Closed
Description
PairwiseDistancesReductions have been introduced as a hierarchy of Cython classes to implement back-ends of some scikit-learn algorithms.
Pieces of work include:
- ENH Introduce
PairwiseDistancesReductionandPairwiseDistancesArgKmin(feature branch) #22134 - ENH Introduce
PairwiseDistancesRadiusNeighborhood#22320 - FEA Add support for float32 on
PairwiseDistancesReductionusing Tempita #23865 - Support sparse-dense data-sets pairs:
- FEA CSR support for all
DistanceMetric#23604 - FEA Fused sparse-dense support for
PairwiseDistancesReduction#23585 - MAINT Introduce
MiddleTermComputer, an abstraction generalizingGEMMTermComputer#24807 by @Vincent-Maladiere - EHN Optimized CSR-CSR support for
Euclideanspecializations ofPairwiseDistancesReductions#24556 by @Vincent-Maladiere - ENH Add the fused CSR dense case for Euclidean Specializations #25044
- FEA CSR support for all
- MAINT Adapt scheduling of
PairwiseDistancesRadiusNeighborhood#22829 - DOC Fix docstrings and comments for
PairwiseDistancesReduction#23978 - MAINT Adapt
PairwiseDistancesReductionheuristic forstrategy="auto"#24043 - Document and communicate about the changes of behaviour regarding
n_jobs:- DOC Update "Parallelism, resource management, and configuration" section #24997
- Add a comment in impacted user-facing
PairwiseDistancesReductions-backed API's doc-strings regarding the non-use ofn_jobs
- Dedicated ASV benchmark suite:
scikit-learn/pairwise-distances-reductions-asv-suite
Note that this needs not be personal work, I would be really glad having others help on this subject, proposing changes and implementations! 🙂