Getting substantially different results when run on local system vs when run on hf inference api
#6
by
Gopalkrishna
- opened
@MoritzLaurer please help resolve this issue at the earliest. Thank you.
@MoritzLaurer I am currently using transformers version 4.40.2 is that causing the problem?
@Gopalkrishna
the difference is that for the first output you've most likely set multi_label=False
while for the widget you have set multi_label=True
MoritzLaurer
changed discussion status to
closed
@MoritzLaurer Nope. Have kept multi=True for both. What do you think is the reason now?
@MoritzLaurer thank you for the suggestion i had a typo in naming the multi_label = True attribute. Issue has been resolved thank you.