-
Notifications
You must be signed in to change notification settings - Fork 54
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
Bug in few-shot sampling process #15
Comments
Hi, I just find the released episodes for train/dev/test contain episodes that have support count > 2*K shot. |
Thanks for reminding! We have noticed the bug here and it does lead to some ">2*K shot" issues. We will fix the bug soon. The influence should be small, however, since it is not a frequent case. |
@cyl628 Hi! Why you closed the bug? Did you fix the bug? |
Hi folks, Thanks! |
Hi!
It seems like function
Few-NERD/util/fewshotsampler.py
Line 62 in 3935544
has a bug, which affects sampling process.
You can see "for loop" and 4 "if" inside it. If flag
isvalid
is set toTrue
orFalse
in some iterations, it can be overwritten in next iterations of the loop. So, for example, first class breaks the rule shots_count <= 2*K, but second class is new and flagisvalid
will be returned asTrue
from__valid_sample__
, but sample breaks validation rules.Thus, with this bug, we can sample samples, which are breaking sampling rules, e.g. with shots_count > 2*K.
What do you think? Does this bug affects results somehow? Maybe in the good way even.
The text was updated successfully, but these errors were encountered: