Trian the CLEAR model

python -m src.model.CLEAR

Load dataset

Now run following line then we will have a cache graph that contains encoded contexts and retrieved topk contexts.

python -m src.dataset.pubmedqa --retriever bm25 --topk 10

Name		Name	Last commit message	Last commit date
Latest commit History 43 Commits
dataset		dataset
src		src
.gitignore		.gitignore
README.md		README.md
requirements.txt		requirements.txt
test.ipynb		test.ipynb