Skip to main content
Unlisted page
This page is unlisted. Search engines will not index it, and only users having a direct link can access it.

Webinar 27.11.2025
Build Your Own AI Quality Control in <1 Day

Materials​

Your task​

For the second part of the webinar, you will build an AI Quality Control system using One AI.

  1. Set up OneWare Studio: Follow the instructions in our setup guide to install OneWare Studio, the OneAI extension and to create an account.
  2. Download the exercise dataset
  3. Create a new project by clicking on AI > OneAI Project Generator
  4. Import the exercise dataset as a COCO dataset. You can check whether the import was successful by clicking on one of the images and checking the annotations.
  5. Set up the dataset splits:
    • The dataset doesn't contain a validation split, so we leave the default settings for creating one turned on.
    • We do have a test split, so we turn off the option to create a test split.
  6. Set up filters and augmentations: The default settings will already create a good starting point, but we can improve the model by tuning the filters and augmentations to our application.
  7. Configure the model settings: provide additional information about the application to obtain the best possible model.
  8. Train the model for 5 minutes.
  9. Test the model and look for cases where the performance can be improved.

Additional resources​