energy consumption.py

from keras.models import Sequential
from keras.layers import LSTM
from keras.layers import Dropout
from keras.layers import Dense
from sklearn.metrics import mean_squared_error,r2_score
import matplotlib.pyplot as plt
import numpy as np

model = Sequential()
model.add(LSTM(100, input_shape=(train_x.shape[1], train_x.shape[2])))
model.add(Dropout(0.1))
model.add(Dense(1))
model.compile(loss='mean_squared_error', optimizer='adam')

# Network fitting
history = model.fit(train_x, train_y, epochs=50, batch_size=70, validation_data=(test_x, test_y), verbose=2, shuffle=False)

# Loss history plot
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper right')
plt.show()

size = df_resample.shape[1]

# Prediction test
yhat = model.predict(test_x)
test_x = test_x.reshape((test_x.shape[0], size))

# invert scaling for prediction
inv_yhat = np.concatenate((yhat, test_x[:, 1-size:]), axis=1)
inv_yhat = scaler.inverse_transform(inv_yhat)
inv_yhat = inv_yhat[:,0]

# invert scaling for actual
test_y = test_y.reshape((len(test_y), 1))
inv_y = np.concatenate((test_y, test_x[:, 1-size:]), axis=1)
inv_y = scaler.inverse_transform(inv_y)
inv_y = inv_y[:,0]

# calculate RMSE
rmse = np.sqrt(mean_squared_error(inv_y, inv_yhat))
print('Test RMSE: %.3f' % rmse)

aa=[x for x in range(500)]
plt.figure(figsize=(25,10)) 
plt.plot(aa, inv_y[:500], marker='.', label="actual")
plt.plot(aa, inv_yhat[:500], 'r', label="prediction")
plt.ylabel(df.columns[0], size=15)
plt.xlabel('Time step for first 500 hours', size=15)
plt.legend(fontsize=15)
plt.show()