Skip to content
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

tag2ts problem #15

Open
NourAlMerey opened this issue Sep 7, 2020 · 2 comments
Open

tag2ts problem #15

NourAlMerey opened this issue Sep 7, 2020 · 2 comments

Comments

@NourAlMerey
Copy link

NourAlMerey commented Sep 7, 2020

When will this condition be satisfied,since it's impossible for the sentiments list to be bigger than one in this condition?

elif pos == 'B': beg = i if len(sentiments) > 1: # remove the effect of the noisy I-{POS,NEG,NEU} #print(sentiments) sentiments = [sentiments[-1]]

@NourAlMerey NourAlMerey reopened this Sep 7, 2020
@NourAlMerey NourAlMerey changed the title What's the difference between max_steps and num_train_epochs? tag2ts problem Sep 7, 2020
@lixin4ever
Copy link
Owner

Sorry for the late reply.

When tagging_schema is BIEOS, the sentiment tag list (i.e., the variable sentiments) will be emptied if and only if the current boundary tag is "S" or "E".

In the ideal case, we will not meet the next "B-POS/NEG/NEU" until the previous span is completed (i.e., encounter "S-POS/NEG/NEU" or "E-POS/NEG/NEU"), however, this is not guaranteed in the testing phase. Considering the example that the current prediction is "B-POS" and the previous predictions are {"I-NEG", "I-NEG", "I-NEG", "I-POS"}. Since the sentiments will accumulate the noisy sentiment tags, we need to remove them and start from the sentiment tag associated with the current "B".

@NourAlMerey
Copy link
Author

Oh now I get it. Thank you

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants