Published components
ML Toolbox
About the ML Toolbox
The ML Toolbox supports the development of ML software by providing tools for common tasks in the development process of an Machine Learning (ML) application. In the future, it will provide tools for diverse application areas of ML – e.g. Computer Vision (CV), Natural Language Processing (NLP), Time Series Prediction – as well as for the different phases of an ML project (data generation, pre-processing, modeling, post-processing).
In the current version of the ML Toolbox, the focus is on computer vision applications with the main component MLCVZoo. The MLCVZoo enables the integration of ML models from different frameworks via a unified API and thus saves the effort of integrating the frameworks individually for each new computer vision project. Thus, this component simplifies the challenge of finding the best model for the task under consideration.
The components config-builder and related-mltoolbox provide additional support for integrating configurations for training ML models.
Involved People
Maximilian Otten, maximilian.otten@iml.fraunhofer.de; Anike Murrenhoff, anike.murrenhoff@iml.fraunhofer.de