Lecture 6 - A Hands-on Training for Machine Learning in Python

Image Classification of Drone Imagery

In this course you will use a training dataset from Radiant MLHub repository with drone imagery of crops in Rwanda, and build a classification model to identify the crop type of each image.

To learn more about the dataset, checkout its registry page on Radiant MLHub.

Python Packages Used

• radiant-mlhub
• pandas
• numpy
• matplotlib
• sklearn
• tensorflow
• keras
• seaborn

Homework

After this lecture, you are encouraged to work on two other image classification datasets available on Radiant MLHub.

1. SEN12-FLOOD: This dataset contains imagery of Sentinel-2 and Sentinel-1 satellites with labels for presence of flood in the imagery.

2. BigEarthNet: This is a land cover classification dataset, and it contains ~570,000 images with multiclass labels for each image. This is a very large dataset, and if you are running it on a local computer, you may need to select a subset of the data for training.