Performance Benchmarking of Parallel Hyperparameter Tuning for Deep Learning based Tornado Predictions

Simple item page
dc.contributor.authorBasalyga, Jonathan N.
dc.contributor.authorBarajas, Carlos A.
dc.contributor.authorGobbert, Matthias K.
dc.contributor.authorWang, Jianwu
dc.date.accessioned2020-07-27T18:25:06Z
dc.date.available2020-07-27T18:25:06Z
dc.date.issued2020-05-31
dc.description.abstractPredicting violent storms and dangerous weather conditions with current models can take a long time due to the immense complexity associated with weather simulation. Machine learning has the potential to classify tornadic weather patterns much more rapidly, thus allowing for more timely alerts to the public. To deal with class imbalance challenges in machine learning, different data augmentation approaches have been proposed. In this work, we examine the wall time difference between live data augmentation methods versus the use of preaugmented data when they are used in a convolutional neural network based training for tornado prediction. We also compare CPU and GPU based training over varying sizes of augmented data sets. Additionally we examine what impact varying the number of GPUs used for training will produce given a convolutional neural network on wall time and accuracy. We conclude that using multiple GPUs to train a single network has no significant advantage over using a single GPU. The number of GPUs used during training should be kept as small as possible for maximum search throughput as the native Keras multi-GPU model provides little speedup with optimal learning parameters.en_US
dc.description.sponsorshipThis work is supported by the grant “CyberTraining: DSE: Cross-Training of Researchers in Computing, Applied Mathematics and Atmospheric Sciences using Advanced Cyberinfrastructure Resources” from the National Science Foundation (grant no. OAC–1730250). Co-author Carlos Barajas additionally acknowledges support as HPCF RA. The hardware used in the computational studies is part of the UMBC High Performance Computing Facility (HPCF). The facility is supported by the U.S. National Science Foundation through the MRI program (grant nos. CNS–0821258, CNS–1228778, and OAC–1726023) and the SCREMS program (grant no. DMS–0821311), with additional substantial support from the University of Maryland, Baltimore County (UMBC).en_US
dc.description.urihttps://www.sciencedirect.com/science/article/pii/S2214579621000290en_US
dc.format.extent35 pagesen_US
dc.genrejournal articlesen_US
dc.genrePreprints
dc.identifierdoi:10.13016/m2mm94-pnfu
dc.identifier.citationJonathan N. Basalyga, Carlos A. Barajas, Matthias K. Gobbert, Jianwu Wang. Performance Benchmarking of Parallel Hyperparameter Tuning for Deep Learning based Tornado Predictions. Big Data Research, vol. 25, no. 100212, 2021. DOI:10.1016/j.bdr.2021.100212en_US
dc.identifier.urihttp://hdl.handle.net/11603/19251
dc.identifier.urihttps://doi.org/10.1016/j.bdr.2021.100212
dc.language.isoen_USen_US
dc.publisherElsevieren_US
dc.relation.isAvailableAtThe University of Maryland, Baltimore County (UMBC)
dc.relation.ispartofUMBC Mathematics Department Collection
dc.relation.ispartofUMBC Faculty Collection
dc.relation.ispartofUMBC Student Collection
dc.relation.ispartofUMBC Information Systems Department
dc.rightsThis item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.
dc.subjectUMBC High Performance Computing Facility (HPCF)
dc.titlePerformance Benchmarking of Parallel Hyperparameter Tuning for Deep Learning based Tornado Predictionsen_US
dc.typeTexten_US
dcterms.creatorhttps://orcid.org/0000-0003-1745-2292
dcterms.creatorhttps://orcid.org/0000-0002-9933-1170

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2021-Performance Benchmarking of Data Augmentation and GPU Count Variability for Deep Learning Tornado Predictions.pdf
Size:
799.75 KB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
2.56 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

UMBC Mathematics and Statistics Department
UMBC Faculty Collection
UMBC Information Systems Department
UMBC Student Collection