ax = plt.axes(projection=projection)

ax.add_feature(cartopy.feature.OCEAN, facecolor='#add8e6')

ax.outline_patch.set_edgecolor(edgecolor)

values = sorted(list(df.unique()))

shpfilename = shpreader.natural_earth(resolution='110m', category='cultural', name='admin_0_countries')

reader = shpreader.Reader(shpfilename)

countries = reader.records()

for country in countries:

attribute = 'NAME_EN'

# get classification

try:

classification = df[country.attributes[attribute]] #ici

print(country.attributes)

break

ax.add_geometries(country.geometry, ccrs.PlateCarree(),

facecolor=(colors[values.index(classification)]),

label=country.attributes[attribute],

edgecolor='#000000',

linewidth=.25)

except:

# print(country.attributes[attribute])

pass

# legend

import matplotlib.patches as mpatches

handles = []

values = ['< à ' + str(v) for v in values]

for i in range(len(values)):

handles.append(mpatches.Rectangle((0, 0), 1, 1, facecolor=colors[i]))

plt.legend(handles, values,

loc='lower left', bbox_to_anchor=(0.025, -0.0),

fancybox=True, frameon=False, fontsize=5)

plt.title(title, fontsize=8)

title = 'maps/'+title+'.png'

plt.savefig(title, bbox_inches='tight', pad_inches=.2, dpi=300)

df = pd.read_csv('countries.csv', index_col='ISO_CODE')

projection = ccrs.Robinson()

title = 'Four Regions With The Same Population'

colors = ['#f4b042', '#92D050','#71a2d6','#b282ac','#DDDDDD']

plot_countries(df,projection,colors,title,edgecolor='white')

print('Done.\n')