Plot flight data with matplolib

mpl.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import numpy as np
import pandas as pd
from mpl_toolkits.basemap import Basemap
import matplotlib.pyplot as plt
from matplotlib.colors import Normalize, LinearSegmentedColormap, PowerNorm


def plot_map(in_filename, color_mode='screen',
             out_filename='flights_map_mpl.png', absolute=False):
    """Plots the given CSV data files use matplotlib basemap and saves it to
    a PNG file.

    Args:
        in_filename: Filename of the CSV containing the data points.
        out_filename: Output image filename
        color_mode: Use 'screen' if you intend to use the visualisation for
                    on screen display. Use 'print' to save the visualisation
                    with printer-friendly colors.
        absolute: set to True if you want coloring to depend on your dataset
                  parameter value (ie for comparison).
                  When set to false, each coordinate pair gets a different
                  color.

    """

    if color_mode == 'screen':
        bg_color = (0.0, 0.0, 0, 1.0)
        coast_color = (204/255.0, 0, 153/255.0, 0.7)
        color_list = [(0.0, 0.0, 0.0, 0.0),
                      (204/255.0, 0, 153/255.0, 0.6),
                      (255/255.0, 204/255.0, 230/255.0, 1.0)]
    else:
        bg_color = (1.0, 1.0, 1.0, 1.0)
        coast_color = (10.0/255.0, 10.0/255.0, 10/255.0, 0.8)
        color_list = [(1.0, 1.0, 1.0, 0.0),
                      (255/255.0, 204/255.0, 230/255.0, 1.0),
                      (204/255.0, 0, 153/255.0, 0.6)
                      ]

    # define the expected CSV columns
    CSV_COLS = ('dep_lat', 'dep_lon', 'arr_lat', 'arr_lon',
                'nb_flights', 'CO2')

    routes = pd.read_csv(in_filename, names=CSV_COLS, na_values=['\\N'],
                         sep=';', skiprows=1)

    num_routes = len(routes.index)

    # normalize the dataset for color scale
    norm = PowerNorm(0.3, routes['nb_flights'].min(),
                     routes['nb_flights'].max())
    # norm = Normalize(routes['nb_flights'].min(), routes['nb_flights'].max())

    # create a linear color scale with enough colors
    if absolute:
        n = routes['nb_flights'].max()
    else:
        n = num_routes
    cmap = LinearSegmentedColormap.from_list('cmap_flights', color_list,
                                             N=n)
    # create the map and draw country boundaries
    plt.figure(figsize=(27, 20))
    m = Basemap(projection='mill', lon_0=0)
    m.drawcoastlines(color=coast_color, linewidth=1.0)
    m.fillcontinents(color=bg_color, lake_color=bg_color)
    m.drawmapboundary(fill_color=bg_color)

    # plot each route with its color depending on the number of flights
    for i, route in enumerate(routes.sort_values(by='nb_flights',
                              ascending=True).iterrows()):
        route = route[1]
        if absolute:
            color = cmap(norm(int(route['nb_flights'])))
        else:
            color = cmap(i * 1.0 / num_routes)

        line, = m.drawgreatcircle(route['dep_lon'], route['dep_lat'],
                                  route['arr_lon'], route['arr_lat'],
                                  linewidth=0.5, color=color)
        # if the path wraps the image, basemap plots a nasty line connecting
        # the points at the opposite border of the map.
        # we thus detect path that are bigger than 30km and split them
        # by adding a NaN
        path = line.get_path()
        cut_point, = np.where(np.abs(np.diff(path.vertices[:, 0])) > 30000e3)
        if len(cut_point) > 0:
            cut_point = cut_point[0]
            vertices = np.concatenate([path.vertices[:cut_point, :],
                                      [[np.nan, np.nan]],
                                      path.vertices[cut_point+1:, :]])
            path.codes = None  # treat vertices as a serie of line segments
            path.vertices = vertices

    # save the map
    plt.savefig(out_filename, format='png', bbox_inches='tight')

if __name__ == '__main__':
    # use 'screen' color mode for on-screen display. Use 'print' if you intend
    # to print the map
    plot_map('data.csv', 'screen', absolute=False)

Hi Hugo, first of all thank you for sharing! I am playing with it and I get an error in your line 75 saying "name 'absolute' is not defined", what could this be?

Thank you in advance!