Skip to content

fix(medcat): fix tokenisation - and correct PredicatedAnnotations text - #639

Open
adam-sutton-1992 wants to merge 6 commits into
mainfrom
tokenization_fix
Open

adam-sutton-1992 wants to merge 6 commits into
mainfrom
tokenization_fix

Conversation

@adam-sutton-1992

Copy link
Copy Markdown
Contributor

Hi,

Short and sweet. Bisect left on token usage of creating entities from tokens where you got gold standard labels making:

Label: vitals in context: "...has Vitals: ..." ->

Tokens: ["Vitals", ":"]

Instead be:

Tokens: ["Vitals"]

I also edited the PredictedAnnotations in stats contain raw text, and detected names for clarity. Instead of just having the detected name.

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

Successfully merging this pull request may close these issues.

1 participant