# How to do NER with a BERT type models?

**URL:** <https://discuss.huggingface.co/t/how-to-do-ner-with-a-bert-type-models/16932>\
**Category:** Beginners\
**Created:** [April 18, 2022, 2:40pm UTC](https://discuss.huggingface.co/t/how-to-do-ner-with-a-bert-type-models/16932 "2022-04-18T14:40:11Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![srishti-hf1110](https://avatars.discourse-cdn.com/v4/letter/s/9fc29f/32.png) [@srishti-hf1110](https://discuss.huggingface.co/u/srishti-hf1110)\
**Post date:** [April 18, 2022, 2:40pm UTC](https://discuss.huggingface.co/t/how-to-do-ner-with-a-bert-type-models/16932/1 "2022-04-18T14:40:11Z")

</div>

Hi, please help me with my question - I’m very new to using BERT based models 7 it’s the first time I’m doing this.  
I want to get a disease tagger (NER) for some pubmed files I have converted to .txt format. I’m specifically interested in using BioBert or PubmedBert both of which are available in HF. However,  
I think there’s no way I can accomplish this without fine tuning. But I do not seem to have any labelled data for fine tuning the models for my downstream task. Is there any way I could get around this? Please help.  
Or maybe if any one knows of any other pre-trained models to accomplish this task?