HVAC_Load_Calc.py

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

import tensorflow as tf
import keras
from keras import models
from keras import layers
from keras.utils import np_utils


# Load of buildings energy data from file. Save ENB2012_data.csv file into local drive and add below the path.
dataset = pd.read_csv('/ ... /ENB2012_data.csv')

print(dataset)

# Separation of training data into X and Y sets
X_train = dataset.ix[0:667,1:9].values.astype('float32')
Y1_train = dataset.loc[0:667,'Y1'].values.astype('float32')
Y2_train = dataset.loc[0:667,'Y2'].values.astype('float32')

# Separation of test data into X and Y sets
X_test = dataset.ix[668:767,1:9].values.astype('float32')
Y1_test = dataset.loc[668:767,'Y1'].values.astype('float32')
Y2_test = dataset.loc[668:767,'Y2'].values.astype('float32')


# Training and test input data Normalization
mean = X_train.mean(axis=0)
X_train -= mean
std = X_train.std(axis=0)
X_train /= std

X_test -= mean
X_test /= std


# NN model definition
def build_model():
    # Train of the DL network using keras
    model =models.Sequential()
    # The Input Layer:
    model.add(layers.Dense(64, input_dim=X_train.shape[1], activation='relu'))
    # The Hidden Layers
    model.add(layers.Dense(64,activation='relu'))
    # The Output Layer
    model.add(layers.Dense(1))

    model.compile(optimizer='rmsprop', loss='mse', metrics=['mae'])
    return model

model = build_model()
model.fit(X_train, Y1_train, epochs=300, batch_size=10, verbose=0)
test_mse_score, test_mae_score = model.evaluate(X_test, Y1_test)