• @[email protected]
    link
    fedilink
    English
    40
    edit-2
    4 days ago

    If you are inferring 32 pixels from 1 pixel that is because the model has been trained on billions of computed pixels. You cannot infer data in vacuum. The statement is bullshit.

    • @loutr
      link
      English
      13 days ago

      Yeah but it does it much faster and more efficiently (according to him).