CoreNLP
CoreNLP copied to clipboard
stanfordnlp
Enable Keeping Nested/Overlapping Matches from CoreMapExpressionExtractor and Expose the Name
In my work we have used tokens-regex, but I'm looking to use the lower-level CoreMapExpressionExtractor and to modify it to allow for matches that are either nested, or overlapping.
For my use case I want to see all possible matches, not just the largest or non-overlapping matches. I'm only using the result functionality of the language, not doing any annotating.
I would also like to expose the name of the ExtractFunc of the MatchedExpression.
I have changes checked in to a local branch where I expose the name, and allow for the Env of the CoreMapExpressionMatcher to allow for all the matches to be returned or not.
Fork the repo and push it to your own local repo
On Mon, Oct 10, 2022 at 10:37 AM jrsperry @.***> wrote:
In my work we have used tokens-regex, but I'm looking to use the lower-level CoreMapExpressionExtractor and to modify it to allow for matches that are either nested, or overlapping.
For my use case I want to see all possible matches, not just the largest or non-overlapping matches. I'm only using the result functionality of the language, not doing any annotating.
I would also like to expose the name of the ExtractFunc of the MatchedExpression.
I have changes checked in to a local branch where I expose the name, and allow for the Env of the CoreMapExpressionMatcher to allow for all the matches to be returned or not. I'm getting access denied in pushing to a new branch however.
— Reply to this email directly, view it on GitHub https://github.com/stanfordnlp/CoreNLP/issues/1307, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2AYWOUMKWTIHDVNHCQCR3WCRH4RANCNFSM6AAAAAARBRPMWU . You are receiving this because you are subscribed to this thread.Message ID: @.***>
i've already done that, I figured it may be of use to others using the CoreMapExpressionExtractor. I'll attach a PR for consideration.
Not sure how to assign it to this issue but here's the PR.
https://github.com/stanfordnlp/CoreNLP/pull/1308