Here we will be using random forests to predict a bird's species. We will use the Caltech-UC San Diego dataset (http://www.vision.caltech.edu/visipedia/CUB-200-2011.html), which contains about 12,000 photos of birds from 200 different species. Here we are not going to look at the pictures because that would need a convolutional neural network (CNN) and this will be covered in later chapters. CNNs can handle pictures much better than a random forest. Instead, we will be using attributes of the birds such as size, shape, and color.
Here are just some of the species in the dataset:
Some, such as the ...