Downloads · 30 days
10
2% of all-time downloads
roshnir/bert-multi-uncased-trained-squadv2
bert-multi-uncased-trained-squadv2 is a question answering model from roshnir. Use it when the input is a question plus a passage. It is set up for transformers.
mBERT base uncased model trained on 50% SQUAD data. This model can further be used to fine-tune using dev data for QA system on a specific language. The process is similar to what was followed in MLQA paper[https://ac…
Downloads · 30 days
10
2% of all-time downloads
All-time downloads
465
Public
Repo size
1.3 GB
Likes
0
Public
Click a slice to open those files.
.bin667 MB · 99%
From the Hugging Face model README
mBERT base uncased model trained on 50% SQUAD data. This model can further be used to fine-tune using dev data for QA system on a specific language. The process is similar to what was followed in MLQA paper[https://aclanthology.org/2020.acl-main.421.pdf].