# Embedding sequence layer with LSTM

**URL:** <https://community.konduit.ai/t/embedding-sequence-layer-with-lstm/1404>\
**Category:** DL4J\
**Created:** [May 26, 2021, 8:18am UTC](https://community.konduit.ai/t/embedding-sequence-layer-with-lstm/1404 "2021-05-26T08:18:21Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![raviteja777](https://yyz1.discourse-cdn.com/flex035/user_avatar/community.konduit.ai/raviteja777/32/269_2.png) [@raviteja777](https://community.konduit.ai/u/raviteja777)\
**Post date:** [May 26, 2021, 8:18am UTC](https://community.konduit.ai/t/embedding-sequence-layer-with-lstm/1404/1 "2021-05-26T08:18:21Z")

</div>

I am trying char modelling in LSTM s and referring to the dl4j example in github -

> <https://github.com/eclipse/deeplearning4j-examples/blob/master/dl4j-examples/src/main/java/org/deeplearning4j/examples/advanced/modelling/charmodelling/generatetext/GenerateTxtModel.java>

I would like to know if it is desirable to provide input to an embedding sequence layer , instead of directly providing to LSTM … , if possible how to provide the time dimension in the embedding ? (any example for the same is appreciated )

---

<div class="post-metadata">

**Author:** ![jijiji](https://yyz1.discourse-cdn.com/flex035/user_avatar/community.konduit.ai/jijiji/32/1420_2.png) [@jijiji](https://community.konduit.ai/u/jijiji)\
**Post date:** [May 27, 2021, 10:55am UTC](https://community.konduit.ai/t/embedding-sequence-layer-with-lstm/1404/2 "2021-05-27T10:55:14Z")

</div>

Check input shape description. It might contain data you’re looking for depend on a layer type.