Cycle for averages with Pandas, frequency every 5 minutes

Hi, I have this df. I am trying to obtain average values assigned to each hour of measured data with a frequency of 5 minutes. This is how I charge the df from MySQL:

import mysql.connector
from mysql.connector import errorcode
import pandas as pd
def data_base_connection(user,password,host,database):
config = {
    'user': user,
    'password': password,
    'host': host,
    'database': database,
    'raise_on_warnings': True,
}

try:
    cnx = mysql.connector.connect(**config)
    return cnx
except mysql.connector.Error as err:
    if err.errno == errorcode.ER_ACCESS_DENIED_ERROR:
        print("Something is wrong with your user name or password")
    elif err.errno == errorcode.ER_BAD_DB_ERROR:
        print("Database does not exist")
    else:
        print(err)

   cnx=data_base_connection("root","cigefi","127.0.0.1","a_siquirres")

   # Create cursor
   cursor=cnx.cursor()
   query1=("select fh,tc10, tc20, tc30, hr10, hr20, hr30, prec10, prec20,        prec30, vs10, vu10, dir10, desU10, vmax10, vs20, vu20, dir20, desU20, vmax20, vs30, vu30, dir30, desU30, vmax30, tsuelo, vw, paus from siqoriginal")
   # Excute Query

   cursor.execute(query1)

   rows = cursor.fetchall()
   df = pd.DataFrame([[ij for ij in i] for i in rows])  # Se cargan los       rows pandas, de la consulta que se hizo.
   df.rename(columns={0: 'fh', 1: 'tc10', 2:'tc20', 3:'tc30', 4: 'hr10', 5: 'hr20', 6: 'hr30', 7:'prec10', 8: 'prec20', 9: 'prec30', 10: 'vs10', 11: 'vu10', 12: 'dir10', 13: 'desU10', 14: 'vmax10', 15: 'vs20', 16: 'vu20', 17: 'dir20', 18: 'desU20', 19: 'vmax20', 20: 'vs30', 21: 'vu30', 22: 'dir30', 23: 'desU30', 24: 'vmax30', 25: 'tsuelo', 26: 'vw', 27: 'paus'}, inplace=True)
   df['fh'] = pd.to_datetime(df['fh'], infer_datetime_format=True)  #     Convertir datetime.

df

Then I try to use Pandas' gruopby to make the hourly averages:

medias = df.groupby([df.index.year,
                 df.index.month,
                 df.index.hour])['tc10'].mean()

medias.index.names = ["Año", "Mes", "Hora"]

However, I miss this error:

  

'RangeIndex' object has no attribute 'year'.

I would like to know if there is a problem with the frequency of the data, and if there is any way in which you can assign a time average to each hour. Like a for, or some module of numpy or pandas.

Thank you very much in advance.