Skip Navigation

SweetAIBelle
SweetAIBelle @ SweetAIBelle @kbin.social
Posts
0
Comments
4
Joined
2 yr. ago

‘Star Trek: Lower Decks’ Cast Talks About The Big Change For Season 4

Jump

  • I'd have loved something like that.

    I'd personally have liked them to have not let that go, and have a whole character arc with Harry struggling with the knowledge that none of them are the same people he knew, though they act identically, differentiating between things he did and the original and so on...

    Resetting back to status quo after events that should have had a lasting impact is one of these things that bothers me in Voyager.

    • ‘Star Trek: Lower Decks’ Cast Talks About The Big Change For Season 4

    Jump

  • Well, to be fair, their original Harry Kim is dead, making the current one the newest member of the crew until 7 of 9 shows up.

    And for all we know, the ship records may still show him as dead, therefore unpromotable...

    • How Many Star Trek Episodes Pass the Bechdel Test? (TOS to ENT) | The Mary Sue

    Jump

  • Well, you know, thinking about it, an awful lot of Kes's life revolves around Neelix and the Doctor, with occasional training by Tuvok. I suspect a lot of her dialogue was about one of the three...

    • Greg Rutkowski Was Removed From Stable Diffusion, But AI Artists Brought Him Back - Decrypt

    Jump

  • Generally speaking, the way training works is this:
    You put together a folder of pictures, all the same size. It would've been 1024x1024 in this case. Other models have used 768z768 or 512x512. For every picture, you also have a text file with a description.

    The training software takes a picture, slices it into squares, generates a square the same size of random noise, then trains on how to change that noise into that square. It associates that training with tokens from the description that went with that picture. And it keeps doing this.

    Then later, when someone types a prompt into the software, it tokenizes it, generates more random noise, and uses the denoising methods associated with the tokens you typed in. The pictures in the folder aren't actually kept by it anywhere.

    From the side of the person doing the training, it's just put together the pictures and descriptions, set some settings, and let the training software do its work, though.

    (No money involved in this one. One person trained it and plopped it on a website where people can download loras for free...)