Re:Exploring 12M of the 2.3B images used to train Stable Dif...
by rom1504 on Wednesday August 31, 2022 @07:49AM
(#32660960)
This is due to the aesthetic scoring in the UI.
Simply disable it if you want precise results rather than aesthetic ones.
It works for your example
I guess I'll disable it by default since it seems to confuse people
Re:Exploring 12M of the 2.3B images used to train Stable Dif...
by rom1504 on Wednesday August 31, 2022 @07:53AM
(#32660997)
Done https://github.com/rom1504/clip-retrieval/commit/53e3383f58b...
Using clip for searching is better than direct text indexing for a variety of reasons but here for example because it matches better what stable diffusion sees
Still interesting to have a different view over the dataset!
If you want to scale this out, you could use elastic search
Re:Exploring 12M of the 2.3B images used to train Stable Dif...
by yreg on Wednesday August 31, 2022 @09:34AM
(#32661860)
I see, thanks! I didn't realize that as I thought I want to keep aesthetic scoring enabled since Stable Diffusion was trained on LAION-Aesthetics.
---
Also: There is a joke to be made at Jony's expense regarding the need to turn off aesthetic scoring to see his face.
When my prompt isn't working, I often want to check whether the concepts I use are even present in the dataset.
For example, inputting `Jony Ive` returns pictures of Jony Ive in Datasette and pictures of apples and dolls in clip retrieval.
(I know laion 5B is not the same as laion aesthetic 6+, but that's a lesser issue.)
[0] - https://rom1504.github.io/clip-retrieval/
[1] - https://laion-aesthetic.datasette.io/laion-aesthetic-6pls/im...