# Question Answering no-context

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fquestion-answering-no-context%2F83148%3E%5C
**Category:** Beginners\
**Created:** [April 22, 2024, 5:31pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fquestion-answering-no-context%2F83148 "2024-04-22T17:31:34Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![rasdlasdl](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/rasdlasdl/32/26099_2.png) [@rasdlasdl](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Frasdlasdl)%5C
**Post date:** [April 22, 2024, 5:31pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fquestion-answering-no-context%2F83148%2F1 "2024-04-22T17:31:34Z")

</div>

Hello everyone,  
I have dataset of medicine questions it looks like [{“question”:question, “answer”:answer}, {“question”:question, “answer”:answer}, …]  
I neeed to train a model which would give me answer based on my question  
It should be something like FAQ

As input I will provide new question and model should give me one of the answers in dataset

I tried to find something similar but only thing I could find was QA with context and Text Classsification Positive or Negative(

Any thoughts?

Thanks

---

<div class="post-metadata">

**Author:** ![Namitg02](https://avatars.discourse-cdn.com/v4/letter/n/e8c25b/32.png) [@Namitg02](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FNamitg02)%5C
**Post date:** [May 22, 2024, 4:55pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fquestion-answering-no-context%2F83148%2F2 "2024-05-22T16:55:34Z")

</div>

Hello,

You can probably consider a RAG approach to do this.  
Thanks

---

<div class="post-metadata">

**Author:** ![vpkprasanna](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/vpkprasanna/32/26436_2.png) [@vpkprasanna](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fvpkprasanna)%5C
**Post date:** [May 22, 2024, 5:09pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fquestion-answering-no-context%2F83148%2F3 "2024-05-22T17:09:08Z")

</div>

You can either convert this into csv or a table structure use a [TAPAS](https://huggingface.co/docs/transformers/en/model_doc/tapas) model from google to get the answer or use LLM to get answer using the technique called RAG
