# Adding small data in fine tune model - bert

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fadding-small-data-in-fine-tune-model-bert%2F24722%3E%5C
**Category:** Models\
**Created:** [October 20, 2022, 9:58am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fadding-small-data-in-fine-tune-model-bert%2F24722 "2022-10-20T09:58:18Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![chintanshrinath](https://avatars.discourse-cdn.com/v4/letter/c/bbe5ce/32.png) [@chintanshrinath](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fchintanshrinath)%5C
**Post date:** [October 20, 2022, 9:58am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fadding-small-data-in-fine-tune-model-bert%2F24722%2F1 "2022-10-20T09:58:18Z")

</div>

Hi everyone

I have trained bert model for multi-label text classification for 15+ classes. So sometime, on inference side, it is not able to predict as expected for some sentence / paragraph.

So, currently I am fine-tuning entire model with train data + extra data. Is ther any approach, where I can add only new data instead of everything. So time can be saved.

Thanks