[Question] How realize ranking (scoring) images?

Hi! I didn't have any problems, but I want to ask a question.
I don't know where it is better to ask a question, so I decided to do it in the developer repository.
Sorry if I shouldn't have done this here.

I train in ml and came up with a task for myself.
The essence of the task:
- **Find images with cars (Using ImageClassification with 2 labels)**
- **Then the images with the cars are additionally tagged, for example, brand, color, purity, etc.**

I think the best way would be to use a separate model for each tag. Check it out if I'm wrong.
I think that the task of determining the brand or model will be more difficult and decided to start with the ranking.

The question is how to implement image ranking in ml.net. Should I use ImageClassification with 10 labels (1, 2, 3, ...), which are the evaluation of the beauty/cleanliness of the car?
Or is there another more concise way to get the output of a single normalized number from 0 to 1 that represents the beauty/cleanliness score of the machine