me sorting out this issue? QSAppendix:AI_DECLARE_STATICai_bool cnn1_configure_activations(ai_network* net_ctx, const ai_network_params* params){AI_ASSERT(net_ctx) if (ai_platform_get_activations_map(g_cnn1_activations_map, 1, params)) {/* Updating activations (byte) offsets */ serving_default_Combined_Input0_output_array.data = AI_PTR(g_cnn1_activations_map[0] + 632);serving_default_

..

Hi, When ST will release more information about the STM32N6? Will it use the new cortex-m55? Can it be used in motor control applications?

Hello everyone, I am trying to use the ST Edge AI Developer Tool and I am encountering the following error when I try to quantize my model (a RNN with an input layer, 3 GRU layers and a Fully Connected layer):    2024-12-11 07:25:35.451880: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. 2024-12-11 07:25:35.479788: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. ONNX Runtime Model Optimization Failed! Consider rerun with option `--skip_optimization'. Traceback (most recent call last): File "/usr/local/lib/pytho

Good afternoon,I want to upload a Machine Learning algorithm on the ISPU part of the LSM6DSO16IS. My alghoritm has to identify is the subject is stationary, walking or running. I used NanoEdgeAI Studio to develop the classification algorithm. By doing this I obtain these files, but don't the .ucf file, which is the file that I have to upload on the ISPU part of the sensor. By uploading this files on the Nucleo, it performs the classification correctly, but I want the classification to be done by the sensor. Which is the easyest way to obtain it starting from the files that I obtained?  

Dear Sir/Madam,I am trying to deploy a trained AI model (CNN+LSTM) to STM32F767. When I tried to analyse the TFlite file with the STM32CubeMX, it stopped and gave the following message: Could you please help me sorting out this issue?Thanks QS  

 /* ============= Copyright (c) 2023, 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.

Hi,I'm working with X-CUBE AI.The code seems working but I have a doubt:status_init = ai_network_create_and_init(&network_handle, act_addr, NULL);Is NULL parameter correct ?Why souldn't I need to pass a pointer also for weigths like 'act_addr' ?If it is required how am I supposed to achieve this point without occcupy other RAM space ? (Note that inference output is actually weird, but it could be correct cause I used mockup input data to feed the network)Please, take a look to code and log below:  ....network_data_params.h.... #ifndef NETWORK_DATA_PARAMS_H #define NETWORK_DATA_PARAMS_H #include "ai_platform.h" /* #define AI_NETWORK_DATA_WEIGHTS_PARAMS \ (AI_HANDLE_PTR(&ai_network_data_weights_params[1])) */ #define AI_NETWORK_DATA_CONFIG (NULL) #define AI_NETWORK_DATA_ACTIVATIONS_SIZES \ { 306432, } #define AI_NETWORK_DATA_ACTIVATIONS_SIZE (306432) #define AI_NETWORK_DATA_ACTIVATIONS_COUNT (1) #define AI_NETWORK_DATA_ACTIVATION_1_SIZE (306432) #define

I am trying to validate a CNN on STM32CubeMX with X-Cube-AI 9.1.0. I provided a datainput.CSV and a dataoutput.csv files to  "Validation inputs:  & Validation outputs: " respectively. When ran Validate on desktop or Validate on target, they all failed >>  "E204(ValParamError): number of data files is not enough, 4 are expected."Could you please help me sort out this problem?Do data files in .csv  correct?Only have 2 boxes for 2 files providing, why expected 4? QS 

First , the github project is thishttps://github.com/Xiaobin-Rong/gtcrn/tree/main   It has converted to onnx and tensorRT for ppl to usebut when i load the onnx model into stm32 for optimize/import (setting , compression : none , optimization : balance)it just jumps out an error X-cube-AI version is 9.1.0     ST Edge AI Core v1.0.0-19894 INTERNAL ERROR: Trying to remove node_339 which has multiple inputs     is it because the multi input and output network limitation or something else?and if i success to import this modelhow to deal with the multi dimensional  tensor arrays that the github author demo ?   if you want to test on the modelthere is the link for the onnx and tensorRT modelhttps://github.com/Xiaobin-Rong/TRT-SE/tree/main/online/models   Thanks in advance     Uveuvenouve

Hi everyone,   I have a STWIN.BOX (STEVAL-STWINBX1) and I have played around with the code examples in Python provided on the tfp-sns-datalog2 repository. The idea is to use this device to provide readings to a Raspberry Pi board and make some decisions based on the data. I am not an expert in Python, but I coded a simple script to read the sensors using the repository, getting the hsdatalog_cli.py as an example, and using this data as input to the Raspberry Pi board. The thing is, I can read the sensors only if I start logging (using the start_log method provided by the script hsdatalog_cli.py), and since the log has started, it will generate .dat files. So, every time I run the algorithm inside the Raspberry Pi, new .dat files are generated, and the file sizes depend on how long the board is running the algorithm. I have tried to get data from the sensors without starting logging, but I have failed. Is there any way of getting data from the sensors without generating the dat fil

I am encountering an issue when deploying a classification model generated with NanoEdge AI Studio 4.6.0 to my STM32H7 development board. Below are the details:Issue: When calling neai_classification_init(knowledge), the function returns NEAI_COMPILER_ERROR. Debugging further, I found the following comment in the generated NanoEdgeAI.h file:  enum neai_state { NEAI_OK = 0, NEAI_INIT_FCT_NOT_CALLED = 123, NEAI_BOARD_ERROR, NEAI_KNOWLEDGE_BUFFER_ERROR, NEAI_NOT_ENOUGH_CALL_TO_LEARNING, //This is a fail-safe to prevent users from learning one or even no signals. NEAI_MINIMAL_RECOMMENDED_LEARNING_DONE, NEAI_UNKNOWN_ERROR, /* Error due to a potential compatibility issue affecting user of the ARMCC toolchain. * Contact our support for further details or consider using a different toolchain */ NEAI_COMPILER_ERROR };  Setup:Toolchain: Keil MDK 5.37 with ARMCC (AC5) compiler.Initialization Code: Generated with STM32CubeMX.NanoEdge Versions:Models generated with N

When I select the library in the [Validation] tab of NANOEDGE AI STUDIO and run [Serial emulator], I get the error message "No data are being received. Please ensure the correct COM port is selected."However, I am able to obtain data if I select the same COM port using TeraTerm (terminal emulator).Incidentally, the same error occurs whether the baud rate is 9600 or 115200.The board I am using is B-L475E-IOT01A2.Does anyone have any information on the cause of this?