From the course: Grasshopper: Generative Design for Architecture
Unlock the full course today
Join today to access over 22,500 courses taught by industry experts or purchase this course individually.
Regression and predictive statistics
From the course: Grasshopper: Generative Design for Architecture
Regression and predictive statistics
- [Instructor] Regression Algorithms may be the most intuitive tool in the machine learning toolbox. They predate machine-learning all together playing a critical role in statistics since the early 1800s. Fundamentally, regression analysis extracts a transfer function from the results of previous test samples. This transfer function then provides a method to predict future output for untested input. There's a wide variety of regression strategies. We will look at some of the options included in lunchbox machine learning tool set for Grasshopper. When picking a regression strategy, typically, you should consider the number and type of inputs and the type of transfer functions to consider. Most importantly linear or nonlinear. I have the exercise file open already. In Grasshopper, you can see there's a data set here. This has been downloaded from a Github repository. The link is included in the script. In the data, we have height, weight, and age and gender for a whole bunch of…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.