Hands-Off Machine Learning with Google AutoML

Tabular data is omnipresent nowadays and can provide us with meaningful insights into both business and engineering problems. A common way of extracting these insights is by applying machine learning (ML) techniques to this data. The process of applying ML to a dataset consists of various steps, e.g., data preprocessing, feature engineering, and hyper-parameter optimisation, with each of these steps often being a time consuming trial and error process in and of themselves. Additionally, one needs to be an expert in the domain of ML in order to be efficient and effective at each of these steps. It can take quite some time for an organisation to either find these domain experts externally, or grow this expertise in-house.