CAP213_PDE_DataScience_for_Business_sesion4.py

# Modulos y datos
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
uber = pd.read_csv("https://raw.githubusercontent.com/robintux/Datasets4StackOverFlowQuestions/refs/heads/master/Datos_Uber_2024.csv")

# Preprocesamiento 
uber['Date'] = pd.to_datetime(uber['Date'], format='%Y-%m-%d')
uber['Year'] = uber['Date'].dt.year
uber['Month'] = uber['Date'].dt.month
uber['MonthName'] = uber['Date'].dt.month_name()
uber['DayOfWeek'] = uber['Date'].dt.dayofweek  # 0 = lunes, 6 = domingo
uber['DayName'] = uber['Date'].dt.day_name()
uber['WeekOfYear'] = uber['Date'].dt.isocalendar().week
uber['Time'] = pd.to_datetime(uber['Time'], format='%H:%M:%S').dt.time
uber['Datetime'] = pd.to_datetime(uber['Date'].astype(str) + ' ' + uber['Time'].astype(str))


# Extraer hora del día como entero (0 a 23)
uber['Hour'] = uber['Datetime'].dt.hour

# Categorizar horas del día (opcional)
def hora_bloque(h):
    if 5 <= h < 9: return 'Mañana (5-9)'
    elif 9 <= h < 13: return 'Media mañana (9-13)'
    elif 13 <= h < 17: return 'Tarde (13-17)'
    elif 17 <= h < 21: return 'Pico (17-21)'
    elif 21 <= h < 24: return 'Noche (21-24)'
    else: return 'Madrugada (0-5)'

uber['HourBlock'] = uber['Hour'].apply(hora_bloque)