# Concept To text

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fconcept-to-text%2F16377%3E%5C
**Category:** Languages at Hugging Face\
**Created:** [April 1, 2022, 4:50pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fconcept-to-text%2F16377 "2022-04-01T16:50:14Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![shaoniana1997](https://avatars.discourse-cdn.com/v4/letter/s/b4bc9f/32.png) [@shaoniana1997](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fshaoniana1997)%5C
**Post date:** [April 1, 2022, 4:50pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fconcept-to-text%2F16377%2F1 "2022-04-01T16:50:14Z")

</div>

Hello, if i want to train a model while the input is serval concepts like “today family happy mountain”, and i want the output is a sentence based on the serval concepts like" My fanily go to mountain and we have a happy game".  
can you tell me what i should start?  
Thank you very much

---

<div class="post-metadata">

**Author:** ![NimaBoscarino](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/nimaboscarino/32/4358_2.png) [@NimaBoscarino](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FNimaBoscarino)%5C
**Post date:** [April 1, 2022, 10:57pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fconcept-to-text%2F16377%2F2 "2022-04-01T22:57:12Z")

</div>

Welcome! For something like that, I think you could start with looking into “text generation with keywords”. This [Medium post could be useful for you](https://medium.com/mlearning-ai/generating-sentences-from-keywords-using-transformers-in-nlp-e89f4de5cf6b), as well as the related [GitHub repository here](https://github.com/gagan3012/keytotext). They even have some [models available on the Hugging Face Hub](https://huggingface.co/models?filter=keytotext)! You may be able to follow their method to finetune a model ([looking at their training code, they originally finetune the t5-small model](https://github.com/gagan3012/keytotext/blob/master/notebooks/Trainer.ipynb)) to make a model that fits your use-case.