-
Notifications
You must be signed in to change notification settings - Fork 2.7k
New issue
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
Squad inference example #390
Conversation
Signed-off-by: Yang Zhang <[email protected]>
Signed-off-by: Yang Zhang <[email protected]>
"start_positions": NeuralType(tuple('B'), ChannelType(), optional=True), | ||
"end_positions": NeuralType(tuple('B'), ChannelType(), optional=True), |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@okuchaiev What exactly does optional mean? How does it work if I only return end_positions but not start_positions?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
start_positions, end_positions are the "labels", they are optional to make the data_layer more general for inference cases where i do not have the groundtruth.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That makes sense, I just didn't know we had this option, and want to know how it works.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Please update changelog
examples/nlp/question_answering/question_answering_squad_infer.py
Outdated
Show resolved
Hide resolved
examples/nlp/question_answering/question_answering_squad_infer.py
Outdated
Show resolved
Hide resolved
…in script Signed-off-by: Yang Zhang <[email protected]>
Signed-off-by: Yang Zhang <[email protected]>
Closes NVIDIA#390 and NVIDIA#391 Signed-off-by: Trevor Grant <[email protected]> Signed-off-by: Mark Sturdevant <[email protected]> Co-authored-by: Mark Sturdevant <[email protected]>
No description provided.