PlasmoDataPlots

PlasmoDataPlots.jl is a package for Julia designed for visualizing graphs from the PlasmoData.jl package.

Bug Reports and Support

This package is functional and can be installed as is. It is still under development, and significant changes will continue to come. If you encounter any issues or bugs, please submit them through the Github issue tracker.

Installation

To install this package, you can use

using Pkg
Pkg.add(url="https://github.com/zavalab/PlasmoDataPlots.jl")

or

pkg> add https://github.com/zavalab/PlasmoDataPlots.jl

Overview

PlasmoDataPlots.jl is designed for plotting the DataGraph object. Node positions are saved under the NodeData of the DataGraph object. For these positions to be recognized, they must be called "x_positions" and "y_positions". If the user does not define these, the node positions are determined by NetworkLayout.jl's sfdp function (stands for Scalable Force Directed Placement). Alternatively, when the DataGraph was constructed from a matrix or a tensor, the user can set the node positions using set_matrix_node_positions! or set_tensor_node_positions! functions which take the argument of the DataGraph and the original matrix or tensor.

The primary function in PlasmoDataPlots.jl is plot_graph, which takes the following, optional keyword arguments:

• get_new_positions::Bool: If true, calculates new positions using NetworkLayout.jl's sfdp function.
• plot_edges::Bool: If false, only the nodes of the DataGraph will be plotted. The edges take the longest time to plot, so this can reduce the plotting time for graphs with many nodes where the edges are less important to visualize (e.g., large matrices formed as graphs)
• C, K, iterations, tol: These are arguments used by NetworkLayout.jl's sfdp function for determining node positions.
• xdim, ydim: Determing the dimensions of the final plot (default is 800 for both)
• linewidth: Determines the width of the edges
• linealpha: Determines the alpha value (how transparent the edge is).
• linecolor: The color used for the edges. When line_z is defined, this should be a color gradient.
• line_z: The set of data corresponding to edge weights. This should be a vector of length equal to the number of edges. Can be used if the edges should be color coded by weight
• nodecolor: Color of the node markers. When node_z is defined, this should be a color gradient.
• nodesize: size of the node markers
• nodestrokewidth: Width of the border on the node markers
• nodestrokecolor: Color of the border on the node markers
• nodestrokealpha: alpha (transparency) of the border on the node markers
• node_z: Set of data corresponding to node weights. This should be a vector of length equal to the number of nodes. Can be used if the nodes should be color coded by weight
• framestyle: framestyle of plot
• rev: Boolean indicating, if node_z or line_z is defined, that the color gradient from node or line color will be reversed
• legend: Boolean determining whether there should be a colorbar when node_z is defined
• save_fig: Boolean to determine if the figure should be saved or not
• fig_name: Name of the file name that the saved figure should have.

Example

Any DataGraph can be plotted simple by calling

plot_graph(datagraph)

However, it may be useful to use keyword arguments depending on the visualization you desire. For example:

random_matrix = rand(15, 15)

matrix_graph = PlasmoData.matrix_to_graph(random_matrix)

set_matrix_node_positions!(matrix_graph, random_matrix)

plot_graph(
  matrix_graph, 
  nodesize = 8, 
  linecolor = :grey, 
  linewidth = 4, 
  node_z = get_node_data(matrix_graph).data[:],
  nodecolor = :algae
)