Loading data

The package is instantiated by providing a dataframe containing the geodata ,e.g. dataframe=df, a list of columns of the dataframe, which should be shown as information for every geo feature, e.g. properties=df.columns to show everything, and the column titles for longitudal and latitudal data (standard parameter names are lat='latitude' and lon='longitude').

data = Convert2GeoJson(
          df,
          df.columns,
          lat='latitude',
          lon='longitude'
       )

The geo data is expected to be gievn as float numbers. Rows with empty longitudal or latitudel data, or marked as NaN are droped from the supplied dataframe.

The original dataframe can however still be accessed as data.rawDF.

Generate geoJson

To generate the geoJSON simple run

data.convert()

This generates a geoJSON formated dictionary which is available at

data.geojson()

To save the geoJSON to a file, run

data.save()

You can provide parameters for the name and path, standards are name='data.geojson' and path='.'.

Plotting

For plotting the package is leveraging the ipyleaflet package which is also available as a JupyterLab extension.

For a first overview use

data.display()

This plots the geoJSON features as a layer on a basemap.

The basic plotting can be changed by parameters for the map layout (mapLayout=dict()), and the basemap data (basemap=dict()).

To use a custom basemap, you can provide a dictionary of the format

customBasemap = {
          'url': 'https://{s}.URL_to_mapdata/{z}/{x}/{y}.png',
          'max_zoom': 'max available zoom level',
          'attribution': 'Attribution',
          'name': 'Name for layer control'
          }

Screenshot

Grouped Plotting

For densely distributed geographical data, to styling options are available.

By choosing style='grouped' ipyleaflets MarkerCluster is used to show groups of markers depending on the zoom level. By clicking on a cluster, the map zooms to the level, which contains the selected markers in the cluster .

Additionally, since markers are now single entities, by clicking on any marker a popup shows the information of the dataframe belonging to the geographical point.

Screenshot

Categorical plotting as pie chart distribution

By choosing style='pie' and providing a category found in the dataframe columns, by setting groupBy='Category', the package generates a standalone map showing the clustered markers as pie charts separated into sub-groups by the chosen category.

Screenshot