Recommenders 1.1.1
New algorithms or improvements
- Reduce iterations of W&D to reduce the integration tests time in #1698
- Implementation of most frequent recommendation in #1666
- Implement time_now for sarplus in #1719 #1721
- Add a fast failure in SAR+ if the similarity metric is not within the options in #1743
- SAR item similarity dtype correction in #1751
- Simplify SAR test data loading functions in #1752
- Reformat SAR+ SQL queries in #1772
- Add new item similarity metrics for SAR in #1754
New utilities or improvements
- Rewrite get_top_k_items() to improve runtime in #1748
- Optimized Spark recall_at_k time performance in #1796
New notebooks or improvements
- Fix missing import in FastAI notebook #1708
- Review NCF notebook in #1703 #1712
- Review LigthFM notebook and add test in #1706
- Review BPR notebook in #1704
- Review LightGCN notebook in #1714
- Review DKN notebook in #1722
- Review SAR notebook #1738 #1768
Other features
- Enable distributed tests with AzureML #1696 #1717 #1729 #1733 #1739 #1747 #1732 #1755 #1763 #1771 #1773 #1775 #1787 #1788 #1794
- Added tests for Python 3.8 and 3.9 in #1756
- Image of contributors in #1692
- Update README.md in #1709 #1711 #1767
- Error in codeowners file in #1699
- Add test to check if CuDNN is enabled in #1715
- Update docker image reference to internal registry in #1727
- Fixed a link error in data_transform.ipynb in #1736
- Added tests for ranking function get_top_k_items() in #1757
- Fix memory error in CPU nightly workflow in #1759
- Update test infrastructure explanation #1776 #1777
- Added time performance tests in #1765
- Add path filter to avoid triggering unit tests when we change a markdown in #1791
Full Changelog: 1.1.0...1.1.1