search

maxjcohen / transformer

Implementation of Transformer model (originally from Attention is All You Need) applied to Time Series.
https://timeseriestransformer.readthedocs.io/en/latest/
GNU General Public License v3.0
842 stars 165 forks source link

Can time series A be used to predict time series B？ #53

Closed chuzheng88 closed 8 months ago

chuzheng88 commented 2 years ago

I have read the docs of this github project. In some notebook, I found that time series A was used to predict time series B. I want to know that the Transformer proposed in this project could use time series A to predict time series B.

maxjcohen commented 2 years ago

Hi, yes it can. It actually for this kind of application this model was designed.

chuzheng88 commented 2 years ago

Oh, I'm so sorry, my question was writen wrong. The right question is described as follows:

In some notebook, I found that time series A was used to predict the same time series A. I want to know that the Transformer proposed in this project could use time series A to predict time series B.

In addition, I want to use your project code to deal with a regression problem, and it is described as follows: X = [ [[time series sequence 1]], [[time series sequence 2]],

[[time series sequence s]], ] and y = [ [[label 1]], [[label 2]],

[[label s]], ] where sequences are not the same length. So I want to just use your model definition (transformer.py) and paddding all sequences to the same length in my dataset before input the model.

Is there any thing I need to pay attention to in this job ? Or do you have other suggestions ?

Thansk.

YangJoy750 commented 2 years ago

are you solve the problem?

chuzheng88 commented 2 years ago

are you solve the problem?

No