# Mutli-label classification for large free text input

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fmutli-label-classification-for-large-free-text-input%2F128863%3E%5C
**Category:** Beginners\
**Created:** [December 2, 2024, 3:29pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fmutli-label-classification-for-large-free-text-input%2F128863 "2024-12-02T15:29:18Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Hisham81](https://avatars.discourse-cdn.com/v4/letter/h/e68b1a/32.png) [@Hisham81](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FHisham81)%5C
**Post date:** [December 2, 2024, 3:29pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fmutli-label-classification-for-large-free-text-input%2F128863%2F1 "2024-12-02T15:29:18Z")

</div>

I have a dataset with 2 columns,

1. around 100k lines
2. large amount of free text (around 10k words per cell).
3. a list of labels that this free text belong to. usually between 5 to 30 list items.
4. the full unique list items (labels) are around 12k.

I was wondering what would be the best way to train a model on this dataset, as majority of transformers has a maximum of 512 or 1024 only tokens as input.

---

<div class="post-metadata">

**Author:** ![juhoinkinen](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/juhoinkinen/32/65095_2.png) [@juhoinkinen](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fjuhoinkinen)%5C
**Post date:** [December 3, 2024, 8:02pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fmutli-label-classification-for-large-free-text-input%2F128863%2F2 "2024-12-03T20:02:08Z")

</div>

Sounds like [Annif](https://annif.org/) could be usable for you. See also [Annif-wiki](https://github.com/NatLibFi/Annif/wiki) and [Annif-tutorial](https://github.com/NatLibFi/Annif-tutorial) for instructions.

(FYI, I’m one of Annif developers.)

---

<div class="post-metadata">

**Author:** ![Hisham81](https://avatars.discourse-cdn.com/v4/letter/h/e68b1a/32.png) [@Hisham81](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FHisham81)%5C
**Post date:** [December 10, 2024, 6:54am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fmutli-label-classification-for-large-free-text-input%2F128863%2F3 "2024-12-10T06:54:36Z")

</div>

I was looking for ready model but for best practice concept that need to be used in this case.

Thank you for your input.