Загрузить данные из Excel в dataset keras

from keras.datasets import boston_housing
from keras.models import Sequential
from keras.layers import Dense
import openpyxl
import numpy as np
 
#Для начала загрузим библиотеку и выберем нужную книгу и рабочий лист:
wb = openpyxl.load_workbook(filename = 'E:/Fast_Water_project/Gidrolodge_.xlsx')
sheet_oren = wb['abc']
sheet_orsk = wb['dsd']
sheet_ilek = wb['hgt']
#считываем значение определенной ячейки
#val = sheet_oren['A1'].value
 
#считываем заданный диапазон
factor_days = [v[0].value for v in sheet_oren['E2':'E2920']]
factor_osadky = [v[0].value for v in sheet_oren['F2':'F2920']]
factor_level_snow = [v[0].value for v in sheet_oren['G2':'G2920']]
factor_chandge_lvl_snow = [v[0].value for v in sheet_oren['H2':'H2920']]
factor_reserv_water = [v[0].value for v in sheet_oren['I2':'I2920']]
factor_frez_ground = [v[0].value for v in sheet_oren['J2':'J2920']]
factor_ice = [v[0].value for v in sheet_oren['K2':'K2920']]
factor_max_t = [v[0].value for v in sheet_oren['L2':'L2920']]
factor_min_t = [v[0].value for v in sheet_oren['M2':'M2920']]
factor_humidity = [v[0].value for v in sheet_oren['N2':'N2920']]
factor_deficit_humidity = [v[0].value for v in sheet_oren['O2':'O2920']]
factor_solar = [v[0].value for v in sheet_oren['P2':'P2920']]
factor_level_water = [v[0].value for v in sheet_oren['Q2':'Q2920']]
resolve_chandge_lvl_water = [v[0].value for v in sheet_oren['R2':'R2920']]
 
#factors_resolve=[factor_days,factor_osadky,factor_level_snow,factor_chandge_lvl_snow,factor_reserv_water,factor_frez_ground,factor_ice
#                    ,factor_max_t,factor_min_t,factor_humidity,factor_deficit_humidity,factor_solar,factor_level_water,resolve_chandge_lvl_water]
 
factors_list=list()
resolve_list=list()
factors_list.append(factor_days)
factors_list.append(factor_osadky)
factors_list.append(factor_level_snow)
factors_list.append(factor_chandge_lvl_snow)
factors_list.append(factor_reserv_water)
factors_list.append(factor_frez_ground)
factors_list.append(factor_ice)
factors_list.append(factor_max_t)
factors_list.append(factor_min_t)
factors_list.append(factor_humidity)
factors_list.append(factor_deficit_humidity)
factors_list.append(factor_solar)
factors_list.append(factor_level_water)
resolve_list.append(resolve_chandge_lvl_water)
 
#factors_resolve=np.hstack([factor_days],[factor_osadky],[factor_level_snow],[factor_chandge_lvl_snow],[factor_reserv_water],[factor_frez_ground],[factor_ice],[factor_max_t],[factor_min_t],[factor_humidity],[factor_deficit_humidity],[factor_solar],[factor_level_water])
factors=np.array(factors_list)
resolve=np.array(resolve_list)
 
(x_train, y_train), (x_test, y_test) = boston_housing.load_data(factors,resolve)
 
#Нормализация данных
#Вычитаем среднее значение и делим на стандартное отклонение
# Среднее значение
 mean = x_train.mean(axis=0)
# Стандартное отклонение
 std = x_train.std(axis=0)
 x_train -= mean
 x_train /= std
 x_test -= mean
 x_test /= std
 
print(factors)
print()
print(resolve)