User avatar
Two Hollywood Phonies @autumn@cafe.autumn.town
1y
would there be a way to fine tune a model to detect bangers and not like trained on 1k posts
2
0
1
0
User avatar
max ✦ @max@duke.social
1y
@autumn you can like try to keep the first hundred posts but remove everything else in the post export json
1
0
0
0
User avatar
Two Hollywood Phonies @autumn@cafe.autumn.town
1y
@max was thinking like have a dataset labelled "funny" and "cringe" for 1k posts and then feeding the rest of the json into that
0
0
1
0