Samfunnsøkonomi med datavitenskap

#!/usr/bin/python
# -*- coding: UTF-8 -*-

from bokeh.plotting import figure, output_file, show
import pandas as pd
import numpy as np

gapminder_filepath = "https://titlon.uit.no/hht/data/gapminder.csv"
g = pd.read_csv(gapminder_filepath)#reading data

#creating figure:
p = figure(
        title = "", 
        x_axis_type="log",
        x_axis_label='BNP per innbygger (NOK)',
        y_axis_label='Forventet antall leveår',
        tools="hover", 
        tooltips=[
            ("Land","@country"),
            ("BNP/innbygger","@gdp_capNOK"),
            ("Forventet alder","@life_exp_rnd"),
            ("Befolkning","@pop_mill_str")
            ],
        plot_height=580,
        plot_width=980)

#creating a dictionary of continents with norwegian translation:
continents={'Africa': 'Afrika', 'Americas': 'Sør/Nord-Amerika', 'Asia': 'Asia', 'Europe': 'Europa', 'Oceania': 'Oseania'}
#assigning a color hex-code to each continent
colors={'Africa': '#E14827', 'Americas': '#84E127', 'Asia': '#2792E1', 'Europe': '#BC27E1', 'Oceania': '#E04A6C'}

#setting the info displayed when hoovering the plot
g['colors']=g['continent'].apply(lambda cont: colors[cont])
g['size']=g['population'].apply(lambda pop: pop**0.5/300)
g['life_exp_rnd']=g['life_exp'].apply(lambda l: int(l))
g['pop_mill_str']=g['population'].apply(lambda pop:  '{:,}'.format(int(pop/1000000)))
g['gdp_capNOK']=g['gdp_cap'].apply(lambda gdp:  '{:,}'.format(int(8.9*gdp)).replace(',',' '))
g['continent_no']=g['continent'].apply(lambda cont:  continents[cont])

#obtaining a scatterplot:
p.scatter(x = 'gdp_cap', y = 'life_exp', size='size', source=g, color = 'colors', alpha = 0.8,legend_field='continent_no')

#formatting:
p.xaxis.major_label_overrides = { 1000: '1k', 10000: '10k', 100000: '100k' }
p.legend.location = "top_left"
p.legend[0].border_line_alpha=0
p.outline_line_alpha=0
p.grid[0].grid_line_alpha=0
p.grid[1].grid_line_alpha=0

#defining the outputfile:
output_file("./BNP_levealder.html")

#drawing it:
show(p)