Mapsplotlib

Custom Python plots on a Google Maps background. A flexible matplotlib like interface to generate many types of plots on top of Google Maps.

This package was renamed from the legacy tcassou/gmaps due to an unfortunate conflict in names with a package from Pypi.

Setup

Simply install from pip:

pip install mapsplotlib

For now, it seems that the Google Static Maps API can still be hit without using an API key:

from mapsplotlib import mapsplot as mplt

# all plots can now be performed here

This behavior will probably change in the future though... In this case you'll then need to have a Google Static Maps API key, go to https://console.developers.google.com, create a project, enable Google Static Maps API and get your server key. Note that billing details have to be enabled for your account for the API calls to work. Before plotting maps, you'll have to register your key (only once for each session you start):

from mapsplotlib import mapsplot as mplt

mplt.register_api_key('your_google_api_key_here')

# all plots can now be performed here

Examples

Marker Plots

Simply plotting markers on a map. Consider a pandas DataFrame df defined as follows:

|   | latitude | longitude |  color |  size | label |
|---|----------|-----------|--------|-------|-------|
| 0 |  48.8770 |  2.30698  |  blue  |  tiny |       |
| 1 |  48.8708 |  2.30523  |   red  | small |       |
| 2 |  48.8733 |  2.32403  | orange |  mid  |   A   |
| 3 |  48.8728 |  2.30491  |  black |  mid  |   Z   |
| 4 |  48.8644 |  2.33160  | purple |  mid  |   0   |

Simply use (assuming mapsplot was imported already, and your key registered)

mplt.plot_markers(df)

will produce

Density Plots

The only thing you need is a pandas DataFrame df containing a 'latitude' and a 'longitude' columns, describing locations.

mplt.density_plot(df['latitude'], df['longitude'])

Heat Maps

This time your pandas DataFrame df will need an extra 'value' column, describing the metric you want to plot (you may have to normalize it properly for a good rendering).

mplt.heatmap(df['latitude'], df['longitude'], df['value'])

Scatter Plots

Let's assume your pandas DataFrame df has a numerical 'cluster' column, describing clusters of geographical points. You can produce plots like the following:

mplt.scatter(df['latitude'], df['longitude'], colors=df['cluster'])

Polygon Plots

Still with the same DataFrame df and its 'cluster' column, plotting clusters and their convex hull.

mplt.polygons(df['latitude'], df['longitude'], df['cluster'])

More to come!

Requirements

pandas >= 0.13.1
numpy >= 1.8.2
scipy >= 0.13.3
matplotlib >= 1.3.1
requests >= 2.7.0
requests>=2.18.4
pillow>=4.3.0

ccraddock / mapsplotlib