pydlm包“步骤超出范围”

Question

只有当我使用基本功能时，动态模型才起作用。在添加动态组件“升级”来训练模型并使用预测功能进行预测后，预测行“result=myDLM.predictN（N=48，….}）”开始给我一个错误“步骤超出范围”

myDLM = dlm(X_train['price'].tolist())
myDLM = myDLM + trend(degree=1, discount=0.95, name='lineTrend')
myDLM = myDLM + seasonality(period=7, discount=0.99, name='week')
myDLM = myDLM + autoReg(degree=2, data=y_train['price'], name='ar2', w=1.0)

myDLM = myDLM + dynamic(features=pd.DataFrame(data_train['promotion']).values.tolist(), discount=0.95, name='train_promotion')

myDLM.fit()

result = myDLM.predictN(N=48, featureDict={'predict_Promo':data_predict['promotion']})