# Artículos
conexion = sql3.connect('/work/data/articles.db')
sql_query = pd.read_sql_query('SELECT * FROM articles', conexion)
df_articles = pd.DataFrame(sql_query, columns=['article_id', 'article_name', 'unit_price'])
print(df_articles)
# Vendedores
df_sellers = pd.read_excel('/work/data/sellers.xlsx', index_col=0)
print(df_sellers)
# Órdenes
df_orders=pd.read_csv('/work/data/orders.csv')
print(df_orders)
# Exploración del df de artículos
print('Muestra de los datos')
print(df_articles.head())
print('\nFormato del dataframe')
print(df_articles.shape)
# Exploración del df de vendedores
print('Muestra de los datos')
print(df_sellers.head())
print('\nFormato del dataframe')
print(df_sellers.shape)
# Exploración del df de órdenes
print('Muestra de los datos')
print(df_orders.head())
print('\nFormato del dataframe')
print(df_orders.shape)
# Cambio el tipo de dato de 'unit_price' a float
df_articles['unit_price'] = df_articles['unit_price'].astype(float)
print(df_articles.dtypes)
# Cambio el índice del df_articles por la columna article_id
df_articles.set_index('article_id', inplace=True)
print(df_articles.head())
# Creo una copia del df_orders
mi_df = df_orders.copy()
mi_df = mi_df.assign(article_name = mi_df['article_id'])
mi_df = mi_df.assign(total_amount = mi_df['article_id'])
mi_df = mi_df.assign(seller_name = mi_df['seller_id'])
print(mi_df.head())
# Reemplazo los valores en las nuevas columnas del df
for i in range(len(mi_df.index)):
article = df_articles.loc[mi_df.loc[i]['article_name']]['article_name']
mi_df.loc[i,'article_name'] = article
mi_df.loc[i,'total_amount'] = mi_df.loc[i,'quantity'] * df_articles.loc[mi_df.loc[i]['total_amount']]['unit_price']
mi_df.loc[i,'seller_name'] = df_sellers.loc[mi_df.loc[i]['seller_name']]['seller_name']
print(mi_df.head())
# Borrar las columnas que no voy a utilizar
mi_df.drop(['order_id','article_id','seller_id'], axis='columns', inplace=True)
print(mi_df.head())
# Resolución Analítica
df = mi_df.groupby('article_name').sum()
orden_cantidad = df.sort_values('quantity', ascending=False)
print(orden_cantidad['quantity'].head(1))
# Resolución Gráfica
r1= sns.barplot(x=df.index,y=df['quantity'],data=df,order=df.sort_values('quantity', ascending=False).index)
r1.set_xlabel('Artículos')
r1.set_ylabel('Unidades vendidas')
plt.xticks(rotation=90)
plt.show()
# Resolución Analítica
df2 = (mi_df.groupby('article_name').sum()).sort_values('total_amount', ascending=False).head()
print(df2['total_amount'])
# Resolución Gráfica
e = (0.05,0,0,0,0)
c = ['gold','yellowgreen', 'lightcoral','lightskyblue','plum']
plt.pie(x=df2['total_amount'], explode=e, labels=df2.index, colors=c, startangle=90)
plt.show()
# Resolución Analítica
df3 = (mi_df.groupby('seller_name').sum()).sort_values('total_amount', ascending=False)
print('Respuesta:')
print(df3.head(1))
# Resolución Gráfica
r3 = plt.bar(df3.index, df3['total_amount'])
plt.ylabel('Monto total de ventas')
plt.xlabel('Vendedores')
plt.xticks(rotation=60)
plt.show()
# Resolución Analítica
df4 = (mi_df.groupby('week').sum()).sort_values('total_amount',ascending=False)
print(df4)
# Resolución Gráfica
plt.bar(df4.index,df4['total_amount'])
plt.show()
# Resolución Analítica
df5 = (mi_df.groupby('country_name').sum()).sort_values('total_amount', ascending=False).head()
print(df5['total_amount'])
# Resolución Gráfica
r5 = plt.bar(df5.index, df5['total_amount'])
plt.ylabel('Monto total de ventas')
plt.xlabel('Países')
plt.show()
# Resolución Gráfica
df6 = mi_df[(mi_df['article_name'] == 'Full Pc') | (mi_df['article_name'] == 'Notebook') | (mi_df['article_name'] == 'Smartphone')]
df6 = df6[['article_name','country_name','total_amount']]
df6 = df6.groupby(['article_name','country_name'])['total_amount'].sum().reset_index()
plt_order = df6.groupby('country_name')['total_amount'].sum().sort_values(ascending=False).index.values[0:5]
sns.catplot(y='country_name', x='total_amount', hue='article_name', data=df6, kind='bar', order=plt_order)
plt.xticks(rotation=90)
plt.title('Ingresos generados del top 3 de productos por países')
plt.ylabel('Países')
plt.xlabel('Ingresos')
plt.show()
# Resolución Analítica
df7 = (mi_df.groupby('country_name').sum()).sort_values('quantity', ascending=True).head()
print('Repuesta:')
print(df7['quantity'])
# Resolución Gráfica
e = (0.05,0,0,0,0)
c = ['gold','yellowgreen', 'lightcoral','lightskyblue','plum']
plt.pie(x=df7['quantity'], explode=e, labels=df7.index, colors=c, startangle=90)
plt.show()