Function with dates

Question

I created a function with the objective that converts the value of Colombian pesos to dollars depending on the date of the index in a dataframe:

def pesosdolar(fecha):
  fecha=pd.to_datetime(fecha)
  plus_one_day = fecha + timedelta(days=1)
  minus_one_day= fecha - timedelta(days=1)
  try: 
    copusd = yf.Ticker("USDCOP=X").history(start=fecha, end=fecha)
    return copusd['Close'].iloc[0]
  except IndexError:
    alt = yf.Ticker("USDCOP=X").history(start=fecha - timedelta(days=3), end=fecha - timedelta(days=1))
    return alt['Close'].mean()

when I write a simple date ('2022-04-28' for example) it works perfect, the problem is when I try to put the dataframe

pesosdolar(df_ordenes['createdAt'])

The warning:

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

Further, this does not work:

df_ordenes = df_ordenes.assign(dolar = lambda x: (x['amount_in'] / pesosdolar(x['createdAt'])))