|
|
import pandas as pd
|
|
|
import psycopg2 as psy
|
|
|
import getpass
|
|
|
import matplotlib.pyplot as plt
|
|
|
|
|
|
data = pd. read_csv (r'vgsales.csv')
|
|
|
df = pd.DataFrame(data)
|
|
|
df = df.drop_duplicates()
|
|
|
|
|
|
co = None
|
|
|
|
|
|
try:
|
|
|
co = psy. connect(host='londres',
|
|
|
database ='dbanperederi',
|
|
|
user='anperederi',
|
|
|
password = getpass.getpass("Mot de passe:"))
|
|
|
|
|
|
#Q1 Écrire la requête SQL permettant de calculer le total des ventes de jeux dans le monde par genre.
|
|
|
datafr = pd.read_sql('''SELECT round(sum(global_sales),2) as total_vente, genre
|
|
|
FROM VGSales
|
|
|
GROUP BY genre
|
|
|
ORDER BY total_vente DESC;''', con=co)
|
|
|
|
|
|
fig = datafr .plot.pie(y='total_vente', labels=datafr['genre'], autopct='%1.1f%%', figsize=(5, 5), legend=False) # Generation du graphique
|
|
|
plt.show () # Affichage
|
|
|
|
|
|
#Q2 Écrire la requête SQL permettant de calculer e pourcentage des ventes dans le monde par genre.
|
|
|
datafr = pd.read_sql('''SELECT (sum(global_sales)/(SELECT sum(global_sales) FROM VGSales))*100 as pourcentage
|
|
|
FROM VGSales
|
|
|
GROUP BY genre
|
|
|
ORDER BY pourcentage DESC;''', con=co)
|
|
|
|
|
|
fig = datafr.plot(kind='pie', legend=False ,title='Ventes par plateforme', figsize=(5, 5), fontsize=12, subplots=True)
|
|
|
plt.show()
|
|
|
|
|
|
# fig = datafr .plot.pie(y='pourcentage', labels=datafr['genre'], autopct='%1.1f%%', figsize=(5, 5), legend=False) # Generation du graphique
|
|
|
# plt.show () # Affichage
|
|
|
|
|
|
#Q3 Afficher les résultats sous forme de diagramme camembert.
|
|
|
datafr = pd.read_sql('''SELECT (sum(global_sales)/(SELECT sum(global_sales) FROM VGSales))*100 as pourcentage
|
|
|
FROM VGSales
|
|
|
GROUP BY genre
|
|
|
ORDER BY pourcentage DESC;''', con=co)
|
|
|
|
|
|
fig = datafr.plot(kind='pie', legend=False ,title='Ventes par plateforme', figsize=(5, 5), fontsize=12, subplots=True)
|
|
|
plt.show()
|
|
|
|
|
|
# fig = datafr .plot.pie(y='pourcentage', labels=datafr['genre'], autopct='%1.1f%%', figsize=(5, 5), legend=False) # Generation du graphique
|
|
|
# plt.show () # Affichage
|
|
|
|
|
|
#Q4 Afficher le total des ventes par genre sous forme de diagrammes camembert de façon à voir sur la même figure :
|
|
|
# un camembert représentant les ventes avant l’année 1990 (exclue),
|
|
|
# un camembert représentant les ventes entre 1990 et 1999,
|
|
|
# un camembert représentant les ventes entre 2000 et 2009,
|
|
|
# un camembert représentant les ventes à partir de 2010.
|
|
|
datafr1 = pd.read_sql('''SELECT (sum(global_sales)/(SELECT sum(global_sales) FROM VGSales WHERE year < 1990))*100 as pourcentage
|
|
|
FROM VGSales
|
|
|
WHERE year < 1990
|
|
|
GROUP BY genre
|
|
|
ORDER BY pourcentage DESC;''', con=co)
|
|
|
|
|
|
datafr2 = pd.read_sql('''SELECT (sum(global_sales)/(SELECT sum(global_sales) FROM VGSales WHERE year >= 1990 AND year < 2000))*100 as pourcentage
|
|
|
FROM VGSales
|
|
|
WHERE year >= 1990 AND year <= 1999
|
|
|
GROUP BY genre
|
|
|
ORDER BY pourcentage DESC;''', con=co)
|
|
|
|
|
|
datafr3 = pd.read_sql('''SELECT (sum(global_sales)/(SELECT sum(global_sales) FROM VGSales WHERE year >= 2000 AND year < 2010))*100 as pourcentage
|
|
|
FROM VGSales
|
|
|
WHERE year >= 2000 AND year <= 2009
|
|
|
GROUP BY genre
|
|
|
ORDER BY pourcentage DESC;''', con=co)
|
|
|
|
|
|
datafr4 = pd.read_sql('''SELECT (sum(global_sales)/(SELECT sum(global_sales) FROM VGSales WHERE year >= 2010))*100 as pourcentage
|
|
|
FROM VGSales
|
|
|
WHERE year >= 2010
|
|
|
GROUP BY genre
|
|
|
ORDER BY pourcentage DESC;''', con=co)
|
|
|
|
|
|
_, axes = plt.subplots(nrows=2, ncols=2)
|
|
|
|
|
|
fig = datafr1.plot(y=0, kind='pie', legend=False ,title='Ventes par plateforme', ax=axes[0,0])
|
|
|
fig = datafr2.plot(y=0, kind='pie', legend=False ,title='Ventes par plateforme', ax=axes[0,1])
|
|
|
fig = datafr3.plot(y=0, kind='pie', legend=False ,title='Ventes par plateforme', ax=axes[1,0])
|
|
|
fig = datafr4.plot(y=0, kind='pie', legend=False ,title='Ventes par plateforme', ax=axes[1,1])
|
|
|
plt.show()
|
|
|
|
|
|
# fig = datafr .plot.pie(y='pourcentage', labels=datafr['genre'], autopct='%1.1f%%', figsize=(5, 5), legend=False) # Generation du graphique
|
|
|
# plt.show () # Affichage
|
|
|
|
|
|
except (Exception , psy.DatabaseError ) as error :
|
|
|
print ( error )
|
|
|
finally :
|
|
|
if co is not None:
|
|
|
co.close () |
