Innovative Makerspaces: Diffusion Modeling for Makerspace Design

At the time of this writing we are still in the early stages of harnessing the potential of natural language processing and machine learning to support the creative design process. The contents of this book leverage machine learning prompt architecture for laboratory design. Just as modern scientific pursuits are a highly collaborative enterprise so is the modern design process. An increasingly complex relationship is evolving as the tools used for design begin to outpace the designers capacity for creative solutions. This work is a demonstration of the augmentation of the design process with machine learning focused on laboratory design. Using natural language processing on trained models of images to generate renderings is a new process where requiring an understanding of command prompt engineering. Command prompt engineering for diffusion modeling requires trial and error practice for how a string of words will be graphically translated by machines. Today’s designers are still approaching computational design as a drawing exercise and not as a poetic endeavor that can harness the potential of imagination with large data sets of images. The ability to index and have instantaneous accessibility to these data sets differentiates machine learning from human thinking, however comprehension remains in the realm of the natural language users. The idea of developing a design process for space that is based on natural language processing helps make design ideas more accessible for anyone trying to envision space. It is from this angle of adding comprehension to machine generated scenes that we approach the images in Innovative Makerspaces.