A Fortran-Keras Deep Learning Bridge for Scientific Computing

• Published in: Scientific programming Vol. 2020; no. 2020; pp. 1 - 13
• Main Authors: Curcic, Milan, Linstead, Erik, Best, Natalie, Pritchard, Mike, Ott, Jordan, Baldi, Pierre
• Format: Journal Article
• Language: English
• Published: Cairo, Egypt Hindawi Publishing Corporation 2020; Hindawi; John Wiley & Sons, Inc
• Subjects: Application programming interface; Artificial neural networks; Chemistry; Climate models; Computational fluid dynamics; Computer architecture; Computer simulation; Deep learning; Dosimetry; Earthquakes; Emulators; Error reduction; Fluid dynamics; FORTRAN; High level languages; Laboratories; Libraries; Machine learning; Mechanics; Molecular physics; Neural networks; Partial differential equations; Physics; Popularity; Programming languages; Simulation; Software; Training; Weather forecasting; United States > US

Implementing artificial neural networks is commonly achieved via high-level programming languages such as Python and easy-to-use deep learning libraries such as Keras. These software libraries come preloaded with a variety of network architectures, provide autodifferentiation, and support GPUs for f...