Ask questions and find answers on how to deploy, debug, and optimize AI models on ST microcontrollers, microprocessors, and smart sensors.
I am using the STEVAL-STWINKT1B evaluation board to gather microphone data with a 192kHz sampling rate. The data is saved in a .dat file which i want to read and process. However, when reading the IMP23ABSU_MIC.dat file i get completely different res...
Hello everyone, I have a pytorch model (mymodel.pth) obtained by running the mixed-precision quantization of this algorithm: * https://github.com/eml-eda/q-ppg. I followed the instructions in the readme section.Python models are not natively supporte...
Hi, I have a self-trained image classification model with two classes with dynamic range quantization(unquantized model is about 1.5mb and it is 500kb after quantization). And this model takes too much flash and ram after benchmarking. Is there any m...
Hi, I am trying to create an STM32Cube.AI application using X-Cube-AI. I am following the documentation under Content 6. I trained my own model and configure the .ioc file as shown below. I wanted to get input from the ADC and feed directly into the ...
Hi folks,I gently ask you some help related to a computer vision application, based in artificial inteligence, that I have been developed at STM32H747i-DISCO. I am developing my code following the steps presented in a stm tutorial that is available ...
Hello, I have implemented https://wiki.st.com/stm32mcu/wiki/AI:How_to_perform_motion_sensing_on_STM32L4_IoTnode this project. When I run the program I had a hard fault error when function enters ai_platform_network_process(). I have incremented the...
Hi, I am currently using SensorTile.box Pro with FP-SNS-DATALOG2 FW. After acquisition, I would need to convert the data from the microphone from .dat to .wav and to do this I am using the hsdatalog_to_wav.py example provided. Unfortunately, the resu...
Hi, I am using NanoEdgeAIStudio to detect anomalies on microphone signal. I am particularly interested in logging ultrasounds (from 20kHz to 100kHz). I used the Datalogger tool (in NanoEdgeAIStudio) to set up the datalogger and generate the neai_data...
The deploy.py script in stm32ai-modelzoo/image_classification/getting_started generates an extra file, ai_model_config.h which is convenient as it contains the class names and other usefull defines.Is there a way to generate this file from cubeMX?Mor...
I used a quantization tool called ppq to quantize my onnx model because the activation values in the middle layer of this model were too large and I needed to quantize it from float32 type to int8 type. By the way, this model was defined and trained ...
