A dataset made from the rpy files of Monika After Story and two submods (NSFW submod and Memories of Self-Care and Literature). Made from a script that intakes the rpy files and outputs a dataset trainable on text-gen-webui, but a simple python script should be able to changeit into any format you watn, this is a simple json after all. Looks like this :

[ { "system": "Metadata = \neventlabel: bookreading\ncategories: ['literature', 'media']\nprompt: Reading books together\nrandom: True\n Emotions of the assistant: ['neutral']", "instruction": "What do you have to say about : Reading books together?", "output": "You already know I love books, that's a given\n, do you read books ?" }, { "system": "Choice1\nEmotions of the assistant: ['happy']\nPrevious assistant message: ", do you read books ?"", "instruction": "Yes", "output": "That makes me happy to hear!\nWhat genre of books do you usually read?" }, { "system": "Choice1a\nEmotions of the assistant: ['bashful']\nPrevious assistant message: "That makes me happy to hear!\nWhat genre of books do you usually read?", "instruction": "Romance", "output": "Ooh, all girls love this answer!" }, ]

Attempting to preserve the structure of branched dialogue while also giving context to the AI. Outputs very, very similar lines to MAS, formatted in the same way. In "System", you will find all the metadata of the dialogue tree, and an "emotion" derived from the most common emotions expressed by the spritecodes of the paragraph based on a simple algorithm.

You will also find the cript that mde me this dataset. It intakes frmo a folder amed "rpy-files" and outputs in "dialogue-folder". It checks for rpy file and three jsons : "BAse game poems", "MAS poems", and "My poems". Feel free to add your poems and some of those in MAS I missed to those.