# PatchTST issue adding independent variables to the model that help predict one target variable

**URL:** <https://discuss.huggingface.co/t/patchtst-issue-adding-independent-variables-to-the-model-that-help-predict-one-target-variable/78494>\
**Category:** 🤗Transformers\
**Created:** [March 21, 2024, 3:48pm UTC](https://discuss.huggingface.co/t/patchtst-issue-adding-independent-variables-to-the-model-that-help-predict-one-target-variable/78494 "2024-03-21T15:48:33Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![CoCoNuTeK](https://avatars.discourse-cdn.com/v4/letter/c/e495f1/32.png) [@CoCoNuTeK](https://discuss.huggingface.co/u/CoCoNuTeK)\
**Post date:** [March 21, 2024, 3:48pm UTC](https://discuss.huggingface.co/t/patchtst-issue-adding-independent-variables-to-the-model-that-help-predict-one-target-variable/78494/1 "2024-03-21T15:48:33Z")

</div>

Hello guys, i would like to ask is it possible when using the PatchTST for time series prediction, if i have independent features can i include them somewhere without making the model think that they are also target variables (making the model predict multivariate time series)? With Autoformer/Informer it was solved using the **past\_time\_features** , because from the docs it says you can only include **past\_values** and **target\_values** for any of the heads available.
