Turing Geovisualization Engine beta release

LIDA Seminars 2021

L Hama

R Beecham

N Lomax

2021-05-20

TGVE

Project Overview

  • Processing and visualizing spatio-temporal data using open source tech
  • Apply empirically-informed guidelines around visualization design and techniques from Geographic Information Science. Paper (Beecham et al. 2021)
  • Production grade delivery + packaging (Dockerized)

Talk Overview

A damaged Graphics Processing Unit (GPU) Baked GPU

  • Brief outline of the project
  • Brief GitHub pages
  • Walk through (github pages option)

What is an eAtlas?

What is an eatlas?

  • Feature extraction? Beautiful pictures?
  • Data exploration/interrogation to draw insights?
  • Combining different datasets?
  • Decision making and public engagement tool?

TGVE (1)

TGVE

TGVE (2)

TGVE

  • A web application powered by R + React
  • R package geoplumber (not on CRAN and welcomes contributions) can run it
  • Dockerfile to deploy both

Showcases

heatmap

  • Two different projects from UoL both national scale datasets
  • Ability to go from data to vector tiles (?) to data driven map tiles
  • Processing data at scale in realtime using R.data.table on typical servers.

SPENSER

www.geopspenser.com

  • Data Processing & Development pipeline
  • Production deployment (Docker container)

SPENSER

SaferActive

https://map.saferactive.org

  • Various types of data including vector tiles (data driven styling)
  • Production deployment (Docker)
  • Processing national data and visualizing it at scale

SPENSER

COVID19 application

covid19 covid19

Heavily customised code base of the project source code. (not deployed)

TGVE (re)design

  • Modular architecture, examples like CMS (Wordpress), ERP (Odoo) etc.
  • Modules like data wrangling, visualization and AI? etc…
  • Scalable tech and tapping into both R + React ecosystems

Modular design

One area of modular design is how eAtlas uses with Uber’s DeckGL visualization layers:

  • Managing layers (including custom)
  • Using layers

TGVE JSON to widgets

TGVE widgets

GitHub pages

  • Your two minute crash course if you do not use it already.
    • Git beginner tutorial for LIDA interns can be found here
  • Free hosting for Open Source (OS) projects
  • Free Coninuous Delivery again for OS
  • Free Visibility (Shop Window - Nik’s description)
  • Example GH pages:

Let us do it

  • We can use the TGVE in two ways, but in this session can only do one
  • An easy to use front-end only on github
  • A little more advanced way using R (time left?)

Usin github pages

SPENSER

  • Wiki entry to do this
    1. Use the template repository & Setup github pages on new github settings
    2. Get data and add it to your repository or pull it from external source, anywhere with “CORS” permissions
    3. Change the environment variables on the Actions. Commit and have a coffee

Using R (time left?)

  • Get some data and in case of separate geography and “csv” make sure there is a matching geography column
  • Some plumbing using R package plumber
  • Build and Run
  • Reproducible R code example running TGVE: here
  • Wiki entry here

Conclusions

  • We can process and visualize spatio-temporal data at scale
  • We can do this better than alternatives, there is better tech
  • We should contribute to the OS community
  • TGVE has a clear future path for more funding and wider contribution
  • TGVE is built on the right tools hoping to be used

Thank you

graph vis

  • www.turing.ac.uk/research/research-projects/turing-geovisualization-engine
  • repo home: www.github.com/layik/eAtlas

References

Beecham, Roger, Jason Dykes, Layik Hama, and Nik Lomax. 2021. “On the Use of ‘Glyphmaps’ for Analysing the Scale and Temporal Spread of Covid-19 Reported Cases.” ISPRS International Journal of Geo-Information 10 (4). https://doi.org/10.3390/ijgi10040213.