dropout(): argument 'input' (position 1) must be Tensor, not str when using Bert with Huggingface

Question

My code was working fine and when I tried to run it today without changing anything I got the following error:

dropout(): argument 'input' (position 1) must be Tensor, not str

Would appreciate if help could be provided. Could be an issue with the data loader?

Welcome to StackOverflow. Please take a look at the guidelines before posting a question. If you want to get answer from SO, your question needs to be reproducible in order for others to help you. https://stackoverflow.com/help/how-to-ask. This could be a problem anywhere within your code. — Ananda, Dec 01 '20 at 04:58

score 22 · Answer 1 · answered Dec 01 '20 at 19:36

22

if you use HuggingFace, this information could be useful. I have same error and fix it with adding parameter return_dict=False in model class before dropout: outputs = model(**inputs, return_dict=False)

answered Dec 01 '20 at 19:36

Evgeny Vorsin

221
1
2

2

Thank You! Worked well for me. I think this should be the accepted answer – tschomacker Mar 17 '21 at 13:00

score 1 · Answer 2 · edited May 09 '21 at 11:39

1

I was also working on same repo. There is a class probably named Bert_Arch that inherits the nn.Module and this class has a overriden method named forward. Inside forward method just add the parameter 'return_dict=False' to the self.bert() method call. Replace

_, cls_hs = self.bert(sent_id, attention_mask=mask)

with

_, cls_hs = self.bert(sent_id, attention_mask=mask, return_dict=False)

edited May 09 '21 at 11:39

tomerpacific

3,092
8
22
41

answered May 09 '21 at 05:23

Utkarsh Dadhich

11
1

dropout(): argument 'input' (position 1) must be Tensor, not str when using Bert with Huggingface

2 Answers2

Linked