A second example

In this page, we show how one can import data in Karstnet and compute the statistical properties of the karstic network.

Loading a network

First, let us import the karstnet module:

import karstnet as kn

There are two built in functions to import data from ASCII files.

The First one assumes that the data describing the geometry of the network are stored in two ASCII files: one containing the position of the nodes, the other containing the description of the links between the nodes (the edges).

The convention used by Karstnet is that the two files have the same basename, and have different extensions:

• basename_nodes.dat: the matrix of 3D node coordinates, one line per node, one column per coordinates

• basename_links.dat: the list of links (edges), one edge per line, two nodes ID for each link

To load the karstic network geometry from these files, use this command:

myobject = kn.from_nodlink_dat('basename')

After reading the two data files, karstnet creates the internal structure of the graph and precomputed all the information that it requires, like for example the distance between the nodes, the orientation of the edges, and stored everything in the object.

In particular, the simplified graph is created during the import. The simplified graph is used to quantify the topological characteristics of the network.

A second possibility is to import the network from a GOCAD ASCII Pline file:

myobject = kn.from_pline('filename.pl')

More information in the API documentation base.from_pline()

Characterizing the network

To check the import and the network simplification, it is possible to plot the map of the original and simplified networks side by side in 2D:

myobject.plot()

Other functions are available to plot the network in 3D:

myobject.plot3()

Finally, to compute the statistical characteristics of this network, just call the characterize function:

results = myobject.characterize_graph( verbose = True )