GTRI
LightGraphs offers both (a) a set of simple, concrete graph implementations – Graph (for undirected graphs) and DiGraph (for directed graphs), and (b) an API for the development of more sophisticated graph implementations under the AbstractGraph type. The data structures and algorithms are now decoupled in order to allow users to extend the basic graph implementations with novel data structures which include support for weights, properties, and more efficient storage. We have integrated with EvolvingGraphs.jl, MetaGraphs.jl, and SimpleWeightedGraphs.jl. These packages allow you to use graphs with various forms of metadata attached to the nodes or edges. From traditional weighted graphs, to dynamic graphs with changing edge sets, to MetaGraphs.jl which allows you to ingest a DataFrame as a property graph. MetaGraphs provides the foundation for an in memory GraphDB. Demo! We will live-code a simple graph type defined by its adjacency matrix and leverage LightGraphs for different applications. This talk will show how anyone can extend JuliaGraphs with a custom graph implementation for their application.
Mathieu Besançon is a PhD student in a double program between Ecole Polytechnique of Montréal, at the GERAD lab and the INOCS team at INRIA Lille, in mathematical optimization and game theory for smart grids. James Fairbanks earned a Ph.D in Computational Science and Engineering at Georgia Tech. My research focuses on numerical, statistical, and streaming algorithms for data analysis. The applications include complex networks, online media, medical data, and sensor data.
