2018 icon indicating copy to clipboard operation
2018 copied to clipboard

National Novel Generation Month, 2018 edition.

Results 100 2018 issues
Sort by recently updated
recently updated
newest added

Markov mashup of trail race reports and psychoactive experiences. [Example output](https://github.com/nmifsud/ultra/raw/master/ultra-181129-075926.pdf). [Repo](https://github.com/nmifsud/ultra). Opening sample: > I never stopped to dig some food out of camp as quickly as possible. A...

completed
preview

Her Majesty's Aviary is an ornithologist's guide to the (fictional) seasonally rotating collection of birds held by Empress Yezida the Third, the Radiant, the Dove of Yezid. It was created...

completed
preview

I'm going to start off with something easyish, and then see what follows. --- First by reproducing some historical procgen shared by [James Ryan](https://twitter.com/xfoml/status/885584010411253761): > extremely early #procgen: Pfizer used...

[Repository](https://github.com/knoxdw/nanogenmo2018), [text](https://raw.githubusercontent.com/knoxdw/nanogenmo2018/master/makeshift.txt), and [pdf](https://raw.githubusercontent.com/knoxdw/nanogenmo2018/master/makeshift.pdf). This work arranges sentences and phrases from historical sources, making use of sentence similarity calculations based on word embeddings provided by the [SpaCy](https://spacy.io/) NLP library in...

completed
preview

I'm in. Zero ideas about what I'm gonna do and how I'll do it. But still.

completed

I finally got around to reading M. R. James' anthology Ghost Stories of an Antiquarian. While his stories do indeed have the same structure as most of Lovecraft's non-Dreamlands stories...

This book generator is about people, animals and choices.The idea arises from a game of eloquence I did weekly, about thirty years ago, when I practiced with my fellow salesmen...

completed
preview

Develop a new take on the Sherlock Shuffle, using my [POS Tagger that outputs Tracery grammars](https://www.npmjs.com/package/pos2tracery)... and then see what else I can come up with... I really liked combining...

I have a more, what I consider, painterly method in mind. Lots of layers, lots of curation. What I want most is to have the results readable, if not compelling....

completed
preview

This is the classic RNN generator approach. I use the pretrained [ULM-FiT](http://nlp.fast.ai/category/classification.html) model and fine-tune to a corpus of all creepypasta stories. I select the next word randomly weighted by...

completed
preview