# Training on Reference images

**URL:** <https://community.drivendata.org/t/training-on-reference-images/6367>\
**Category:** Image Similarity Challenge\
**Created:** [August 4, 2021, 4:27pm UTC](https://community.drivendata.org/t/training-on-reference-images/6367 "2021-08-04T16:27:27Z")\
**Posts on this page:** 1\
**Showing post:** 5

<div class="post-metadata">

**Author:** ![wenhaowang](https://avatars.discourse-cdn.com/v4/letter/w/b2d939/32.png) [@wenhaowang](https://community.drivendata.org/u/wenhaowang)\
**Post date:** [August 17, 2021, 1:33pm UTC](https://community.drivendata.org/t/training-on-reference-images/6367/5 "2021-08-17T13:33:18Z")

</div>

I do NOT understand why you argue that training on the reference dataset is not prohibited. In the early reply, [Is legal for training on the reference dataset?](http://community.drivendata.org/t/is-legal-for-training-on-the-reference-dataset/6293) and the rule [Competition: Facebook AI Image Similarity Challenge: Matching Track](https://www.drivendata.org/competitions/79/competition-image-similarity-1-dev/page/376/#datause) have already pointed out "Use of augmented reference images for any other reason, including model training, is **prohibited**".

---

_[View the full topic](https://community.drivendata.org/t/training-on-reference-images/6367)._