1. What questions do you still have about the model and the associated data? Are there elements you would propose including in the biography?
   1. I still would like to know what types of people doing the labeling are, where they are from, age, gender etc. It is important to know this to be able to understand the biases in the labelling set. I would like to see information relating to the people doing the labelling in the biography. I also would like to know the difference between CoCoSSD and the default ML5 image recognition algorithm.
2. How does understanding the provenance of the model and its data inform your creative process?
   1. It is good to know how old a data set is as it can help with keeping my expectations in line. It is also good to know who labelled the images and how - which can inform my understanding of the biases that exist in the model. Additionally, it is useful to know what the model is built for and what use cases it is made for. And finally it is good to know if models are built on the same platform (like the multiple shown that were built on TensorFlow), which helps if I wanted to combine them together.

https://editor.p5js.org/liam-mahogany/sketches/JpSosP2SU

2022-09-30 14-58-47.mp4

I wanted to create a sketch that would allow me to control the sketch using my hands. I landed on using posenets wrist variable to change the height and direction of a rectangle, emulating a volume or loading bar.