This first assignment will help to familiarize you with the Python programming language and Google Colab notebooks. For this assignment you will primarily be running code examples, but you will also write a little bit of Python code yourself.
The Colab Notebook for HW0 is available here.
You should make a copy of this Notebook and save it to your own Google Drive (File > Save a copy in Drive) in order to run and edit the Notebook. The data sets for this assignment can be found here.
Please see Canvas for detailed submission instructions.
Homework 1 covers methods for Unsupervised Learning. This assignment is split into two Colab Notebooks. Part A will cover methods for cluster analysis, while Part B will focus on linear dimensionality reduction methods.
The Colab Notebook for HW1 Part A is available here.
The Colab Notebook for HW1 Part B is available here.
You should make a copy of this Notebook and save it to your own Google Drive (File > Save a copy in Drive) in order to run and edit the Notebook. The data sets for this assignment can be found here.
Please see Canvas for detailed submission instructions.
Homework 2 covers baseline models for Supervised Learning. In the first activity, you will train linear regression, ridge regression and LASSO regression models to predict wind and solar energy production based on weather variables. In the second activity, you will use K-Nearest Neighbors to classify birds based on their bone measurements. In both activities, you will use cross-validation to select model parameters.
The Colab Notebook for HW2 is available here. See Canvas for submission instructions.
Homework 3 covers classification and regression methods including SVMs, neural networks, decisions trees and random forests. Activity 1 will compare multiple classifiers on the same data set. Activity 2 will focus on tree-based methods. Activity 3 is an open-ended landcover classification problem where students will select the classification methods to use to create their prediction model.
The Colab Notebook for HW3 is available here. See Canvas for submission instructions.