# LayoutLM for table detection and extraction

**URL:** <https://discuss.huggingface.co/t/layoutlm-for-table-detection-and-extraction/7015>\
**Category:** Beginners\
**Created:** [June 20, 2021, 5:41pm UTC](https://discuss.huggingface.co/t/layoutlm-for-table-detection-and-extraction/7015 "2021-06-20T17:41:45Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![ujjayants](https://avatars.discourse-cdn.com/v4/letter/u/aeb1de/32.png) [@ujjayants](https://discuss.huggingface.co/u/ujjayants)\
**Post date:** [June 20, 2021, 5:41pm UTC](https://discuss.huggingface.co/t/layoutlm-for-table-detection-and-extraction/7015/1 "2021-06-20T17:41:45Z")

</div>

Can the LayoutLM model be used or tuned for table detection and extraction?  
The paper says that it works on forms, receipts and for document classification tasks.

---

<div class="post-metadata">

**Author:** ![g3casey](https://avatars.discourse-cdn.com/v4/letter/g/e19b73/32.png) [@g3casey](https://discuss.huggingface.co/u/g3casey)\
**Post date:** [November 8, 2021, 5:15pm UTC](https://discuss.huggingface.co/t/layoutlm-for-table-detection-and-extraction/7015/2 "2021-11-08T17:15:59Z")

</div>

@ujjayants LayoutLM is not designed for that but you may want to look at

> **[GitHub - Layout-Parser/layout-parser: A Unified Toolkit for Deep Learning...](https://github.com/Layout-Parser/layout-parser)**
>
> A Unified Toolkit for Deep Learning Based Document Image Analysis - GitHub - Layout-Parser/layout-parser: A Unified Toolkit for Deep Learning Based Document Image Analysis

It may do what you are looking for.

---

<div class="post-metadata">

**Author:** ![PackElend](https://avatars.discourse-cdn.com/v4/letter/p/90ced4/32.png) [@PackElend](https://discuss.huggingface.co/u/PackElend)\
**Post date:** [July 6, 2022, 7:38am UTC](https://discuss.huggingface.co/t/layoutlm-for-table-detection-and-extraction/7015/3 "2022-07-06T07:38:12Z")

</div>

[TAPAS](https://huggingface.co/docs/transformers/model_doc/tapas) could be the right choice.  
Have you made any progress?  
My idea [Serialize bank statements from PDF to CSV - Beginners - Hugging Face Forums](https://discuss.huggingface.co/t/serialize-bank-statements-from-pdf-to-csv/19728) has the same requirements but still don’t know how to start.

---

<div class="post-metadata">

**Author:** ![PackElend](https://avatars.discourse-cdn.com/v4/letter/p/90ced4/32.png) [@PackElend](https://discuss.huggingface.co/u/PackElend)\
**Post date:** [July 11, 2023, 2:51pm UTC](https://discuss.huggingface.co/t/layoutlm-for-table-detection-and-extraction/7015/4 "2023-07-11T14:51:12Z")

</div>

there could be another option but I could not find a example yet how to convert the output to csv  
[Table Transformer (huggingface.co)](https://huggingface.co/docs/transformers/main/en/model_doc/table-transformer)