KoBART-base-v2
With the addition of chatting data, the model is trained to handle the semantics of sequences longer than KoBART.
from transformers import PreTrainedTokenizerFast, BartModel
tokenizer = PreTrainedTokenizerFast.from_pretrained('hyunwoongko/kobart')
model = BartModel.from_pretrained('hyunwoongko/kobart')
Performance
NSMC
- acc. : 0.901
hyunwoongko/kobart
- Added bos/eos post processor
- Removed token_type_ids
- Downloads last month
- 31,912
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.