From the course: Grasshopper: Generative Design for Architecture

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

Clustering

Clustering

From the course: Grasshopper: Generative Design for Architecture

Start my 1-month free trial

Clustering

“

- [Instructor] Clustering is one of the most flexible techniques in the machine learning toolbox. It's easy to apply, and requires no sample data or predetermined classifications. As a result it's often used as a starting point to organize data sets, or as a way to sort continuous data into useful chunks. For example, grouping customers by preference to target marketing. In this video, we will look at two example clustering problems to see how simple the technique is to set up, and how careful data input design can facilitate more sophisticated clustering. I have the exercise file open already. See in Grasshopper, I have 2D clustered points. Roughly grouped into three circular areas. In this first example, we're going to use the gaussian mixture component. So I'll go to LunchBox, machine learning, gaussian mixture. Look at the component, it has an input, which is the original data, the points, the number of components, which is how many groups to create in a random seed. So we'll go…

Contents