Code

Entity Mention Linker#3388

Code

Merged

alanakbik merged 58 commits intomasterflairNLP/flair:masterfrom bf/bio-entity-normalizationflairNLP/flair:bf/bio-entity-normalizationCopy head branch name to clipboardFeb 8, 2024

+2,417-100

Commits

Commits on Feb 4, 2024

• Initial version (already adapted to recent Flair API changes)

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  a01e060

• Revise mention text pre-processing: define general interface and adapt basic text and Ab3P pre-processing to the new structure; fix bug in Ab3P abbreviation detection

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  22bc93a

• Refactor entity linking model structure

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  acf5fb6

• Update documentation

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  56c89ba

• Introduce separate methods for pre-processing (1) entity mentions from text and (2) entity / concept names from an knowledge base or ontology

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  51fe951

• Fix formatting

  

  alanakbik

  authored and

  helpmefindaname

  committed
  19f74fb

• feat(test): biomedical entity linking

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  e5342d5

• fix(test): hold on w/ automatic tests for now

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  4ef5924

• fix(bionel): start major refactoring

  Show description for abc42b5
  

  Samule Garda

  authored and

  helpmefindaname

  committed
  abc42b5

• fix(bionel): major refactor

  Show description for bdc3e8a
  

  Samule Garda

  authored and

  helpmefindaname

  committed
  bdc3e8a

• fix(bionel): assign entity type

  Show description for 48e8ae7
  

  Samule Garda

  authored and

  helpmefindaname

  committed
  48e8ae7

• fix(biencoder): set sparse encoder and weight

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  e6b57eb

• fix(bionel): address comments

  Show description for 0d3cec2
  

  Samule Garda

  authored and

  helpmefindaname

  committed
  0d3cec2

• fix(candidate_generator): container for search result

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  99a109f

• fix(predict): default annotation layer iff not provided by use

  Show description for 301988e
  

  Samule Garda

  authored and

  helpmefindaname

  committed
  301988e

• fix(label): scores can be >= or <=

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  c14d6ce

• fix(candidate): parametrize database name

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  c66789a

• feat(candidate_generator): cache sparse encoder

  Show description for 70c0c7d
  

  Samule Garda

  authored and

  helpmefindaname

  committed
  70c0c7d

• fix(candidate_generator): minor improvements

  Show description for 37a2458
  

  Samule Garda

  authored and

  helpmefindaname

  committed
  37a2458

• feat(linking_candidate): pretty print

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  bc801fc

• fix(candidate_generator): check sparse encoder for sparse search

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  3677658

• feat(candidate_generator): add sparse index

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  414b5a8

• fix(candidate_generator): KISS: sparse search w/ scipy sparse matrices

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  1783eab

• Minor update to comments and documentation

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  8c908ba

• Fix tests and type annotations

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  c6273e6

• code format and fix ruff & most mypy errors.

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  9a3fff6

• refine interface of BiomedicalEntityLinker

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  fd9507f

• refactor knowledgebase datasets

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  a374040

• refactor CandidateSearchIndex

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  2c12f14

• fix rebase errors

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  e398213

• WIP: add save functionality

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  51a1a57

• add load & save functionality

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  ce85e3d

• fix naming

  helpmefindaname

  committed
  750ba7d

• add hf model download

  helpmefindaname

  committed
  c5fa9c9

• fix ruff & mypy errors

  helpmefindaname

  committed
  94db5b3

• fix entity linking test

  helpmefindaname

  committed
  1d55fec

• fixed selection of knowledge base identifiers for entity_mention_linking

  

  Xing Wang

  authored and

  helpmefindaname

  committed
  a4a27dc

• fix(dictionary): corrections in file parsing

  

  Samule Garda

  authored and

  helpmefindaname

  committed
  6689297

• fix: preprocessing, candidate generator, linker

  Show description for c2c53a9
  

  Samuele Garda

  authored and

  helpmefindaname

  committed
  c2c53a9

• feat(tests): test preprocessing

  

  Samuele Garda

  authored and

  helpmefindaname

  committed
  1efe956

• Minor fix in pre-processing pipeline

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  5da7494

• Add support for label and entity type definition + fix tests

  

  Mario Sänger

  authored and

  helpmefindaname

  committed
  4cd86b0

• fix: formatting and type checking

  

  Samuele Garda

  authored and

  helpmefindaname

  committed
  aaa5f39

• add batchsize to prediction instead of only embedding to reduce memory usage

  helpmefindaname

  committed
  def9905

• add evaluation function

  helpmefindaname

  committed
  03092c2

• fix(linker): extraction of entity mentions in predict

  Show description for bf88d0a
  

  Samuele Garda

  authored and

  helpmefindaname

  committed
  bf88d0a

• fix(predict): ensure mentions extraction works with legacy classifier

  Show description for 7b824a1
  

  Samuele Garda

  authored and

  helpmefindaname

  committed
  7b824a1

• chore: update tests

  

  Samuele Garda

  authored and

  helpmefindaname

  committed
  c8cc120

• fix(tests): normalized entity type name

  

  Samuele Garda

  authored and

  helpmefindaname

  committed
  94aaca1

• fix(logging): deprecated logger.warn

  

  Samuele Garda

  authored and

  helpmefindaname

  committed
  75ea402

• improve typing and run black

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  d9805be

• add datasets for nel-bioner evaluation

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  8ab67d2

• mark heavy test as integration test

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  654ed8d

• add metadata to labels for cnadidate names

  

  Benedikt Fuchs

  authored and

  helpmefindaname

  committed
  efb1018

• fix black ruff and mypy

  helpmefindaname

  committed
  44c1413

Commits on Feb 6, 2024

• make test more memory efficient by only loading the smallest model

  helpmefindaname

  committed
  d7fa7dd

Commits on Feb 7, 2024

• chore(docs): add docstrings fro datasets

  

  Samuele Garda

  committed
  ba833f0

Commits on Feb 8, 2024

• fix(bigbio): better naming & fix ruff

  

  Samuele Garda

  committed
  44d73a2