Wednesday, 2 May 2018

Python, REST, Json and gmaps : plotting data

Hi all,

Let’s continue with another small and quick Python use case on data.

Here is a use case to plot gmaps heatmap with Python. The data is coming from a live REST service (New York city bike, because it is live, free and easy) and a simple JSON parsing is done to catch the location data.

I’m using Jupyter Notebook, which is THE Python platform for data manipulation, exploration, testing, science etc …

Here is the code, fairly simple. Of course you need a gmaps api key.

I decided to render a heatmap because it is eye candy but you can render anything you want (marks…). Next time, I will show you how to embed data into the map and add some real time features (read the previous article about real time plotting : http://open-bi.blogspot.ch/2018/04/simple-and-compact-python-bitcoin-ticker.html).


import gmaps
import requests
import gmaps.datasets
import numpy as np

gmaps.configure(api_key="xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx") # Your Google API key

url = 'https://gbfs.citibikenyc.com/gbfs/en/station_information.json'
print('Processing url request ....')
r = requests.get(url)
data = r.json()

coords = []
locations = []

for i in data['data']['stations']:
    lat = float(i['lat'])
    lon = float(i['lon'])
    coords = [lat, lon]
#hard coded coordinates for New York below, but you can do something better …
fig = gmaps.figure(center=(40.71448, -74.00598), zoom_level=12, layout={
        'width': '700px',
        'height': '800px',
        'padding': '3px',
        'border': '1px solid black'
Processing url request ....

No comments: