THINGS THAT SHOULD HAVE RHYMED 
A custom generative solution combining classical generative algorithms with a trained machine learning model, TTSHR is the name of the solution as much as the final series of 300 artworks itself. TTSHR uses a machine learning model trained on a custom generative dataset to output a simple, abstract combination of strokes of colour. These strokes are imposed onto a generative grid that makes up the canvas composition of each artwork.

This underlying grid is accompanied by a set of traits that make up the aesthetic settings for each of the 300 artworks. Across the artworks a variety of colours come into play, defining the background, strokes and details of each piece. Examples demonstrate the variety, from monochrome artworks to multi-coloured ones.