Adding a Marker To Ndlib Graph Viz

[amshoo]

Ndlib_Marker_Viz

Adding a Marker To Ndlib Graph Viz

Adding marker to Ndlib ComparisonViz and DiffusionViz to distinguish multiple lines and improve visual understanding.

NDlib is a Python software package that allows to describe, simulate, and study diffusion processes on complex networks. The package was design and documentation by Giulia Rossetti you can find examples, tutorials and a complete reference here ndlib.

Matplotlib Marker: a basic matplotlib Marker plot, this shows a few optional features, like defining legend labels, legend size, and maker size.

#Code to plot a simple demo with a “*” marker.
 
import matplotlib.pyplot as plt
x = [1,2,3,4,5]
fig, ax = plt.subplots()
ax.plot(x, '*' , markersize=12, label='(*)')
ax.axis('equal')
leg = ax.legend(fontsize=12);

Below are some basic example plots for markers (‘*’, ‘o’ , ’v’ , ‘+’) respectively.

Assign a unique marker for each plot

Form Solution 4in StackOverflow, which answer the question of how to us unique markers for each plot in matplotlib. the solution uses ‘itertools.cycle’ to iterate over a list or tuple indefinitely which picks markers randomly for you. The answer below.

Python 2.x

import itertools
marker = itertools.cycle((',', '+', '.', 'o', '*')) for n in y:
    plt.plot(x,n, marker = marker.next(), linestyle='')

Python 3.x

import itertools
marker = itertools.cycle((',', '+', '.', 'o', '*')) for n in y:
plt.plot(x,n, marker = next(marker), linestyle='')

To find the set of All possible markers you can visit matplotlib.markers[2] in matplotlib website (https://matplotlib.org/3.1.0/api/markers_api.html).

Ndlib Graph Viz

From the documentation of Ndlib, it shown that the model uses matplotlib and bakeh to visualize the diffusion/spreading process and did a good job but Adding markers to a line can be a useful way to distinguish multiple lines or to highlight particular data points. So above solution to add unique markers to all lines.
Instead of visualizing the diffusion process like this:

It can be visualize like this:

[coveralls]

Coverage decreased (-0.05%) to 94.578% when pulling 1c843ba on amshoo:master into 1927b58 on GiulioRossetti:master.

[Bisma1]

I agree, this is very helpful in distinguishing the plots but I need to specifically replace legend names to depict the name of dataset or method. Is there any way out to alter those as I am naïve to viz plots . thankyou