Training course for spatial analysis, data science and machine learning at CIMMYT
CIMMYT staff participate in a R training course to develop capabilities for spatial analysis, automatizing image processing, machine learning, and other approaches for working on large climate data sets.
As part of CIMMYT’s move towards big data and data science, Robert Hijmans, Associate Professor at the Department of Environmental Science and Policy at the University of California, Davis, gave a two-day (17 and 18 December) R training course dedicated to spatial analysis, data science and machine learning. The training course—organized and funded by the Geospatial Data Community of Practice of the CGIAR Platform for Big Data in Agriculture— was attended by 20 staff members from SIP, SEP, IDP, WHEAT and GRP at CIMMYT HQ.
R is an open source programing language and environment for statistical computing and graphics used by researchers and students all over the world. CIMMYT uses R for biometrics and statistics for GxE analysis and trainings. The programming language also has powerful capabilities for spatial analysis, automatizing image processing, machine learning and other modeling approaches as well as for working on large climate and other data sets used in CIMMYT’s research projects.
Robert Hijmans has developed several of these tools within R. He is highly respected in the GIS world for several global data products like Worldclim — a monthly climate data set with over 16.000 citations since its release in 2005 — and the Database of Global Administrative Areas (GADM).
The training course’s first day was designed to give the participants an overview of tools, routines and procedures within R. On the second day, participants applied tools to specific data sets and research questions from their current projects. This covered areas like high throughput phenotyping in wheat, land use change detection in Mexico and Bolivia, climate data processing for analysis, and targeting and crop modeling in the foresight group.
Robert Hijmans is currently collaborating with the TAMASA project in Africa. Organizers expect that further collaboration and data analysis will result as an additional benefit of the training.