Material for my talk "Approaching geovisualization and remote sensing with GeoViews" @ GeoPython 2018.

The best way to install all the dependencies for this notebook is to create a conda environment. Either Miniconda or Anaconda are good.

You can use the environment.yml file included in this repository to create a conda environment identical to the one I used:

conda env create --file environment.yml

Otherwise you can create and activate a new conda environment with:

conda create --name geopython-2018 python=3.6 --yes
source activate geopython-2018

And install all the dependencies with:

# geoviews (it depends on holoviews)
conda install -c conda-forge -c ioam holoviews geoviews --yes
# Additional packages (for the examples in the notebook)
conda install xarray  -y
conda install -c conda-forge iris -y
conda install -c conda-forge iris-sample-data
conda install -c conda-forge geopandas -y

You will also need to download the Bokeh sample data:

import bokeh.sampledata
bokeh.sampledata.download()

If you run into issues when importing geopandas, try manually downgrading fiona:

conda install -c conda forge fiona=1.7.10

In order to run the notebook you will also need some data files.

Here is where to find the data used in the notebook:

• Shapefiles of Basel districts @ https://wiki.openstreetmap.org/wiki/City_of_Basel_Suburb_Import (raw data, not the .osm file)
• Shapefiles of Basel parks and cafes @ https://extract.bbbike.org/
• netCDF file (Air quality) @ NOAA
• Shapefile of American Indian/Alaska Native Areas/Hawaiian Home Lands @ United States Censun Bureau

Create a data directory in the root of the repository and drop the files there.

Open a terminal and type:

jupyter nbconvert Jupyter Notebook.ipynb --to slides --post serve

You could freeze your environment with:

conda env export > environment.yml

To remove this conda environment, run:

conda env remove -n geopython-2018