On Wed, Aug 6, 2014 at 11:37 AM, AlexanderRiggers < [email protected]> wrote:
> To perform the page rank I have to create a graph object, adding the edges > by setting sourceID=id and distID=brand. In GraphLab there is function: g = > SGraph().add_edges(data, src_field='id', dst_field='brand') > > Is there something similar in GraphX? It sounds like you're trying to parse an edge list file into a graph, where each line is a comma-separated pair of numeric vertex ids. There's a built-in parser for tab-separated pairs (see GraphLoader) and it should be easy to adapt that to comma-separated pairs. You can also drop the header line using RDD#filter (and eventually using https://github.com/apache/spark/pull/1839). Ankur <http://www.ankurdave.com/>
