a bunch of mastodon markov bots based on the entire corpus of text from various futurama characters
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
mark da651f689f remove unused config variable 7 months ago
character_models first working version 7 months ago
characters first working version 7 months ago
.gitignore first working version 7 months ago
README.md update readme with instructions for each file 7 months ago
UNLICENSE init 7 months ago
config.example.json remove unused config variable 7 months ago
corpus.json Clean up corpus 7 months ago
main.py first working version 7 months ago
process.py first working version 7 months ago
scrape.py init 7 months ago

README.md

the entire corpus of various futurama characters, turned into shitposting robots

transcripts source from the infosphere

the files:

scrape.py

scrapes the infosphere for the transcript files and downloads them, creating corpus.json which is a JSON file of the following format:

{ 
    "character_name": [ "line1", "line2", "line3", ... ],
    ...
}

Please use this responsibly, don't pound the wiki too hard thanks.

process.py

Takes all the characters in the characters list and pulls them from the corpus and generates markov model json files, suitable for use with Markovify

main.py

Generally run from the commandline like:

$ python main.py fry

Which will generate a sentence for fry and post it to mastodon according to config.json. Check out config.example.json for creating config.json

Be careful with your config.json, it contains a bunch of app keys.