Benchmark Datasets

Benchmark datasets serve as summary descriptions of problem areas, providing a simple interface between disciplines without requiring extensive background knowledge.  Each benchmark consists of a data set, along with a document that provides a quick introduction to the topic, provides a wishlist of analysis tasks and a quick interface (e.g. in Matlab or Python) to load and visualize the data.

FOR A GENERAL INTRODUCTION to the benchmark creation effort (short 4-page paper):

  • I. Ebert-Uphoff, D.R. Thompson, I. Demir, Y.R. Gel, M.C. Hill, A. Karpatne, M. Guereque, V. Kumar, E. Cabral-Cano, P. Smyth, A Vision for the Development of Benchmarks to Bridge Geoscience and Data Science, Proceedings of the Seventh International Workshop on Climate Informatics (CI 2017), NCAR Technical Note NCAR/TN-536+PROC, Sept 2017.  Get PDF here

Interested in shaping or contributing a benchmark?  Join the Geoscience Case Studies working group!

Benchmark Catalog

…Table in prep…