未来的时间序列预测

Question

#Keras LSTM model 
n_batch=1
model = Sequential()
model.add(LSTM(2, batch_input_shape=(1, None,1),activation='sigmoid',return_sequences=True, stateful=True))
model.add(LSTM(2,return_sequences=False,stateful=True))
model.add(Dense(1))
model.compile(loss='mse', optimizer = 'adam')

model.fit(Xtrain,Ytrain, batch_size=n_batch, epochs=3, verbose=1)
model.reset_states()

#model accuracy   
predictions = model.predict(Xtrain,batch_size=n_batch)

plt.figure(figsize=(30,10))
plt.plot(predictions)
plt.plot(Ytrain)
plt.show()

这是我的时间序列分析。此模型正在预测下一步。但我想计算到下一个36步。我该怎么办？