# Generative Adversarial Network (GAN) for graph generation

**URL:** <https://community.neo4j.com/t/generative-adversarial-network-gan-for-graph-generation/68591>\
**Category:** Graph Data Science / Graph Analytics\
**Tags:** ai\
**Created:** [June 27, 2024, 2:58pm UTC](https://community.neo4j.com/t/generative-adversarial-network-gan-for-graph-generation/68591 "2024-06-27T14:58:58Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![mmelek](https://avatars.discourse-cdn.com/v4/letter/m/838e76/32.png) [@mmelek](https://community.neo4j.com/u/mmelek)\
**Post date:** [June 27, 2024, 2:58pm UTC](https://community.neo4j.com/t/generative-adversarial-network-gan-for-graph-generation/68591/1 "2024-06-27T14:58:58Z")

</div>

Hi everyone,  
I am new to neo4j data science. I have use case where I am looking for ways to generate new graphs based on a dataset of graphs. I found a post about drug compound generation using GANs ([https://community.neo4j.com/t/how-to-find-similarity-between-two-graphs-in-neo4j/](https://community.neo4j.com/t/how-to-find-similarity-between-two-graphs-in-neo4j/)). Is there a way to train and deploy a GAN model using neo4j data science platform? Has anyone come across similar examples? Thank you.

Murat

---

<div class="post-metadata">

**Author:** ![adam\_schill-col](https://sea1.discourse-cdn.com/flex021/user_avatar/community.neo4j.com/adam_schill-col/32/19938_2.png) [@adam\_schill-col](https://community.neo4j.com/u/adam_schill-col)\
**Post date:** [July 3, 2024, 12:40pm UTC](https://community.neo4j.com/t/generative-adversarial-network-gan-for-graph-generation/68591/2 "2024-07-03T12:40:39Z")

</div>

Hi @mmelek,

Unfortunately there is no way to train and deploy a GAN model with the Neo4j Graph Data Science library at this time. But we shall make note of your request.

Regards,  
Adam