The Helmholtz Analytics Framework (HAF) is a joint project of the six Helmholtz centers DLR, KIT, FZ Jülich, DESY, DKFZ and HMGU. The main objective of the Helmholtz Analytics Framework is to develop a common tool and methods for data analysis. The latter shall be employed in a variety of applications that can be found within the Helmholtz association, ranging from the design of an aircraft, the prediction of solar radiation on photovoltaic, to image analysis in neuroscience and data fusion in structural biology.
In particular, machine learning methods are of interest, in order to extract essential information from large amounts of data. Furthermore, to be able to cope with extremely large amounts of data at all, an efficient data analysis on supercomputers and graphics cards, as performed by the department for High-Performance Computing, is essential. More specifically, we are concerned with the parallelized training of the algorithms, as well as the easy provision of distributed data.