Property Prices

A map of property prices.

Datasets

This project uses the following datasets:

• Ordnance Survey Boundary-Line
• Ordnance Survey Code-Point Open
• HM Land Registry Price Paid Data

These datasets are released under the Open Government Licence. The following attribution statements apply:

• Contains OS data © Crown copyright and database right 2020
• Contains Royal Mail data © Royal Mail copyright and database right 2020
• Contains National Statistics data © Crown copyright and database right 2020
• Contains HM Land Registry data © Crown copyright and database right 2020

Directory structure

This project uses the Cookiecutter Data Science directory structure.

.
├── bin              <- QA scripts
├── data             <- Data processing stages
│   ├── external     <- Stage 1, data that are external to this project
│   ├── interim      <- Stage 3
│   ├── processed    <- Stage 4
│   └── raw          <- Stage 2
├── models           <- Data for visualization
├── reports          <- Analysis
│   └── figures      <- Figures for analysis
├── specs            <- Vega specifications for visualization
├── src              <- Python source code
│   └── data         <- Data processing scripts
└── tests            <- Tests for Python source code

(Run tree -d -L 2 -I __*__ to create the above.)

Setup

First, install Poetry, topojson-server, topojson-simplify, and vega-cli. Then, run:

poetry install & poetry run dvc pull

Motivations

• To make a map with Vega. How much control does Vega give over, for example, titles, legends, and labels? To what extent is command-line cartography possible without using D3.js directly? How easy is it to switch between in-browser rendering, using the Vega Editor, and offline rendering, using vg2svg?

• To use Vega with projected spatial data. The Ordnance Survey datasets contain eastings and northings using the British National Grid, rather than longitude and latitude using WGS84. How to configure a Vega projection with projected spatial data?

• To use the Cookiecutter Data Science directory structure for a project that doesn't involve modelling. To what extent is the data/<stage>/<dataset> directory structure effective?

• To use the Cookiecutter Data Science directory structure with Poetry. When a project isn't a library, it's convenient to keep Python source code in the src directory. However, Poetry has to be tricked into initializing this directory structure: Name the package src and rename the parent directory, after running poetry new src. Are there any pitfalls?

• To learn more about DVC. What are the strengths and limitations of data pipelines, for example?

Licence

This project, but not its datasets (see above), is released under the MIT Licence.