targets[14] - 0.07
) / 7 #normalising output/target values into range [0,0.99] approx
n_potential_nrg = (targets[15] - 0.07) / 7
n_current_co2 = targets[16] / 19
n_potential_co2 = targets[17] / 19
n_current_light = targets[18] / 165
n_potential_light = targets[19] / 165
n_current_hhw = targets[20] / 1656
n_potential_hhw = targets[21] / 1656
NN1.train(normal_inputs, n_current_nrg)
NN2.train(normal_inputs, n_potential_nrg)
NN3.train(normal_inputs, n_current_co2)
NN4.train(normal_inputs, n_potential_co2)
NN5.train(normal_inputs, n_current_light)
NN6.train(normal_inputs, n_potential_light)
NN7.train(normal_inputs, n_current_hhw)
NN8.train(normal_inputs, n_potential_hhw)
pass
pass
pass
NN1.save('a6', 'b6')
NN2.save('c6', 'd6')
NN3.save('e6', 'f6')
NN4.save('g6', 'h6')
NN5.save('i6', 'j6')
NN6.save('k6', 'l6')
NN7.save('m6', 'n6')
NN8.save('o6', 'p6')
print('The neural networks have successfully been trained.')
targets[11] - 0.07
) / 7 #normalising output/target values into range [0,0.99] approx
n_potential_nrg = (targets[12] - 0.07) / 7
n_current_co2 = targets[13] / 19
n_potential_co2 = targets[14] / 19
n_current_light = targets[15] / 165
n_potential_light = targets[16] / 165
n_current_hhw = targets[17] / 1656
n_potential_hhw = targets[18] / 1656
NN1.train(normal_inputs, n_current_nrg)
NN2.train(normal_inputs, n_potential_nrg)
NN3.train(normal_inputs, n_current_co2)
NN4.train(normal_inputs, n_potential_co2)
NN5.train(normal_inputs, n_current_light)
NN6.train(normal_inputs, n_potential_light)
NN7.train(normal_inputs, n_current_hhw)
NN8.train(normal_inputs, n_potential_hhw)
pass
pass
pass
NN1.save('a7', 'b7')
NN2.save('c7', 'd7')
NN3.save('e7', 'f7')
NN4.save('g7', 'h7')
NN5.save('i7', 'j7')
NN6.save('k7', 'l7')
NN7.save('m7', 'n7')
NN8.save('o7', 'p7')
print('The neural networks have successfully been trained.')
