v23.02.00
github-actions
released this
08 Mar 16:39
·
110 commits
to main
since this release
What’s Changed
🚀 Features
📄 Documentation
- update README.md/docs @radekosmulski (#804)
- add links to docs @radekosmulski (#802)
- Review getting started examples @radekosmulski (#779)
- Support matrix updates for 22.12 @nvidia-merlin-bot (#784)
- Replace next(iter(loader)) with loader.peek() @edknv (#778)
- Address review feedback @mikemckiernan (#755)
- Inconsistent order of columns in getting-started-movielens tutorial @mikemckiernan (#758)
- Remove legacy Criteo examples @bschifferer (#757)
- Review Criteo Examples and Added Unittest @bschifferer (#756)
- Add SageMaker example to docs @mikemckiernan (#754)
- Support matrix updates for 22.11 @nvidia-merlin-bot (#747)
🔧 Maintenance
- container size increase to 16gb @jperez999 (#847)
- Install hps backend and hps trt plugin in hugectr/tf/pytorch @EmmaQiaoCh (#783)
- Add transformers==4.26.1 to CI runner image @oliverholworthy (#843)
- Pin dask/distributed package versions after other installs in CI image @karlhigley (#841)
- Fix a variety of minor issues with the CI container build @karlhigley (#840)
- Adjust Merlin library installs in the nightly images @karlhigley (#839)
- Fix string literal comparison in CI container test script @karlhigley (#838)
- Consolidate size checks in container builds @karlhigley (#837)
- fix size check to exclude mounts @jperez999 (#836)
- Skip the container size check for the CI container @karlhigley (#832)
- Remove NVT Triton back-end install from Docker base image @karlhigley (#831)
- Pin pynvml to 11.4.1 @oliverholworthy (#830)
- fix syntax error @nv-alaiacano (#821)
- Shallow Clone HugeCTR submodules and remove
.git/modules
@oliverholworthy (#817) - clean up .git from cloned repos after installing @nv-alaiacano (#812)
- Use
--no-cache-dir
on all pip install commands in dockerfiles @oliverholworthy (#808) - Add container size check to check script @jperez999 (#790)
- Container size fix for <15gb @jperez999 (#789)
- Fix docs build to use tox, as described @mikemckiernan (#803)
- Install protobuf==3.20.3 for hugectr @EmmaQiaoCh (#777)
- update dockerfile merlin base container @jperez999 (#781)
- save env var for systems test @jperez999 (#780)
- Replace next(iter(loader)) with loader.peek() @edknv (#778)
- adding workflow for gpu ci on gha runner @jperez999 (#775)
- revert ci change @bschifferer (#774)
- Remove data manipulation in integration test and run the first notebo… @edknv (#773)
- Pin pyarrow version for integration tests @oliverholworthy (#770)
- Replace
ensemble_model
withexecutor_model
@oliverholworthy (#772) - Remove hugectr test @jperez999 (#771)
- Increase timeout of scaling-criteo/03-Training-with-HugeCTR test @oliverholworthy (#769)
- Bump
treelite
to 2.4.0 to match with Triton 22.11 @karlhigley (#750) - Pin protobuf version in merlin base container to 3.20.3 @oliverholworthy (#765)
- add gcp label to jenkinsfile @AyodeAwe (#762)
- Update hps_tf folder name @EmmaQiaoCh (#763)
- fixes support library lookups for torch @jperez999 (#761)
- Migrate the legacy examples to the Merlin repo @karlhigley (#742)
- Replace sklearn by scikit-learn in pip install @edknv (#753)
- Adjust package paths for
torch
,numpy
,numba
in Torch Docker image @karlhigley (#751) - update drafter to work on tags @jperez999 (#743)
- Add Jenkinsfile @AyodeAwe (#734)