Jupy-Geospatial

A complete jupyter lab setup for geospatial development in python. Builds and installs all dependencies in a docker container, then runs a jupyter lab server which can be accessed from the host machine.

Why

Geospatial development in python should be fun! Instead every-time I find myself reinstalling packages (Cartopy, PROJ and GEOS) and getting installation conflicts along the way. This docker image aims to handle the installation process and provide a clean working environment.

Installation

Requires docker

Clone

git clone https://github.com/HarveyBates/jupy-geospatial
cd jupy-geospatial

Build docker container

Runs once, takes about 20 mins.

docker build -t pygeo .

Run

Installs required python dependencies (if required).

docker run --name pygeo_container -it -p 8888:8888 -v ${PWD}:/root/jupy-geospatial pygeo

The terminal will output a url that can be navigated to on the host computer. Something like: 127.0.0.1:8888/lab/?token=<unique_token>. Or you can navigate to localhost:8888/lab and enter the token provided to you.

License

This work is MIT licensed, as found in the LICENSE file.