.tfs labels needed for st_yolo training

[dogg]

Hi,

I previously successfully trained an st_yolo model for the stm32h7 board using VSCode.

I now got the N6 disco board and cloned the new github page for modelzoo.

After running the stm32ai_mai.py program and having pointed the .yaml file to the correct training path, I get the following error: ValueError: Could not find any .tfs labels files in directory

My dataset is yolo formatted.

I did not get this error previously and am a bit confused by it. Am I in the right road at least and what should I do next?

thanks

[Julian E.]

Hello @dogg,

You can use this script to convert your yolo data to tfs files:

stm32ai-modelzoo-services/object_detection/datasets/dataset_create_tfs at main · STMicroelectronics/stm32ai-modelzoo-services · GitHub

Then in your yaml, use the path to these data for the training, validation etc.

I didn't know it was required to use tfs and it is not written anywhere, I am gathering information, I'll let you know.

Have a good day,

Julian

[dogg]

Ok thanks  for that. I have trained a mock model to test and now I am trying to load the model onto the N6 but I get this error instead: 

File "c:\Users\dogg\Desktop\zooooo\stm32ai-modelzoo-services\common\stm32ai_local\board_config.py", line 208, in __init__
raise STMAICFileError(f'"{project_path}" is not a regular file')
common.stm32ai_local.utils.STMAICFileError: E300: "../application_code/image_classification/STM32N6/stmaic_STM32N6570-DK.conf" is not a regular file

The deployment structure was originally for the H7 board so I had to change it to: 

deployment:

   c_project_path: ../application_code/image_classification/STM32N6/

   IDE: GCC

   verbosity: 1

   hardware_setup:

      serie: STM32N6

      board: STM32N6570-DK #STM32H747I-DISCO, STM32N6570-DK

thanks