PI Name: Romit Maulik, LCF

Description: We are building deep learning emulators for the NASA DayMET dataset (https://daac.ornl.gov/cgi-bin/dsviewer.pl?ds_id=1328) for forecasting maximum daytime temperatures across North America. These emulators will enable fast ensemble predictions at very fine spatial resolutions (a grid-size of 10km^2).

Testbed: Nvidia DGX and gpu_v100_smx2 queues