Abstract
In this paper we propose a distributed architecture to provide machine learning practitioners with a set of tools and cloud services that cover the whole machine learning development cycle: ranging from the models creation, training, validation and testing to the models serving as a service, sharing and publication. In such respect, the DEEP-Hybrid-DataCloud framework allows transparent access to existing e-Infrastructures, effectively exploiting distributed resources for the most compute-intensive tasks coming from the machine learning development cycle. Moreover, it provides scientists with a set of Cloud-oriented services to make their models publicly available, by adopting a serverless architecture and a DevOps approach, allowing an easy share, publish and deploy of the developed models.
Highlights
The impact of emerging computing techniques together with an increasing dimension of large datasets and the availability of more and more performing and accessible computing resources is transforming many research areas
Whilst many of the described services focus their activity on a subset of the phases of the machine learning cycle (Figure 1), our work proposes a comprehensive framework that comprises a set of services and tools aimed at covering all the relevant phases of the development and deployment of a machine learning application: from the model creation to the serving and sharing phases, passing trough the training, testing and evaluation phases
The DEEP as a Service4 solution (DEEPaaS) API is a REST endpoint, based on the OpenAPI specification [58] that provides a thin layer over a machine learning model, ensuring consistency across all the modules that are published into the marketplace
Summary
ÁLVARO LÓPEZ GARCÍA 1, JESÚS MARCO DE LUCAS 1, MARICA ANTONACCI 3, WOLFGANG ZU CASTELL 9,10, MARIO DAVID 2, MARCUS HARDT 6, LARA LLORET IGLESIAS 1, GERMÁN MOLTÓ 7, MARCIN PLOCIENNIK 5, VIET TRAN 8, ANDY S. ALIC 7, MIGUEL CABALLER 7, ISABEL CAMPOS PLASENCIA 1, ALESSANDRO COSTANTINI 4, STEFAN DLUGOLINSKY 8, DOINA CRISTINA DUMA 4, GIACINTO DONVITO 3, JORGE GOMES 2, IGNACIO HEREDIA CACHA 1, KEIICHI ITO 9, VALENTIN Y. KOZLOV 6, GIANG NGUYEN 8, PABLO ORVIZ FERNÁNDEZ 1, ZDENĚK ŠUstr 11, AND PAWEL WOLNIEWICZ 5.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
