Managing Projects

KEY TERMS

Projects are the basic framework that allows you to organize and structure your images on the Deepomatic platform. Each project therefore corresponds to a precise set of images, and therefore a given subset of your visual universe.

Concepts are the way to decompose and describe your visual universe. The term is actually a shortcut, as you should think more about visual concepts, i.e. concepts that have a visual unity of their own.

Project listing

The home page for the Projects section is the listing of all the projects that have been created for the organization.

You can either select one of the projects that have already been created or create a new one.

Create a new project

To create a new project, you need to click on Add a new project and choose a name for it.

Now you can select the type of the project you want to create:

  • Classification: each image is labelled with exactly one concept and all concepts are then exclusive.

  • Tagging: each image is labelled positively or negatively for all concepts. A single image can then validate several concepts (or none).

  • Detection: concepts are localised in the image with bounding boxes. The information is then much richer as you can get several instances of a same concept in the image.

For more information on the different kinds of project, see our Guidebook on how to build your custom AI or our blog post series "AI for Dummies".

To design your image recognition application, you need to specify the concepts that you want to recognize. You can do so by clicking on New concept and giving a name to your new concept. Don't worry, you can add more concepts and/or edit them after you have ended the project creation process.

You can also use existing concepts by importing a JSON file in which the concepts and images are defined. To do so, you must precise the field "tags" in the JSON. Check the JSON Upload page to know more about the JSON format accepted.

Project Home Page

When selecting a given project, you arrive on its Home Page.

Here are the different elements that you will find on this page:

  • A navigation bar on the left hand side to navigate through all main functionalities.

  • On the central panel, you can see the number of images and the image distribution for each of your concepts.

  • Finally, you have the validation split function, which enables you to randomly select a portion of your images and put them in your validation set.

  • By entering an integer strictly greater than 1, you can also have a random validation split that will make sure to send the corresponding number of images in your validation set.

After changing your validation set, it is no longer possible to compare two models.