Simplifying Hyperparameter Tuning in Online Machine Learning -- The spotRiverGUI
This tool simplifies hyperparameter tuning for researchers and practitioners working with online machine learning on data streams, though it is incremental as it builds on existing packages like river and spotRiver.
The paper tackles the challenge of hyperparameter tuning in online machine learning by introducing spotRiverGUI, a graphical user interface that automates the search for optimal hyperparameter settings, enabling users to efficiently compare and tune algorithms from the river package for streaming data.
Batch Machine Learning (BML) reaches its limits when dealing with very large amounts of streaming data. This is especially true for available memory, handling drift in data streams, and processing new, unknown data. Online Machine Learning (OML) is an alternative to BML that overcomes the limitations of BML. OML is able to process data in a sequential manner, which is especially useful for data streams. The `river` package is a Python OML-library, which provides a variety of online learning algorithms for classification, regression, clustering, anomaly detection, and more. The `spotRiver` package provides a framework for hyperparameter tuning of OML models. The `spotRiverGUI` is a graphical user interface for the `spotRiver` package. The `spotRiverGUI` releases the user from the burden of manually searching for the optimal hyperparameter setting. After the data is provided, users can compare different OML algorithms from the powerful `river` package in a convenient way and tune the selected algorithms very efficiently.