mastodon.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
The original server operated by the Mastodon gGmbH non-profit

Administered by:

Server stats:

366K
active users

this is a little weird and worked a little better than I expected it to: using scikit-learn to fit gaussian mixtures to the vectors for all words following n-grams in the text, then doing nearest-neighbor lookups on the concatenated vectors for the n-grams, nearest-neighbor lookup on a sample from the mixture model, and chaining predictions like a markov chain

Allison Parrish

"trained" on _Frankenstein_. the idea of using a gaussian mixture model on the word vectors is to make it possible to predict words similar in meaning that aren't necessarily in the original text. I'm not sure if the variability in the text is due to (a) not enough data to really get the mixtures right (b) the lack of ability to control the degree of randomness in the sample function or (c) you shouldn't really do this kind of thing with word vectors

@aparrish *twitch* *grind teeth* "You mean 'trained on Frankenstein's Monst--" *wrestles self to ground, strangles self*

@Borogove i mean the monster does a lot of talking in the novel for sure, so your self-strangled statement is at least partially true