pytorch-widedeep is a python package from London based machine learning specialist Javier Rodriguez Zaurin. It can be used for building neural networks that use structured (tabular) data combined with images and text. It is based on Google’s 'Wide and Deep' algorithm and uses the PyTorch framework. The algorithm was first used to power the recommender system of Google's Android Play store.

Over the last decade, neural networks have achieved dominance in the analysis of text and images. However, they have tended to lag behind on structured data like you might find in a SQL table or an Excel spreadsheet. I am not keen on the increasing tendency to use neural nets for everything. They are good at some things but are often not the best choice. However, sometimes I want to be able to combine data types like images and text with structured data. This approach should be useful for doing this.