Dashboard de Ventas Minoristas con Python

import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
# Cargar el dataset
file_path = "./online_retail.csv"
df = pd.read_csv(file_path)
# Rellenar valores nulos
df = df.fillna("Sin Dato")
# Limpiar los datos de los valores duplicados y los almaceno en un nuevo df
df_cleaned = df.drop_duplicates()
# Convertir columna fechas en formato fecha
df_cleaned["InvoiceDate"] = pd.to_datetime(df_cleaned["InvoiceDate"])
df_cleaned["Description"] = df_cleaned["Description"].astype(str)
# funcion para categorizar datos
def categorize(amount):
    if amount < 25:
        return "Low"
    elif amount < 200:
        return "Medium"
    else:
        return "Hight"
# Creacion de nuevas columnas con nuevos datos a obtener
df_cleaned["TotalAmount"] = df_cleaned["Quantity"] * df_cleaned["UnitPrice"]
df_cleaned["Year"] = df_cleaned["InvoiceDate"].dt.year
df_cleaned["Month"] = df_cleaned["InvoiceDate"].dt.month
df_cleaned["Day"] = df_cleaned["InvoiceDate"].dt.day
df_cleaned["Semester"] = df_cleaned["Month"].apply(lambda x:1 if x<=6 else 2)
df_cleaned["Trimestre"] = df_cleaned["Month"].apply(lambda x:1 if x<=3 else 2 if x<=6 else 3 if x<=9 else 4)
df_cleaned["Categorize"] = df_cleaned["TotalAmount"].apply(categorize)
# Creamos un nuevo df con agrupaciones o informaciòn especifica 
sales_by_year = df_cleaned.groupby("Year")["TotalAmount"].sum().reset_index()
sales_by_semester = df_cleaned.groupby(["Year","Semester"])["TotalAmount"].sum().reset_index()
sales_by_trimestre = df_cleaned.groupby(["Year","Trimestre"])["TotalAmount"].sum().reset_index()
sales_by_month = df_cleaned.groupby(["Year","Month"])["TotalAmount"].sum()
# Obtenemos el top 10 de prodcutos
top_products = df_cleaned.groupby("StockCode")["Quantity"].sum().sort_values(ascending=False).head(10)
top_products = top_products.reset_index()
top_products = pd.merge(top_products, df_cleaned[["StockCode", "Description"]].drop_duplicates(), on="StockCode", how="left")
total_returns = df_cleaned[df_cleaned["Quantity"] < 0].shape[0]
total_non_returns = df_cleaned[df_cleaned["Quantity"] >= 0].shape[0]
total_low = df_cleaned[df_cleaned["Categorize"] == "Low"].shape[0]
total_medium = df_cleaned[df_cleaned["Categorize"] == "Medium"].shape[0]
total_hight = df_cleaned[df_cleaned["Categorize"] == "Hight"].shape[0]
sales_by_country = df_cleaned.groupby("Country")["TotalAmount"].sum()
sales_by_country_top10 = sales_by_country.sort_values(ascending=False).head(10)
sales_by_categorize = df_cleaned.groupby("Categorize")["TotalAmount"].count()

# Crea graficos con matplotlib con funciones
def graph_returns():
    labels = ["Devoluciones", "No Devoluciones"]
    sizes = [total_returns, total_non_returns]
    colors = ["skyblue", "orange"]
    plt.figure(figsize=(8,8))
    plt.pie(sizes, labels=labels, colors=colors , autopct="%1.1f%%", startangle=140)
    plt.title("% transacciones con y sin Devolucion")
    plt.axis("equal")
    plt.show()   
def graph_categorize():
    labels = ["Low", "Medium", "Hight"]
    sizes = [total_low, total_medium, total_hight]
    colors = ["skyblue", "orange", "gold"]
    plt.figure(figsize=(8,8))
    plt.pie(sizes, labels=labels, colors=colors , autopct="%1.1f%%", startangle=140)
    plt.title("% por categoria de monto")
    plt.axis("equal")
    plt.show()
def graph_sales_month():
    sales_by_month.index = sales_by_month.index.map(lambda x: f"{x[0]}-{x[1]:02d}")
    plt.figure(figsize=(8,6))
    plt.plot(sales_by_month.index, sales_by_month.values, marker="o", color="green")
    for i, value in enumerate(sales_by_month.values):
        plt.text(i, value, f"{value/1e6:.1f}M", ha='center', va='bottom', fontsize=9)
    plt.title("Ventas por mes")
    plt.xlabel("Meses")
    plt.xticks(rotation=45)
    plt.ylabel("Ventas (millones)")
    plt.show()
def table_sales_semester():
    sales_by_semester["TotalAmount"] = sales_by_semester["TotalAmount"].map(lambda x: f"$ {x:,.0f}")
    fig, ax = plt.subplots(figsize=(6, 2))
    ax.axis("off")  # Oculta los ejes
    # Dibujar tabla
    tabla = ax.table(
        cellText=sales_by_semester.values,
        colLabels=sales_by_semester.columns,
        cellLoc="center",
        loc="center"
    )
    tabla.auto_set_font_size(False)
    tabla.set_fontsize(10)
    tabla.scale(1.2, 1.2)
    plt.show()
def graph_sales_country():
    plt.figure(figsize=(10,4))
    bars = plt.bar(sales_by_country_top10.index, sales_by_country_top10.values, color = "skyblue", label = "Ventas por Pais")
    for bar in bars:
        yval = bar.get_height()
        plt.text(bar.get_x() + bar.get_width()/2, yval, f"{yval/1e6:.1f}M", ha='center', va='bottom', fontsize=9)
    plt.title("Top 10 Paises con Mayores Ventas")
    plt.xlabel("Top 10 Paises")
    plt.xticks(rotation=45)
    plt.ylabel("Ventas (millones)")
    plt.show()
def graph_top10_products():
    plt.figure(figsize=(12,8))
    plt.barh(top_products["Description"], top_products["Quantity"], color = "salmon")
    plt.title("Top 10 Prodcutos mas Vendidos")
    plt.xlabel("Cantidad Vendida")
    #plt.xticks(rotation=45)
    plt.ylabel("Prodcuto")
    plt.gca().invert_yaxis()
    plt.show()
graph_sales_month()
graph_top10_products()
graph_categorize()
graph_returns()
graph_sales_country()
table_sales_semester()