# Past dynamic features in Time Series Transformer

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fpast-dynamic-features-in-time-series-transformer%2F98812%3E%5C
**Category:** 🤗Transformers\
**Created:** [July 22, 2024, 7:05pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fpast-dynamic-features-in-time-series-transformer%2F98812 "2024-07-22T19:05:36Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![javiermsm12](https://avatars.discourse-cdn.com/v4/letter/j/53a042/32.png) [@javiermsm12](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fjaviermsm12)%5C
**Post date:** [July 22, 2024, 7:05pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fpast-dynamic-features-in-time-series-transformer%2F98812%2F1 "2024-07-22T19:05:36Z")

</div>

Can someone provide some example using past dynamic features?

I’ve been following the blog below and now I want to implement a similar model but using past dynamic features in addition to future dynamic features. It seems that the model in the blog treats all covariates as features known in advance.

> **[Probabilistic Time Series Forecasting with 🤗 Transformers](https://huggingface.co/blog/time-series-transformers)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Thanks!  
Javier