# Is there a way to choose the order of property vectorization using text2vec-\* modules?

**URL:** <https://forum.weaviate.io/t/is-there-a-way-to-choose-the-order-of-property-vectorization-using-text2vec-modules/596>\
**Category:** General\
**Created:** [August 30, 2023, 3:10pm UTC](https://forum.weaviate.io/t/is-there-a-way-to-choose-the-order-of-property-vectorization-using-text2vec-modules/596 "2023-08-30T15:10:51Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![junbetterway](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.weaviate.io/junbetterway/32/174_2.png) [@junbetterway](https://forum.weaviate.io/u/junbetterway)\
**Post date:** [August 30, 2023, 3:10pm UTC](https://forum.weaviate.io/t/is-there-a-way-to-choose-the-order-of-property-vectorization-using-text2vec-modules/596/1 "2023-08-30T15:10:51Z")

</div>

Based from here: [Vectorizers and Rerankers | Weaviate - vector database](https://weaviate.io/developers/weaviate/modules/retriever-vectorizer-modules)

Unless specified otherwise in the schema, the default behavior is to:

- Only vectorize properties that use the `text` data type (unless [skip](https://weaviate.io/developers/weaviate/configuration/schema-configuration#property-level-module-settings)ped)
- Sort properties in alphabetical (a-z) order before concatenating values
- … so on

Is there a way to sort the properties based on our pre-defined order when vectorizing?

The reason behind is that for example, we are using Huggingface **multi-qa-MiniLM-L6-cos-v1** which has the following [note](https://huggingface.co/sentence-transformers/multi-qa-MiniLM-L6-cos-v1#intended-uses):

> Note that there is a limit of 512 word pieces: Text longer than that will be truncated. Further note that the model was just trained on input text up to 250 word pieces. It might not work well for longer text.

So we would want to prioritize the some fields before truncation in case we hit the limit. Thanks!

---

<div class="post-metadata">

**Author:** ![DudaNogueira](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.weaviate.io/dudanogueira/32/7846_2.png) [@DudaNogueira](https://forum.weaviate.io/u/DudaNogueira)\
**Post date:** [August 31, 2023, 11:31am UTC](https://forum.weaviate.io/t/is-there-a-way-to-choose-the-order-of-property-vectorization-using-text2vec-modules/596/2 "2023-08-31T11:31:10Z")

</div>

Hi!

There is no way to select this order. What you can do is to intentionally name your properties so they fall in the order you want.

Please, feel free to [open a new feature request in our Github](https://github.com/weaviate/weaviate/issues/new) 🙂

Thanks!

---

<div class="post-metadata">

**Author:** ![junbetterway](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.weaviate.io/junbetterway/32/174_2.png) [@junbetterway](https://forum.weaviate.io/u/junbetterway)\
**Post date:** [August 31, 2023, 1:10pm UTC](https://forum.weaviate.io/t/is-there-a-way-to-choose-the-order-of-property-vectorization-using-text2vec-modules/596/3 "2023-08-31T13:10:03Z")

</div>

Thanks for confirming. As suggested, I have filed a feature request here:

> <https://github.com/weaviate/weaviate/issues/3487>
>
> Based from this \[documentation\](https://weaviate.io/developers/weaviate/modules/…retriever-vectorizer-modules#vectorization-with-text2vec--modules
> ):
> 
> Unless specified otherwise in the schema, the default behavior is to:
> 
> \- Only vectorize properties that use the text data type (unless skip)
> \*\*- Sort properties in alphabetical (a-z) order before concatenating values\*\*
> 
> Is there a way to sort the properties based on our pre-defined order when vectorizing?
> 
> The reason behind is that for example, we are using Huggingface multi-qa-MiniLM-L6-cos-v1 which has the following \[note\](https://huggingface.co/sentence-transformers/multi-qa-MiniLM-L6-cos-v1#intended-uses):
> 
> \*\*\_Note that there is a limit of 512 word pieces: Text longer than that will be truncated. Further note that the model was just trained on input text up to 250 word pieces. It might not work well for longer text.\_\*\*
> 
> So we would want to prioritize the some fields before truncation in case we hit the limit if possible. Thanks!