How to use Bert_Review for E2E ABSA?

[RalphSchuurman]

Hello @howardhsu ,

This is a very nice repo. I am interested in running Bert_Review for E2E ABSA, which is possible according to the transformers.md file.

I can load the post-training weights using Transformers by running this code
`import torch
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("activebus/BERT_Review")
model = AutoModel.from_pretrained("activebus/BERT_Review")`

but how would I then use BERT_Review for E2E ABSA? For example a text.txt file with the example sentence: 'The food was good' to obtain 'food' as an aspect and 'POS' as a sentiment.

[AndrewBai-Marketing]

Did you figure this out?

[howardhsu]

Thanks for your question. You need to fine-tune the model in e2e setup.

On Tue, Nov 29, 2022 at 2:31 PM AndrewSharng ***@***.***> wrote: Did you figure this out? — Reply to this email directly, view it on GitHub <#26 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACRK37Y5BBRMNBMCKZR2D7DWKZ743ANCNFSM5QJ2Q2WQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Homepage: https://howardhsu.github.io/ Linkedin: https://www.linkedin.com/in/hu-xu-9852403b/ Google Scholar: https://scholar.google.com/citations?user=SaH2yWMAAAAJ Twitter: https://twitter.com/Hu_Hsu Email: ***@***.***