Royal HaskoningDHV launches its first open-source package called SAM for Smart Asset Management and it has been used as an internal Python package for years.
Time series data and anomaly detection
SAM is a Python toolbox to work with time series data and it is used by data scientists, data engineers and data analysts. It contains numerous functions and classes for data loading, preprocessing, validation, visualization and feature building as well as creating machine learning models for forecasting and anomaly detection. SAM makes creating machine learning models easier by simplifying the creation of rolling features and time features. It also allows the users to easily create models that predict multiple steps ahead or predict quantile values, which can be used for anomaly detection.
Data Science meetup
If you’d like to know more about the details of this package, we will present it in more detail during the next Royal HaskoningDHV Data Science meetup, which is planned for October 21st at our office in Amersfoort.
Royal HaskoningDHV has decided to make this package open-source and available to all, because it allows us to work together more effectively and we want to share our experience with time series analysis with the rest of the open-source community. At the same time, we are also working on our Machine Learning as a Service (MLaaS) platform, that will make it easy to deploy multiple models created with SAM.
