transbigdata

English 中文版

TransBigData

Introduction

TransBigData is a Python package developed for transportation spatio-temporal big data processing, analysis and visualization. TransBigData provides fast and concise methods for processing common transportation spatio-temporal big data such as Taxi GPS data, bicycle sharing data and bus GPS data. TransBigData provides a variety of processing methods for each stage of transportation spatio-temporal big data analysis. The code with TransBigData is clean, efficient, flexible, and easy to use, allowing complex data tasks to be achieved with concise code.

For some specific types of data, TransBigData also provides targeted tools for specific needs, such as extraction of Origin and Destination(OD) of taxi trips from taxi GPS data and identification of arrival and departure information from bus GPS data. The latest stable release of the software can be installed via pip and full documentation can be found at https://transbigdata.readthedocs.io/en/latest/. Introduction PPT can be found here and here(in Chinese)

Target Audience

The target audience of TransBigData includes:

• Data science researchers and data engineers in the field of transportation big data, smart transportation systems, and urban computing, particularly those who want to integrate innovative algorithms into intelligent trasnportation systems
• Government, enterprises, or other entities who expect efficient and reliable management decision support through transportation spatio-temporal data analysis.

Technical Features

• Provide a variety of processing methods for each stage of transportation spatio-temporal big data analysis.
• The code with TransBigData is clean, efficient, flexible, and easy to use, allowing complex data tasks to be achieved with concise code.

Main Functions

Currently, TransBigData mainly provides the following methods:

• Data Quality: Provides methods to quickly obtain the general information of the dataset, including the data amount the time period and the sampling interval.
• Data Preprocess: Provides methods to clean multiple types of data error.
• Data Gridding: Provides methods to generate multiple types of geographic grids (Rectangular grids, Hexagonal grids) in the research area. Provides fast algorithms to map GPS data to the generated grids.
• Data Aggregating: Provides methods to aggregate GPS data and OD data into geographic polygon.
• Data Visualization: Built-in visualization capabilities leverage the visualization package keplergl to interactively visualize data on Jupyter notebook with simple code.
• Trajectory Processing: Provides methods to process trajectory data, including generating trajectory linestring from GPS points, and trajectory densification, etc.
• Basemap Loading: Provides methods to display Mapbox basemap on matplotlib figures

Grid processing framework offered by TransBigData

Here is an overview of the gridding framework offered by TransBigData.

See This Example for further details.

Trajectory processing framework offered by TransBigData

Here is an overview of the Trajectory processing framework offered by TransBigData.

See This Example for further details.

Installation

TransBigData support Python >= 3.6

Using pypi

TransBigData can be installed by using pip install. Before installing TransBigData, make sure that you have installed the available geopandas package. If you already have geopandas installed, run the following code directly from the command prompt to install TransBigData:

pip install transbigdata

Using conda-forge

You can also install TransBigData by conda-forge, this will automaticaly solve the dependency, it can be installed with:

conda install -c conda-forge transbigdata

Contributing to TransBigData

All contributions, bug reports, bug fixes, documentation improvements, enhancements and ideas are welcome. A detailed overview on how to contribute can be found in the contributing guide on GitHub.

Examples

Example of data visualization

Visualize trajectories (with keplergl)

Visualize data distribution (with keplergl)

Visualize OD (with keplergl)

Example of taxi GPS data processing

The following example shows how to use the TransBigData to perform data gridding, data aggregating and data visualization for taxi GPS data.

Read the data

import transbigdata as tbd
import pandas as pd
#Read taxi gps data  
data = pd.read_csv('TaxiData-Sample.csv',header = None) 
data.columns = ['VehicleNum','time','lon','lat','OpenStatus','Speed'] 
data

	VehicleNum	time	lon	lat	OpenStatus	Speed
0	34745	20:27:43	113.806847	22.623249	1	27
1	34745	20:24:07	113.809898	22.627399	0	0
2	34745	20:24:27	113.809898	22.627399	0	0
3	34745	20:22:07	113.811348	22.628067	0	0
4	34745	20:10:06	113.819885	22.647800	0	54
...	...	...	...	...	...	...
544994	28265	21:35:13	114.321503	22.709499	0	18
544995	28265	09:08:02	114.322701	22.681700	0	0
544996	28265	09:14:31	114.336700	22.690100	0	0
544997	28265	21:19:12	114.352600	22.728399	0	0
544998	28265	19:08:06	114.137703	22.621700	0	0

544999 rows × 6 columns

Data pre-processing

Define the study area and use the tbd.clean_outofbounds method to delete the data out of the study area

#Define the study area
bounds = [113.75, 22.4, 114.62, 22.86]
#Delete the data out of the study area
data = tbd.clean_outofbounds(data,bounds = bounds,col = ['lon','lat'])

Data gridding

The most basic way to express the data distribution is in the form of geograpic grids. TransBigData provides methods to generate multiple types of geographic grids (Rectangular grids, Hexagonal grids) in the research area. For rectangular gridding, you need to determine the gridding parameters at first (which can be interpreted as defining a grid coordinate system):

#Obtain the gridding parameters
params = tbd.area_to_params(bounds,accuracy = 1000)
params

{'slon': 113.75, 'slat': 22.4, 'deltalon': 0.00974336289289822, 'deltalat': 0.008993210412845813, 'theta': 0, 'method': 'rect', 'gridsize': 1000}

The gridding parameters store the information of the initial position, the size and the angle of the gridding system.

The next step is to map the GPS data to their corresponding grids. Using the tbd.GPS_to_grid, it will generate the LONCOL column and the LATCOL column (Rectangular grids). The two columns together can specify a grid:

#Map the GPS data to grids
data['LONCOL'],data['LATCOL'] = tbd.GPS_to_grid(data['lon'],data['lat'],params)

Count the amount of data in each grids, generate the geometry of the grids and transform it into a GeoDataFrame:

#Aggregate data into grids
grid_agg = data.groupby(['LONCOL','LATCOL'])['VehicleNum'].count().reset_index()
#Generate grid geometry
grid_agg['geometry'] = tbd.grid_to_polygon([grid_agg['LONCOL'],grid_agg['LATCOL']],params)
#Change the type into GeoDataFrame
import geopandas as gpd
grid_agg = gpd.GeoDataFrame(grid_agg)
#Plot the grids
grid_agg.plot(column = 'VehicleNum',cmap = 'autumn_r')

Triangle and Hexagon grids & rotation angle

TransBigData also support the triangle and hexagon grids. It also supports given rotation angle for the grids. We can alter the gridding parameter:

#set to the hexagon grids
params['method'] = 'hexa'
#or set as triangle grids: params['method'] = 'tri'
#set a rotation angle (degree)
params['theta'] = 5

Then we can do the GPS data matching again:

#Triangle and Hexagon grids requires three columns to store ID
data['loncol_1'],data['loncol_2'],data['loncol_3'] = tbd.GPS_to_grid(data['lon'],data['lat'],params)
#Aggregate data into grids
grid_agg = data.groupby(['loncol_1','loncol_2','loncol_3'])['VehicleNum'].count().reset_index()
#Generate grid geometry
grid_agg['geometry'] = tbd.grid_to_polygon([grid_agg['loncol_1'],grid_agg['loncol_2'],grid_agg['loncol_3']],params)
#Change the type into GeoDataFrame
import geopandas as gpd
grid_agg = gpd.GeoDataFrame(grid_agg)
#Plot the grids
grid_agg.plot(column = 'VehicleNum',cmap = 'autumn_r')

Data Visualization(with basemap)

For a geographical data visualization figure, we still have to add the basemap, the colorbar, the compass and the scale. Use tbd.plot_map to load the basemap and tbd.plotscale to add compass and scale in matplotlib figure:

import matplotlib.pyplot as plt
fig =plt.figure(1,(8,8),dpi=300)
ax =plt.subplot(111)
plt.sca(ax)
#Load basemap
tbd.plot_map(plt,bounds,zoom = 11,style = 4)
#Define colorbar
cax = plt.axes([0.05, 0.33, 0.02, 0.3])
plt.title('Data count')
plt.sca(ax)
#Plot the data
grid_agg.plot(column = 'VehicleNum',cmap = 'autumn_r',ax = ax,cax = cax,legend = True)
#Add scale
tbd.plotscale(ax,bounds = bounds,textsize = 10,compasssize = 1,accuracy = 2000,rect = [0.06,0.03],zorder = 10)
plt.axis('off')
plt.xlim(bounds[0],bounds[2])
plt.ylim(bounds[1],bounds[3])
plt.show()

Citation information

Please cite this when using TransBigData in your research. Citation information is as follows:

@article{Yu2022,
  doi       = {10.21105/joss.04021},
  url       = {https://doi.org/10.21105/joss.04021},
  year      = {2022},
  publisher = {The Open Journal},
  volume    = {7},
  number    = {71},
  pages     = {4021},
  author    = {Qing Yu and Jian Yuan},
  title     = {TransBigData: A Python package for transportation spatio-temporal big data processing, analysis and visualization},
  journal   = {Journal of Open Source Software}
}

Introducing Video (In Chinese)

• Bilibili
• Youtube