# Training HF transformer models on custom (not text) data

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftraining-hf-transformer-models-on-custom-not-text-data%2F41064%3E%5C
**Category:** 🤗Transformers\
**Created:** [May 26, 2023, 9:51am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftraining-hf-transformer-models-on-custom-not-text-data%2F41064 "2023-05-26T09:51:33Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Malik7115](https://avatars.discourse-cdn.com/v4/letter/m/c2a13f/32.png) [@Malik7115](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FMalik7115)%5C
**Post date:** [May 26, 2023, 9:51am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftraining-hf-transformer-models-on-custom-not-text-data%2F41064%2F1 "2023-05-26T09:51:33Z")

</div>

Hi,  
is there a way to train data not in text form using the hf models? For example, if I want to train a transformer on some representation of images and use it to generate new images, how exactly would I accomplish this?