# Bert NER model start and end position None after fine-tuning

**URL:** <https://discuss.huggingface.co/t/bert-ner-model-start-and-end-position-none-after-fine-tuning/41790>\
**Category:** Beginners\
**Created:** [June 1, 2023, 3:33pm UTC](https://discuss.huggingface.co/t/bert-ner-model-start-and-end-position-none-after-fine-tuning/41790 "2023-06-01T15:33:28Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![amitsinghkec](https://avatars.discourse-cdn.com/v4/letter/a/4da419/32.png) [@amitsinghkec](https://discuss.huggingface.co/u/amitsinghkec)\
**Post date:** [June 1, 2023, 3:33pm UTC](https://discuss.huggingface.co/t/bert-ner-model-start-and-end-position-none-after-fine-tuning/41790/1 "2023-06-01T15:33:28Z")

</div>

HI All,

I have fine-tuned a BERT NER model to my dataset. The model that I am fine-tuning is “dslim/bert-base-NER”. I have been successfully able to train the model using the following script as refrence:  
[https://colab.research.google.com/github/NielsRogge/Transformers-Tutorials/blob/master/BERT/Custom\_Named\_Entity\_Recognition\_with\_BERT\_only\_first\_wordpiece.ipynb#scrollTo=zPDla1mmZiax](https://colab.research.google.com/github/NielsRogge/Transformers-Tutorials/blob/master/BERT/Custom_Named_Entity_Recognition_with_BERT_only_first_wordpiece.ipynb#scrollTo=zPDla1mmZiax)

The prediction form the base model contained the start and end position of the word in the original text like:  
{‘entity\_group’: ‘ORG’, ‘score’: 0.9992545247077942, ‘word’: ‘A’, **‘start’: 10, ‘end’: 11** }  
{‘entity\_group’: ‘ORG’, ‘score’: 0.998507097363472, ‘word’: ‘##bc Corp Ltd’, ‘start’: 11, ‘end’: 22}

While the prediction from the re-trained model is:  
{‘entity\_group’: ‘ORG’,  
‘score’: 0.747031033039093,  
‘word’: ‘##7’,  
‘start’: None,  
‘end’: None},  
{‘entity\_group’: ‘ORG’,  
‘score’: 0.9055356582005819,  
‘word’: ‘Games , Inc’,  
**‘start’: None,**  
\*\* ‘end’: None\*\*}

I am passing the position ids to the model during the training process. I looked at the model training parameters but, could not find a way to pass start and end position of the words to model training process. I have the start and end position of the tokenized words.