CubeMX AI 5.1.0 BUG: Cannot analyze the neural network
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content
‎2020-07-30 1:51 PM
Hi!
I have to report a bug. I cannot analyze the neural network in STM32CubeMX 1.3.0. I don't use 1.4.0 because they won't work with FreeRTOS.
How to repeat this issue:
- Start a new Project.
- Select STM32 F446RE Nucleo board
- Use CubeMX AI 5.1.0
- Download Mobilenet_V1_0.25_128_quant from https://www.tensorflow.org/lite/guide/hosted_models
- Download keras_cifar10_trained_model_100_epochs.h5 from https://github.com/ABurrello/CubeMX-AI-demo/tree/master/saved_models
- Download mobilenetv2-7.onnx from https://github.com/onnx/models/tree/master/vision/classification/mobilenet/model
As you can see, the mobilenetv2-7.onnx is the only file that can be analyzed with.
Thank you.
Operative system: Lubuntu Linux 20.04
Installation file: Generic Linux
IDE: STM32 CubeMX IDE 1.3.0
AI: CubeMX AI 5.1.0
STM32MP257FAK3 custom board with STM64-OS as operating system: https://github.com/DanielMartensson/STM64-Computer
- Labels:
-
STM32CubeMX
-
STM32F4 Series
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content
‎2020-11-25 1:19 AM
Dear @Daniel Mårtensson​
Could you please share you .ioc file and it's recommended to use the last available version of STM32CubeMX.
Best Regards,
Houssem
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content
‎2020-11-25 2:07 AM
Sorry. I don't have the .ioc file no longer.
You have to reproduce the issue by following these 1 to 6 steps.
STM32MP257FAK3 custom board with STM64-OS as operating system: https://github.com/DanielMartensson/STM64-Computer
