# Interactive learning : load pretrained model

**URL:** <https://forum.rasa.com/t/interactive-learning-load-pretrained-model/6075>\
**Category:** Rasa Open Source\
**Created:** [March 15, 2019, 1:34pm UTC](https://forum.rasa.com/t/interactive-learning-load-pretrained-model/6075 "2019-03-15T13:34:29Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![anojoual](https://avatars.discourse-cdn.com/v4/letter/a/74df32/32.png) [@anojoual](https://forum.rasa.com/u/anojoual)\
**Post date:** [March 15, 2019, 1:34pm UTC](https://forum.rasa.com/t/interactive-learning-load-pretrained-model/6075/1 "2019-03-15T13:34:29Z")

</div>

Hello ! i have a question about interactive learning , is there any chance to avoid ‘retrain’ the model from the beginning during interactive learning and just load a pretrained dialogue model that we previously trained on data stories . I think it’s complecated since the online\_train function defined The agent from scratch and call the train function . Thank you

---

<div class="post-metadata">

**Author:** ![lgrinberg](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/lgrinberg/32/1659_2.png) [@lgrinberg](https://forum.rasa.com/u/lgrinberg)\
**Post date:** [March 17, 2019, 1:18am UTC](https://forum.rasa.com/t/interactive-learning-load-pretrained-model/6075/2 "2019-03-17T01:18:03Z")

</div>

If you already have a trained model and want to skip that step just do this: python -m rasa\_core.train   
interactive --core models/dialogue   
–nlu models/current/nlu   
–endpoints endpoints.yml

As described over here:

> **[Writing Conversation Data](https://rasa.com/docs/rasa/writing-stories/#using-interactive-learning)**
