Due: 5/6
This is an extra credit assignment. Completing this assignment will add one point to the rubric explained in 'Class Requirements and Grading' on the syllabus.
Pick either tutorial 12 or 13, and complete the following for each:
Tutorial 12: Spatial Intelligence
After completing the tutorial, create a new dataset just in part 1. Meaning we will just be creating a new map of manhattan with neighborhood boundaries, no embeddings.
To do this, you will primarily just need to edit the prompt that classifies the data. In the tutorial, the prompt classifies listing as either unit-forward (0) or neighborhood-forward (1), and maps the result. You could ask for another binary classification like that and then simply change the legend and titles in the next cells to fit, or you could try a totally different prompt, and update the code for the map.
When deciding, keep in mind what our data is: the text of about 1500 manhattan apartments. There is an a homogenous quality to it that I encourage you to embrace (e.g. I wouldn't recommend a prompt that asks where luxury apartments all, because all brokers call the apartments they list luxury!)
When finished, download your created map, and arrange it on a designed PDF page that includes the prompt that you used, a title, legend, and a pithy 2-sentence summary. Upload it to canvas.
Tutorial 13: Simulating Trees
After completing the tutorial, tweak some of the variables used in the model to create new predictions. What changes would create drastically different spatial trends? How much carbon could we sequester if every tree became 10 meters in diameter? How would you argue that these changes are in fact, valuable representations of real phenomena?
Don't worry, you certainly won't be the first or the last to paint fanciful realities using the aesthetic objectivity of simulations. Have fun with it.
When finished, download your created maps, and arrange them on a designed PDF page that includes a title, legend, a list of the variables you changed, and a pithy 2-sentence summary describing your simulation.