We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Basically a subject is the data being trained on (images only right now)