Skip to content
This repository has been archived by the owner on Dec 16, 2022. It is now read-only.

Releases: allenai/allennlp-models

Version 1.0.0 Release Candidate 5

27 May 00:50
Compare
Choose a tag to compare
Pre-release

Changed

  • Moved the models into categories based on their format

Fixed

  • Made transformer_qa predictor accept JSON input with the keys "question" and "passage" to be consistent with the reading-comprehension predictor.

Added

  • conllu dependency (previously part of allennlp's dependencies)

Commits

9b6a2b0 Make the linter happy
ac0ed48 removed unused stuff from setup.py (#62)
bd42c2c Update dev-requirements.txt
c7cb218 Update version for release 1.0.0rc5
822202f make transformer qa predictor consistent with other rc predictors (#60)
114c203 add conllu dependency (#61)
888596c Renames the model categories (#58)
d21af08 Update README.md
8ce4d0f add a 'clean' phony make target
e0a590f add a CHANGELOG (#56)
452b213 use tokenless codecov upload (#57)
20fb509 Move seq2seq tests to right test directory (#51)
1de3bac tick version to rc5

Version 1.0.0 Release Candidate 4

14 May 21:56
Compare
Choose a tag to compare
Pre-release

Commits

ef74f37 release workflow fixes
a6bd156 Update version for release v1.0.0rc4
794640f Grammar
ef4df72 add commit and release Dockerfiles with instructions (#54)
f8bdf21 remove pre-commit dev dependency (#53)
54a5df8 specify namespace in roberta mnli (#52)
05f48c7 Be compatible with the new tokenizers code (#44)
3c764c9 Attach nltk and HF caches to docker containers in CI (#49)
96de66f rename CI workflows
08e7655 Add GPU test workflow (#48)
13bebf0 fix new linting errors (#47)
83a14c4 ensure docker images get commit SHA tag
96d623c Adds the pretrained module from allennlp-hub. (#42)
36020d5 fix
98e222d update for new AllenNlpTestCase (#45)
356eacf We need to specify the namespace now (#43)
34204ed Clarify local install instructions (#41)
c833e7b get rid of plugin mechanism (#40)
d764ffc Fixes for ELMo embedder API change (#39)
0449e5b Fix coref interpret test to use the right gradient length (#38)
0b040af Fixes the roberta snli model (#37)
316c909 reduce verbosity of linters (#36)
04a2c73 Depend on an unreleased version of AllenNLP (#33)
3a50436 add install instructions to README (#32)
4e7698b Move default predictors (#31)

Version 1.0.0 Release Candidate 3

28 Apr 15:44
64185b8
Compare
Choose a tag to compare
Pre-release

Commits

dcc9674 tick version
5209e07 Even better token annotation (#27)
2cd9a73 try fix package build (#29)
f6cb9ae small CI improvements (#26)
92e6a85 Handles mismatched token annotation better (#24)

Version 1.0.0 Release Candidate 2

23 Apr 23:38
Compare
Choose a tag to compare
Pre-release

Commits

8cf8621 Bump pre-commit from 2.2.0 to 2.3.0 (#25)
7515207 Add PyPI badge/link
1f841a3 Load allennlp_models with plugins (#23)
0eff217 Update README.md (#20)
e58d657 ensure setuptools and wheel installed first (#22)
2e1454f Ensure code coverage computed correctly (#21)

Release v1.0.0.rc1 corresponding to allennlp release

20 Apr 16:37
Compare
Choose a tag to compare
Adds tag v1.0.0.rc1 for PyPY