Fixed position hierarchical output from NetworkX without graphviz? - python

I'm trying to use Python to plot simple hierarchical tree. I'm using the networkx module. I declared a simple graph
G=networkx.DiGraph()
After adding nodes and edges into G, I tried using
nx.draw(G)
or
nx.draw_networkx(G)
to plot. The output plot hierarchy are all correct, but the position of the nodes appears all random on the graph. Even worse, each time I ran the script, the node position are different.
There is a solution provided in a similar question which requires graphviz package.
pos=nx.graphviz_layout(G,prog='dot')
nx.draw(G,pos,with_labels=False,arrows=False)
Unfortunately I'm not allowed to install graphviz. So I'm seeking alternative solution here.
From graphviz solution, it looks like it's only needed to compute position of the node. Is it correct to say that, as long as I specify a list of coordinates to the draw command, I would be able to plot node at the correct location?
Thanks


UPDATE (15 Apr 2015) Look at my answer here for code that I think will do what you're after.
So networkx doesn't make it particularly easy to use the graphviz layout if you don't have graphviz because it's a better idea to use graphviz's algorithm if you're trying to reproduce graphviz's layout.
But, if you're willing to put in the effort to calculate the position for each node this is straightforward (as you guessed). Create a dict saying where each node's position should be.
pos = {}
for node in G.nodes():
pos[node] = (xcoord,ycoord)
nx.draw(G,pos)
will do it where xcoord and ycoord are the coordinates you want to have.
The various plotting commands are described here

Related

Convert networkx graph to dot format in python

I am converting my networkx graph using the following code.
nx.drawing.nx_pydot.write_dot(G,path)
It creates a correct dot format which I can visualize later using graphviz interface. However, instead of adding multiple existing lines(Arcs, edges whatever you say), it creates a single (or two if there is an edge in the opposite direction). I just want to have all lines to be preserved in the dot format. How can I do that?

The creation of Networkx graph in the first place was not correct. I changed the line G=nx.DiGraph(directed=True) to G=nx.MultiDiGraph(directed=True). Now I do not have that problem.

Adding sub nodes to networkx (with image)

I would like to create a networkx graph that looks more or less like this, but I haven't been able to find a way for it to display the way I need. The large nodes and edges display fine, but I haven't been able to find how to add the small nodes.

networkx.draw() has an optional argument node_size:
node_size (scalar or array, optional (default=300)) – Size of nodes. If an array is specified it must be the same length as nodelist.
If you want to draw nodes with various sizes, you should specify the array of sizes. You can also use some kind of list generator.
P.S. I don't recommend to use basic networkx drawing functional. There are many powerful visualization libraries better than networkx. Even in networkx docs you can find the same opinion. One can use Gephi, Graphviz (with various libraries) or Cytoscape for really HUGE graphs.

Is there a way to control line angle when producing graphs with pyDot

I wrote a python class to display and animate binary search trees. Yet the graphs produced by the pyDot edge and node commands don't seem to allow me to control the angle or direction of the arrows connecting each element of my tree.
There are lots of controls for shape of the nodes but there does not seem to be any control for angle of the lines. I was hoping to find some kind of minimum angle parameter to add to my edges as they are added to the graph?
This is a sample of my code where I draw nodes using the edge command:
def draw(parent_name, child_name):
# color for lines = red
edge = pydot.Edge(parent_name, child_name, color="#ff0000")
graph.add_edge(edge)
The end result is that my graphs for binary search trees don't look like the traditional tree structures when several nodes have only one child.
Here is a link to my project where you can see the otherwise working results:
http://www.embeddedcomponents.com/blogs/2013/12/visualizing-software-tree-structures/

from pydot documentation:
create will write the graph to a temporary dot file and process
it with the program given by 'prog' (which defaults to 'twopi'),
reading the Postscript output and returning it as a string is the
operation is successful.
it is not possible to directly control edges angels using the dot language, but adding invisible edges can give you the result you want, see an example here that generates the following graph:

Placing vertices of graphs + wxpython

I have to draw graphs along with edges according to a file input by the user. I am using wxPython for the same.
Once the positions are clear I can easily create circles and edges between the nodes but
I have a problem that given a panel to draw on is there any way I can get to know optimum positions for the vertices to be if I know the number of vertices ?
By optimum I mean simply that its readable what has been drawn and written along with it.....
So say that I have to draw 3 vertices I just want that I am able to clearly get the coordinates of where to place the nodes and if I can make the system automated....
Please help ....

You want a Graph Drawing algorithm. There is ongoing research in this area, but a simple force-directed algorithm can give good results for small graphs. Look at this wikipedia article for the algorithm. You can also get some open source libraries that handle this problem, like NodeBox and Graphvis.

Also a good lib: igraph
It offers a nice collection of layout algorithms

Arched Relationship Infographic In Python

This is a very specific inforgraphic challange altough the fundemental question is how do you build archs between words using matplotlib, cario or an other python libary.
Given a the following data structure.
me, you, 7 |
me, apple, 9 |
apple, you, 1 |
bike, me, 5
Names would be displayed horizontally the names with the most relationships larger than the others and then there would be weighted archs between the names. A 10 weighted arch would be twice as thick as a 5 weighted arch.
Inspiration comes from: similar diverstiy. http://similardiversity.net/
Let the challange commence!

matplotlib isn't the right library here, since it's not a general purpose graphics library. What you need here is either something like Cairo, or much simpler, you can do with the graphics capabilities of any GUI toolkit, such as PyQt. Another feasible approach is PyGame, which has good drawing capabilities as well.
If you want an example, see this Cairo samples page, the first sample - arc. You just write the text words and then use the arc code for an arc of any width and color between them.

There are several libraries, at least one of which relies on Matplotlib, that will do what you want. I recommend Networkx (www.networkx.lanl.gov) to build your graph structure, and which you can then use to call the relevant Matplotlib methods to plot. Networkx and Matplotlib work very well together.
import networkx as NX
import matplotlib.pyplot as PLT
Gh = NX.Graph()
Gh.add_edge("You", "Bike", weight=1.0)
Gh.add_edge("Bike", "Apple", weight=0.9)
Gh.add_edge("Me", "Bike", weight=1.1)
all_nodes = Gh.nodes()
# to scale node size with degree:
scaled_node_size = lambda(node) : NX.degree(Gh, node) * 700
position = NX.spring_layout(Gh) # just choose a layout scheme
NX.draw_networkx_nodes(Gh, position, node_size=map(scaled_node_size, all_nodes))
NX.draw_network_edges(Gh, position, Gh.edges(), width=1.0, alpha=1.0, edge_color="red")
# now for the Matplotlib part:
PLT.axis("off")
PLT.show()
As you can see, you could scale the edges by applying a factor to vary the 'weight' parameter to any of the 'edge' methods, just the same way as i did it for node scaling.
I would also recommend pygraphviz (obviously using graphviz as its backend). It is very similar to Netwworkx (same lead developer).

I don't really see where the difficulty or challenge is. All you need is a graphics library that can draw text and half circles, which is possible in pretty much everything. There is no need for Bezier curves. For example you could simply create an SVG file (then you don't even need any library and can nicely embed this in a webpage and use some JavaScript to make it interactive).
The actual plotting is simple: the center of a circle is in the middle between the two words, the radius is half the distance between them. You can then adjust the stroke to reflect other quantities.

Categories

Resources