We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Hi Jingtao,
Thanks for open-source this interesting work and detailed implementation.
I have a question for the hard negative generation, is the mode should be train instead of dev?
mode
train
dev
python ./star/prepare_hardneg.py \ --data_type passage \ --max_query_length 32 \ --max_doc_length 256 \ --mode dev \ --topk 200
The text was updated successfully, but these errors were encountered:
No branches or pull requests
Hi Jingtao,
Thanks for open-source this interesting work and detailed implementation.
I have a question for the hard negative generation, is the
mode
should betrain
instead ofdev
?The text was updated successfully, but these errors were encountered: