# T5 Model for Recipe generation

**URL:** <https://discuss.huggingface.co/t/t5-model-for-recipe-generation/14351>\
**Category:** 🤗Transformers\
**Created:** [February 2, 2022, 9:11pm UTC](https://discuss.huggingface.co/t/t5-model-for-recipe-generation/14351 "2022-02-02T21:11:22Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![kokos](https://avatars.discourse-cdn.com/v4/letter/k/3ec8ea/32.png) [@kokos](https://discuss.huggingface.co/u/kokos)\
**Post date:** [February 2, 2022, 9:11pm UTC](https://discuss.huggingface.co/t/t5-model-for-recipe-generation/14351/1 "2022-02-02T21:11:22Z")

</div>

I’d like to use the t5 model to generate recipes, the dataset has a description for every recipe, but I don’t know what should I put as an input for the network since I’d like to just generate a random unique recipe when I put a text like ‘give me a recipe’ as my input!  
I thought of using ‘give me a recipe’ as my input when training for all dataset and output would be the actual recipe, but then how can the model learn the difference and optimize the model!