Train_LSTM_Model

"""# LSTM"""

# fix random seed for reproducibility
tf.random.set_seed(7)

print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)

#Process the data for LSTM
trainX =np.array(X_train)
testX =np.array(X_test)
X_train = trainX.reshape(X_train.shape[0], 1, X_train.shape[1])
X_test = testX.reshape(X_test.shape[0], 1, X_test.shape[1])

trainY =np.array(y_train)

#Building the LSTM Model
model = Sequential()
# 45, 55, 123, 132
model.add(LSTM(units=50, input_shape=(1, trainX.shape[1]), activation='relu', return_sequences=False))
#model.add(LSTM(units=50, dropout=0.2, input_shape=(1, trainX.shape[1]), activation='relu', return_sequences=False))
model.add(Dense(1))
model.compile(loss='mean_squared_error', optimizer='adamax')

history = model.fit(X_train, y_train, epochs=100, batch_size=1)

lstm_prediction = model.predict(X_test)
print(len(lstm_prediction))

MSE = round(mean_squared_error(y_test, lstm_prediction), 9)
RMSE = round(mean_squared_error(y_test, lstm_prediction, squared=False), 9)
MAE = round(mean_absolute_error(y_test, lstm_prediction), 9)

print('MSE: ', MSE)
print('RMSE: ', RMSE)
print('MAE', MAE)

x = np.arange(1)
plt.bar(x-0.2, MSE, width=0.1, color='red')
plt.bar(x, RMSE, width=0.1, color='orange')
plt.bar(x+0.2, MAE, width=0.1, color='blue')

plt.xticks(x, ['MSE', 'RMSE'])
plt.xlabel("Metrics")
plt.ylabel("Scores")
plt.legend(["MSE", "RMSE", "MAE"])
plt.show()