Time Series Forecasting Using Deep Learning Example is wrong

2 次查看（过去 30 天）

Andriy Chubatyy 2020-3-30

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/513900-time-series-forecasting-using-deep-learning-example-is-wrong

评论： Andriy Chubatyy 2020-3-30

I´m new in use of Deep Learning toolbox.

I was try execute the example 'Time Series Forecasting Using Deep Learning'

https://es.mathworks.com/help/deeplearning/ug/time-series-forecasting-using-deep-learning.html;jsessionid=a8465367f3e08aa3e11833e524d7

but it has an error (Matrix dimensions must agree.) in line:

rmse = sqrt(mean((YPred-YTest).^2))

YPred and YTest don´t have the same dimensions, and then if i truncate rmse value, the code has other error (Vectors must be the same length.) in line:

plot(idx,[data(numTimeStepsTrain) YPred],'.-')

So, how i can solve this problem?

Thank you.

Regards.

2 个评论
显示无隐藏无

Mohammad Sami 2020-3-30

Did you accidently change something in the example. did you run from the beginning ? i dont get any error on mine

openExample('nnet/TimeSeriesForecastingUsingDeepLearningExample')

Andriy Chubatyy 2020-3-30

Yes, i changed one think and it is my mistake, in line:

[net,YPred] = predictAndUpdateState(net,YTrain(end)); --> [net,YPred] = predictAndUpdateState(net,YTrain);

because i had error:

The construct "YTrain(...end...)" is ambiguous in this context, because "YTrain" cannot be ascertained to be either the name of a variable or of a function. To make it a variable, assign to it; to allow it to become a function at execution time, replace "end" with a call to LENGTH, SIZE, or NUMEL on the desired array.

So this line: [net,YPred] = predictAndUpdateState(net,YTrain(end)); if i execute it in script editor it give me a error but if i execute it in command window is good.

And yes, i run from the beginning.

Now i executed:

openExample('nnet/TimeSeriesForecastingUsingDeepLearningExample')

step by step and it is working.

thank you Mohammad, regards.

请先登录，再进行评论。

请先登录，再回答此问题。