# What events count as feature for training Rasa core

**URL:** <https://forum.rasa.com/t/what-events-count-as-feature-for-training-rasa-core/19781>\
**Category:** Rasa Open Source\
**Created:** [October 14, 2019, 6:55pm UTC](https://forum.rasa.com/t/what-events-count-as-feature-for-training-rasa-core/19781 "2019-10-14T18:55:29Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![nbeuchat](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/nbeuchat/32/13688_2.png) [@nbeuchat](https://forum.rasa.com/u/nbeuchat)\
**Post date:** [October 14, 2019, 6:55pm UTC](https://forum.rasa.com/t/what-events-count-as-feature-for-training-rasa-core/19781/1 "2019-10-14T18:55:29Z")

</div>

Hi there!

I am trying to better understand the internal mechanics of Rasa core. In the `KerasPolicy` for example, what is considered a feature? What will be part of the `max_history`?

My understanding is that we will have at least the following as features:

- intents
- featurized slots values
- entities (?)

But how about actions, utterances, and forms? Or `form{"name": "xxx"}`? And the `requested_slot`? Will they be used to predict the next actions?

The reason I am asking is that we have quite a few notifications (ie: actions that we execute through the API instead of predicting them) and we want to better control what will happen after they are triggered. Also, I’d just like to better understand what is going on in general.

Thanks a lot for the info! Cheers, Nicolas

---

<div class="post-metadata">

**Author:** ![dakshvar22](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/dakshvar22/32/3993_2.png) [@dakshvar22](https://forum.rasa.com/u/dakshvar22)\
**Post date:** [October 15, 2019, 1:21pm UTC](https://forum.rasa.com/t/what-events-count-as-feature-for-training-rasa-core/19781/2 "2019-10-15T13:21:15Z")

</div>

@nbeuchat You can read more about the featurization for core policies here - [Featurization](https://rasa.com/docs/rasa/api/featurization/#featurization) Both KerasPolicy and EmbeddingPolicies use the featurization methods described above. Architectures for both these policies can be found here - [Policies](https://rasa.com/docs/rasa/core/policies/#keras-policy)

To be precise, yes previous actions and active form is part of the feature bag that’s passed to the model. Actual text of utterances is not part of the featurization.

---

<div class="post-metadata">

**Author:** ![nbeuchat](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/nbeuchat/32/13688_2.png) [@nbeuchat](https://forum.rasa.com/u/nbeuchat)\
**Post date:** [October 17, 2019, 12:09pm UTC](https://forum.rasa.com/t/what-events-count-as-feature-for-training-rasa-core/19781/3 "2019-10-17T12:09:25Z")

</div>

Hi @dakshvar22, thanks a lot for your response, this is very helpful in better understanding what is going when training, especially with manually crafted stories.