Problem with "train.py" file ModelZoo

[Diogomv]

When I try to launch the "train.py" file, in conda virtual environment it gives me an error "ModuleNotFoundError: No module named 'logs_utils'". This is a module called on the code line 23 of the train.py.

 

After that I've proceed with a "pip list" to see the modules instaled, but indeed the 'logs_utils'" isn't there.

I've tried to install it with "pip install logs_utils" but it simply can't find the module.

I've google it and the only module that it seems to exist is the "log_utils".

Any one knows how to turn around this?

Thank you.

 

[absyed1]

it is a custom library by ST located at \stm32ai-modelzoo\common\utils, not the log_utils from standard python library. Try running the demo from notebook stm32ai_model_zoo_colab.ipnyb located in \stm32ai-modelzoo\tutorials\notebooks\. Running train.py is not successful on my system. I have run and deployed using the notebooks. The deploy notebook is at \stm32ai-modelzoo\image_classification\imageclassification_deploy.ipynb. Though the model is not performing as good. 

[Diogomv]

Thank you for your time. Indeed it seems there's some problem with this software pack by ST, becouse I've installed all the dependencys like in the instructions on git repo .Readme file but can't runing it.

[absyed1]

Try running form notebook. it should run.

[Laurent FOLLIOT]

Hello,
This is not a Pb with the software pack.
To do some training, you must not run train.py, but go in /src folder and run the stm32ai_main.py script and use config yaml file. Please don't hesitate to refer to the readme for more details or to config_file_examples folder that provides minimalistic configuration examples

.