This is the first of a two-part lesson introducing deep learning based computer vision methods for humanities research. Using a dataset of historical newspaper advertisements and the fastai Python library, the lesson walks through the pipeline of training a computer vision model to perform image classification.