# Does Rasa need to retrain to learn values of a lookup table?

**URL:** <https://forum.rasa.com/t/does-rasa-need-to-retrain-to-learn-values-of-a-lookup-table/19512>\
**Category:** Rasa Open Source\
**Created:** [October 9, 2019, 12:12pm UTC](https://forum.rasa.com/t/does-rasa-need-to-retrain-to-learn-values-of-a-lookup-table/19512 "2019-10-09T12:12:13Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![ice2tricky](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/ice2tricky/32/4361_2.png) [@ice2tricky](https://forum.rasa.com/u/ice2tricky)\
**Post date:** [October 9, 2019, 12:12pm UTC](https://forum.rasa.com/t/does-rasa-need-to-retrain-to-learn-values-of-a-lookup-table/19512/1 "2019-10-09T12:12:13Z")

</div>

Does anyone know if you need to retrain Rasa when you add a lookup tables and do you only have to make sure that there are enough examples in your NLU training data to make them work?

Or do i need to retrain everything, because it doesn’t retrain if a lookup table has more values?

---

<div class="post-metadata">

**Author:** ![j.mosig](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/j.mosig/32/3588_2.png) [@j.mosig](https://forum.rasa.com/u/j.mosig)\
**Post date:** [October 9, 2019, 12:32pm UTC](https://forum.rasa.com/t/does-rasa-need-to-retrain-to-learn-values-of-a-lookup-table/19512/2 "2019-10-09T12:32:30Z")

</div>

You definitely don’t need to retrain core. Probably also not NLU, because as far as I know a lookup table is just reduced to a regex.

---

<div class="post-metadata">

**Author:** ![IgNoRaNt23](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/ignorant23/32/8029_2.png) [@IgNoRaNt23](https://forum.rasa.com/u/IgNoRaNt23)\
**Post date:** [October 9, 2019, 12:37pm UTC](https://forum.rasa.com/t/does-rasa-need-to-retrain-to-learn-values-of-a-lookup-table/19512/3 "2019-10-09T12:37:42Z")

</div>

You need to retrain NLU. Rasa models are meant to work just by themselves, so all the lookup table data is saved in the model files.

---

<div class="post-metadata">

**Author:** ![IgNoRaNt23](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/ignorant23/32/8029_2.png) [@IgNoRaNt23](https://forum.rasa.com/u/IgNoRaNt23)\
**Post date:** [October 9, 2019, 12:39pm UTC](https://forum.rasa.com/t/does-rasa-need-to-retrain-to-learn-values-of-a-lookup-table/19512/4 "2019-10-09T12:39:50Z")

</div>

Once your lookup tables got a certain size, youll even notice that the training process takes some time for the RegexFeaturizer

---

<div class="post-metadata">

**Author:** ![ice2tricky](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/ice2tricky/32/4361_2.png) [@ice2tricky](https://forum.rasa.com/u/ice2tricky)\
**Post date:** [October 9, 2019, 12:47pm UTC](https://forum.rasa.com/t/does-rasa-need-to-retrain-to-learn-values-of-a-lookup-table/19512/6 "2019-10-09T12:47:13Z")

</div>

Is it possible that you can learn the bot to extract entities while it is running (and to not bring your bot offline for retrain?

---

<div class="post-metadata">

**Author:** ![IgNoRaNt23](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/ignorant23/32/8029_2.png) [@IgNoRaNt23](https://forum.rasa.com/u/IgNoRaNt23)\
**Post date:** [October 9, 2019, 1:19pm UTC](https://forum.rasa.com/t/does-rasa-need-to-retrain-to-learn-values-of-a-lookup-table/19512/7 "2019-10-09T13:19:23Z")

</div>

Yes, you can do this in the Rasa X UI or via the API for example

> **[Rasa & Rasa Pro Documentation](https://rasa.com/docs/rasa/pages/http-api/)**