Plot4GMNS: An open-source academic research tool for visualizing multimodal networks for transportation system modeling and optimization

Authors: Dr. Junhua Chen, Zanyang Cui, Xiangyong Luo

Email: cjh@bjtu.edu.cn, zanyangcui@outlook.com, luoxiangyong01@gmail.com

Table of Contents

• Plot4GMNS: An open-source academic research tool for visualizing multimodal networks for transportation system modeling and optimization
  • Table of Contents
  • Introduction
  • Requirements
  • Install
  • Features
  • Usage
    • Quickstart
    • Advance Usage
  • Contributing
  • Contributors
  • Changelog
    • TODO LIST

Introduction

To enable rapid transportation modeling and optimization, as traffic management researchers, we provide this free open-source tool for visualizing multimodal networks. Based on GMNS data format by Zepha foundation, plot4gmns is designed for reading and plotting multimodal data sets including transportation network files, demand and agent trace files.

Requirements

• pandas
• shapely
• matplotlib
• numpy
• seaborn
• scipy
• chardet
• keplergl==0.3.2

Install

pip install plot4gmns

Note

• For Windows users, the pip method might fail to install some dependencies. If errors occur when you try to install any of those dependencies, try instead to pip install their .whl files, which can be downloaded from the Unoffical Windows Binaries for Python Extension Packages.

Features

• **show GMNS data elements (coordinates or geometry included)
• web-based network visualization
• show networks in different modes
• show network with given node types
• show network by given link types
• show network by given link attributes range
• show network by link attributes distribution
• show network with given POI types
• show network by poi attributes distribution
• show network demand matrix heatmap
• show network demand OD
• Show only network elements of interest
• Show different networks on one diagram
• Set the drawing style

Usage

Before starting, you must have prepared network files, including node.csv, link.csv, poi.csv, demand.csv, and zone.csv. The osm2gmns package will help you quickly obtain node, link, and poi data, and the grid2demand package will help you obtain network demand and zone information.

Quickstart

Step 0: generate multimodal network

import plot4gmns as p4g
mnet=p4g.generate_multi_network_from_csv(r'./datasets')

After executing the above command, you will get an Html file, as shown below. More visual operations are supported on the web site..

**Step 1: Show GMNS data elements (GMNS DATA Elements)

The supported data elements are: node, link, lane, zone, location, geometry, poi, movement, demand

# we can visualize GMNS elements and any combination of each data elements

# GMNS specification: https://github.com/zephyr-data-specs/GMNS/tree/main/spec
# For data elements including geometry: node, link, lane, zone, location, geometry, poi, movement, demand

cf_node = p4g.show_gmns_nodes(mnet)
cf_node.show()

cf_link = p4g.show_gmns_links(mnet)
cf_link.show()

# you can also show the combined network
cf_node_link = p4g.show_gmns_links(mnet, fig_obj=cf_node)
cf_node_link.show()

Step 2: show networks in different modes

# draw 'all' modes network and save to png file
cf = p4g.show_network_by_modes(mnet=mnet)
# show the figure on the current window
cf.show()

# show 'bike' mode network
cf = p4g.show_network_by_modes(mnet=mnet,modes=['bike'])
cf.show() # show the figure on the current window

Step 3: show network with given node types

cf = p4g.show_network_by_node_types(mnet=mnet,ctrl_type=['traffic_signals','crossing'])
cf.show()

Step 4: show network by given link types

# show network by given link types
cf = p4g.show_network_by_link_types(mnet=mnet,link_types=['secondary','footway'])
cf.show()

Step 5: show network by given link attributes range

# show network by given link length range
cf = p4g.show_network_by_link_length(mnet=mnet,min_length=10,max_length=50)
cf.show()

# show network by given link lane range
cf = p4g.show_network_by_link_lanes(mnet=mnet,min_lanes=1,max_lanes=3)
cf.show()

# show network by given link free speed range
cf = p4g.show_network_by_link_free_speed(mnet=mnet,min_free_speed=10,max_free_speed=40)
cf.show()

Step 6: show network by link attributes distribution

# show network by link lane distribution
cf = p4g.show_network_by_link_lane_distribution(mnet=mnet)
cf.show()

# show network by link capacity distribution
cf = p4g.show_network_by_link_capacity_distribution(mnet=mnet)
cf.show()

# show network by link free speed distribution
cf = p4g.show_network_by_link_free_speed_distribution(mnet=mnet)
cf.show()

Step 7: show network with given POI types

cf = p4g.show_network_by_poi_types(mnet=mnet,poi_type=['public','industrial'])
cf.show()

Step 8: show network by poi attributes distribution

# show network by poi attraction distribution
cf = p4g.show_network_by_poi_attraction_distribution(mnet=mnet)
cf.show()

# show network by poi production distribution
cf = p4g.show_network_by_poi_production_distribution(mnet=mnet)
cf.show()

Step 9: show network demand matrix heatmap

cf = p4g.show_network_demand_matrix_heatmap(mnet)
cf.show()

Step 10: show network demand OD

cf = p4g.show_network_by_demand_OD(mnet=mnet,load_network=True)
cf.show()

Advance Usage

Step 1: Show only network elements of interest

The tool displays all elements by default, such as nodes, links, and POIs. If you want to display only the elements of interest, you can refer to the following example to modify the state of other elements before invoking the drawing command.

# not show nodes
mnet.node_loaded = False
cf = p4g.show_network_by_link_lane_distribution(mnet=mnet)
cf.show()

Step 2: Show different networks on one diagram

By default, this tool will clear the original contents before drawing. If you want to draw different content on a graph, you can refer to the following example

mnet.node_loaded = False
mnet.POI_loaded = False
cf = p4g.show_network_by_link_lane_distribution(mnet=mnet)
mnet.link_loaded = False
mnet.POI_loaded = True
cf = p4g.show_network_by_poi_attraction_distribution(mnet,fig_obj=cf)
cf.show()

Step 3: Set the drawing style

Users can refer to the following examples to adjust the color, size and other attributes of the image before drawing.

parameter	Value	Description
mnet.style.figure_size	tuple,(width,height)	Image size
mnet.style.dpi	int	the resolution in dots per inch.
mnet.style.node_style.size	int	node marker size
mnet.style.node_style.colors	dict,{node_type:color}	node color in different types
mnet.style.node_style.markers	dict,{node_type:marker}	node marker in different types
mnet.style.link_style.linecolor	str	link color
mnet.style.link_style.linewidth	float	link width
mnet.style.poi_style.facecolor	str	POI facecolor
mnet.style.poi_style.edgecolor	str	POI edgecolor
mnet.style.demand_style.linecolor	str	demand flow line color
mnet.style.demand_style.linelinewidth	float	demand flow line width
mnet.style.zone_style.linewidth	float	zone grid line width
mnet.style.zone_style.edgecolors	str	zone grid edgecolors
mnet.style.zone_style.fontsize	int	zone label font size
mnet.style.zone_style.fontcolor	str	zone label font color

mnet.style.node_style.size = 3
mnet.style.link_style.linecolor = 'green'
mnet.style.poi_style.facecolor = 'gray'
cf = p4g.show_network_by_modes(mnet=mnet)
cf.show()

Contributing

Feel free to dive in! Open an issue.

Contributors

@PraiseC

@xyluo25

Changelog

2023-01-25 -- v0.1.1:

Support web-based network visualization(Kepler.gl)

TODO LIST

1. Add OD 3D visualization