Work with shapefiles (without Geopandas)

Here we will use numpy and basemap.

pip install numpy
pip install basemap

To install basemap on WIndows, download a package here, according your versions, and install it with wheel.

Thanks to ramiro.org. I got the code below on his website, I just adapted it.

Suppose you have an Excel file with a country field, you get count and map them with a country shape. The country names in the Excel file and in the shape must be the same.

import matplotlib as mpl
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
 
from matplotlib.patches import Polygon
from matplotlib.collections import PatchCollection
from mpl_toolkits.basemap import Basemap
 
from PIL import Image, ImageOps
 
shp_simple_countries = r'C:/Users/Downloads/simple_countries/simple_countries'
 
inputExcelFile = r'C:/Users/Downloads/My file.xslx'
 
df = pd.read_excel(inputExcelFile, sheet_name='Export', engine='openpyxl', usecols=['ID', 'My Country Field'])
 
# Count the countries
df_Country_count = pd.DataFrame(df.groupby(['My Country Field'], dropna=True).size(), columns=['Total']).sort_values(['Total'], ascending=False).reset_index()
df_Country_count = df_Country_count.fillna('Unknow')
 
df_Country_count['Percent'] = (df_Country_count['Total'] / df_Country_count['Total'].sum()) * 100
df_Country_count['Percent'] = df_Country_count['Percent'].round(decimals=2)
 
# Set countries as index
df_Country_count.set_index('My Country Field', inplace=True)
 
my_values = df_Country_count['Percent']
 
num_colors = 30
cm = plt.get_cmap('Blues')
 
scheme = [cm(i / num_colors) for i in range(num_colors)]
my_range = np.linspace(my_values.min(), my_values.max(), num_colors)
 
# -1 TO AVOID SEARCHS IN A PANDAS DATA-FRAME INCLUDING START AND STOP VALUE (I think ...)
df_Country_count['Percent'] = np.digitize(my_values, my_range) - 1
 
map1 = plt.figure(figsize=(14, 8))
ax = map1.add_subplot(111, frame_on=False)
 
map1.suptitle('Countries', fontsize=30, y=.95)
 
m = Basemap(lon_0=0, projection='robin')
m.drawmapboundary(color='w')
 
m.readshapefile(shp_simple_countries, 'units', color='#444444', linewidth=.2, default_encoding='iso-8859-15')
 
# Create the chloro map
for info, shape in zip(m.units_info, m.units):
    shp_ctry = info['COUNTRY_HB']
    if shp_ctry not in df_Country_count.index:
        color = '#dddddd'
    else:
        color = scheme[df_Country_count.loc[shp_ctry]['Percent']]
 
    patches = [Polygon(np.array(shape), True)]
    pc = PatchCollection(patches)
    pc.set_facecolor(color)
    ax.add_collection(pc)
 
# Cover up Antarctica so legend can be placed over it
ax.axhspan(0, 1000 * 1800, facecolor='w', edgecolor='w', zorder=2)
 
# Draw color legend
ax_legend = map1.add_axes([0.2, 0.14, 0.6, 0.03], zorder=3)
cmap = mpl.colors.ListedColormap(scheme)
 
cb = mpl.colorbar.ColorbarBase(ax_legend, cmap=cmap, ticks=my_range, boundaries=my_range, orientation='horizontal')
cb.ax.set_xticklabels([str(round(i, 1)) for i in my_range])
cb.ax.tick_params(labelsize=7)
cb.set_label('Percentage', rotation=0)
#cb.remove()
 
# Set the map footer
# description = 'Bla bla bla'
# plt.annotate(description, xy=(-.8, -3.2), size=14, xycoords='axes fraction')
 
map1.savefig('C:/Users/Downloads/mymap1.png', dpi=100, bbox_inches='tight')
 
plt.show()
plt.clf()
 
im = Image.open('C:/Users/Downloads/mymap1.png')
bordered = ImageOps.expand(im, border=1, fill=(0, 0, 0))
bordered.save('C:/Users/Downloads/mymap1.png')

Zoom on France

map1.suptitle('Académies', fontsize=20, y=.87)
 
m = Basemap(projection='merc', resolution='l', \
llcrnrlon=-7, # Longitude minimale : étend vers l'est \
llcrnrlat=39.5, # Latitude minimale : étend vers le sud \
urcrnrlon=13, # Longitude maximale : étend vers l'ouest \
urcrnrlat=52) # Latitude maximale : étend vers le nord

Add labels or POI

ax.text(0.05, # Vers la droite
0.59, # Vers le haut
'Guadeloupe', fontsize=10, ha='center', transform=ax.transAxes)

Another example

If you use the shape attached named France-Departements-Deformation.shp:

############## Carte df_DepEtablissement
# Set academies as index
df_DepEtablissement.set_index('nom_departement_etablissement', inplace=True)
my_values = df_DepEtablissement['Pourcentage']
num_colors = 30
cm = plt.get_cmap('Blues')
scheme = [cm(i / num_colors) for i in range(num_colors)]
my_range = np.linspace(my_values.min(), my_values.max(), num_colors)
 
# -1 TO AVOID SEARCHS IN A PANDAS DATA-FRAME INCLUDING START AND STOP VALUE (I think ...)
df_DepEtablissement['Pourcentage'] = np.digitize(my_values, my_range) - 1
 
map2 = plt.figure(figsize=(14, 10))
ax = map2.add_subplot(111, frame_on=False)
 
map2.suptitle('Départements', fontsize=20, y=.87)
 
m = Basemap(projection='merc', resolution='l', 
llcrnrlon=-9, \ # Longitude minimale : étend vers l'est 
llcrnrlat=39.5, \ # Latitude minimale : étend vers le sud 
urcrnrlon=15, \ # Longitude maximale : étend vers l'ouest
urcrnrlat=52) \ # Latitude maximale : étend vers le nord
 
m.drawmapboundary(color='w')
 
m.readshapefile(shp_departements, 'units', color='#444444', linewidth=.2, default_encoding='utf-8')
 
# Create the chloro map
for info, shape in zip(m.units_info, m.units):
    shp_departements = info['dep_name']
    if shp_departements not in df_DepEtablissement.index:
        color = '#dddddd'
    else:
        color = scheme[df_DepEtablissement.loc[shp_departements]['Pourcentage']]
    patches = [Polygon(np.array(shape), True)]
    pc = PatchCollection(patches)
    pc.set_facecolor(color)
    ax.add_collection(pc)
 
# Cover up Antarctica so legend can be placed over it
# ax.axhspan(0, 1000 * 1800, facecolor='w', edgecolor='w', zorder=2)
 
# Draw color legend
ax_legend = map2.add_axes([0.2, 0.14, 0.6, 0.03], zorder=3)
cmap = mpl.colors.ListedColormap(scheme)
cb = mpl.colorbar.ColorbarBase(ax_legend, cmap=cmap, ticks=my_range, boundaries=my_range, orientation='horizontal')
cb.ax.set_xticklabels([str(round(i, 1)) for i in my_range])
cb.ax.tick_params(labelsize=7)
cb.set_label('Pourcentage', rotation=0)
# cb.remove()
 
ax.text(0.125, 0.565,'Guadeloupe', fontsize=10, ha='center', transform=ax.transAxes)
 
ax.text(0.175, 0.46, 'Martinique', fontsize=10, ha='center', transform=ax.transAxes)
 
ax.text(0.1, 0.18, 'Guyane', fontsize=10, ha='center', transform=ax.transAxes)
 
ax.text(0.42, 0.155, 'Mayotte', fontsize=10, ha='center', transform=ax.transAxes)
 
ax.text(0.6, 0.155, 'La Réunion', fontsize=10, ha='center', transform=ax.transAxes)
 
ax.text(0.73, 0.15, 'Corse', fontsize=10, ha='center', transform=ax.transAxes)
 
ax.text(0.83, 0.41, 'Nouvelle-Calédonie', fontsize=10, ha='center', transform=ax.transAxes)
 
ax.text(0.8, 0.515, 'Polynésie Française', fontsize=10, ha='center', transform=ax.transAxes)
 
ax.text(0.836, 0.69, 'Inconnu', fontsize=8, ha='center', transform=ax.transAxes)
 
ax.text(0.86, 0.903, 'Étranger', fontsize=8, ha='center', transform=ax.transAxes)
 
map2.savefig(downloadsDir + 'mymap2.png', dpi=80, bbox_inches='tight')
 
# plt.show()
# plt.clf()
 
im = Image.open(downloadsDir + 'mymap2.png')
bordered = ImageOps.expand(im, border=1, fill=(0, 0, 0))
bordered.save(downloadsDir + 'mymap2.png')
 
# INSERT IN EXCEL
img = openpyxl.drawing.image.Image(downloadsDir+'mymap2.png')
img.anchor = 'E4'
workbook['Départements établissements'].add_image(img)
workbook.save(statsFile)
 
################# REMOVE PICTURES
os.remove(downloadsDir + 'mymap2.png')

Liens ou pièces jointes

France-Departements-Deformation.zip	[France-Departements-Deformation]	335 Ko
simple_countries.zip	[simple_countries]	1880 Ko