# Unexpected input type after export

**URL:** <https://discuss.huggingface.co/t/unexpected-input-type-after-export/77979>\
**Category:** 🤗Transformers\
**Created:** [March 18, 2024, 6:14pm UTC](https://discuss.huggingface.co/t/unexpected-input-type-after-export/77979 "2024-03-18T18:14:42Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![JBigelow](https://avatars.discourse-cdn.com/v4/letter/j/f07891/32.png) [@JBigelow](https://discuss.huggingface.co/u/JBigelow)\
**Post date:** [March 18, 2024, 6:14pm UTC](https://discuss.huggingface.co/t/unexpected-input-type-after-export/77979/1 "2024-03-18T18:14:42Z")

</div>

Hello,  
I exported this model: /[bert-large-cased-finetuned-conll03-english](https://huggingface.co/dbmdz/bert-large-cased-finetuned-conll03-english)  
to onnx using this code:

import torch  
from transformers import AutoModelForTokenClassification, AutoTokenizer

#load model/tokenizer  
model\_id = “dbmdz/bert-large-cased-finetuned-conll03-english”  
model = AutoModelForTokenClassification.from\_pretrained(model\_id)  
tokenizer = AutoTokenizer.from\_pretrained(model\_id)  
dummy\_model\_input = tokenizer(“JJ and his friends are going to the Park later.”, return\_tensors=“pt”)

#Export to onnx  
torch.onnx.export(  
model,  
tuple(dummy\_model\_input.values()),  
f= “dbmdz/bert\_large\_cased\_finetuned\_conll03\_english\_onnx/”,  
input\_names=[‘input\_ids’, ‘attention\_mask’],  
output\_names=[‘logits’],  
dynamic\_axes={‘input\_ids’ : {0: ‘batch\_size’, 1: ‘sequence’},  
‘attention\_mask’: {0: ‘batch\_size’, 1: ‘sequence’},  
‘logits’: {0: ‘batch\_size’, 1: ‘sequence’}},  
do\_constant\_folding= True,  
opset\_version= 13,  
)

After export My .onnx model has ‘input\_ids’, ‘attention\_mask’, and ‘input.3’ inputs, and i am unsure how or why i got the input.3 input.