Error about the data type

[chenghan710]

I'm trying to use CEEMDAN_LSTM to predict time series

the code report the AttributeError: Can only use .str accessor with string values!

How can I fix this problem

And one more question, can I predict more than one out-of-sample value by CEEMDAN_LSTM ?

[FateMurphy]

(1) For problem 1: Can only use .str accessor with string values! A: you can try to set the encoding='utf-8' (2) Predict more than one out-of-sample value by CEEMDAN_LSTM A: set DAY_AHEAD=2 or larger, 1 means one-day-ahead

[chenghan710]

Thanks for replying, the first question was solved, for the problem(2), the two output is I set DAY_AHEAD as 2, 3 It seems only predict for one day, here is my code Is there something wrong?

[FateMurphy]

For multi-step prediction, you can try the rolling forecasting method.

[chenghan710]

I tried the rolling forecasting method, the error was reported I've update pandas to the newest version

[FateMurphy]

You can try to set REDECOM_LIST = None firstly. If the error still exists, please let me know and I will fix this error later.

[chenghan710]

The error still exist

[chenghan710]

Sorry, here is another question

I used multiple_keras_predict and set run time=20, and NEXT_DAY=TRUE

Is this mean that it will predict the value of 20 days in the future?

e.g. I run this code at 9/1, it will predict the value until 9/21?

[FateMurphy]

It means that the code forecast the last data point, eg. 9/1, 20 times by respective method.

[chenghan710]

So if I want to predict the value of 20 days in the future, it can only use rolling forecasting method?

[FateMurphy]

In this code, yes. In fact, the rolling forecasting method means running one-day-ahead forecasting multiple times with changing the dataset. Or, you can find other codes directly to make sequence forecasting.

[chenghan710]

It is OK for me to running one-day-ahead forecasting multiple times with changing the dataset, but I still have the same error for running rolling forecasting method