NetworkX Integration with Relational Databases

404 views
Skip to first unread message

Konstantinos Xirogiannopoulos

unread,
Jan 21, 2016, 10:48:42 AM1/21/16
to networkx-discuss
Hi everyone, 

We'd like to introduce and make an initial beta release of this system we've been developing at the University of Maryland called GraphGen, which is geared towards efficiently extracting graphs from relational databases (currently only PostgreSQL). With GraphGen, users can use a declarative query language we've designed in order to define graphs that exist inside their relational datasets and load them into memory. Although GraphGen is natively written in Java, we've also written a wrapper python library called graphgenpy which calls our tool and simply returns a serialized (.gml) version of the specified graph, which can then automatically be parsed into NetworkX and then analyzed! 

To download the initial release graphgen and for more information, demos, example code you can run, and tutorials please visit http://konstantinosx.github.io/graphgen-project/. I'd love to know of any  datasets anyone finds use for graphgen on. Also, please feel free to contact me with any questions, suggestions, feedback as well as to submit any bug reports as issues.

Thank you,
Konstantinos Xirogiannopoulos

suma...@gmail.com

unread,
Feb 10, 2016, 7:26:14 AM2/10/16
to networkx-discuss
is any function or method for computing number of positive(negative) edges in networkX?

Himanshu Mishra

unread,
Feb 10, 2016, 7:33:07 AM2/10/16
to networkx-discuss
@suman1791 By positive edges, do you mean edges with positive weights?
If yes then, I think we can iterate over them and count it ourselves. Would that help?

On Wed, Feb 10, 2016 at 5:56 PM <suma...@gmail.com> wrote:
is any function or method for computing number of positive(negative) edges in networkX?

Himanshu Mishra

Daniel Schult

unread,
Feb 10, 2016, 7:34:08 AM2/10/16
to networkx...@googlegroups.com
len(1 for u,v,d in G.edges(data=True) if d['weight']>0) 

Suman Pandey

unread,
Feb 10, 2016, 10:24:52 AM2/10/16
to networkx...@googlegroups.com
thank you sir but network dataset weights are not included then what i do for this
for example the network dataset is contains only edgelist 

--
You received this message because you are subscribed to a topic in the Google Groups "networkx-discuss" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/networkx-discuss/Yxjm6Dl6lHc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to networkx-discu...@googlegroups.com.
To post to this group, send email to networkx...@googlegroups.com.
Visit this group at https://groups.google.com/group/networkx-discuss.
For more options, visit https://groups.google.com/d/optout.

CA-GrQc.txt

Daniel Schult

unread,
Feb 10, 2016, 3:31:43 PM2/10/16
to networkx...@googlegroups.com
If you don't have edge weights, then all edge weights are 1. So all edges are positive. Or do you mean something else when you say "positive" edge?

--
You received this message because you are subscribed to the Google Groups "networkx-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to networkx-discu...@googlegroups.com.

Suman Pandey

unread,
Feb 10, 2016, 3:41:38 PM2/10/16
to networkx...@googlegroups.com
thank you so much Daniel ,really helped me

Suman Pandey

unread,
Feb 10, 2016, 4:07:30 PM2/10/16
to networkx...@googlegroups.com, Himanshu Mishra
my code is in this file it gives error
error is below
KeyError: 'weight'
why it gives error
degcorelation.py
Reply all
Reply to author
Forward
0 new messages