0.7.0: 20x-40x faster batch training, and a documentation pass - #22

Merged

andremsouza merged 9 commits into

perf/vectorised-batch-training

Jul 31, 2026

Commits on Jul 31, 2026

perf: contract the batch neighborhood by axis instead of per node
andremsouza
committed
00969d8
perf: find every best-matching unit with one matrix product
andremsouza
committed
663b013
perf: add an optional numba kernel behind a fast extra
andremsouza
committed
f8c86bf
bench: re-measure, and retire the benchmark whose subject no longer exists
andremsouza
committed
952ca2b
docs: give the derivations a home before the docstrings lose them
andremsouza
committed
3adac7d
docs: cut the narration from the docstrings and the manifest
andremsouza
committed
c5dec43
chore: release 0.7.0
andremsouza
committed
d5d9670
fix: install numba directly rather than through an extra
andremsouza
committed
166d043
docs: correct the pages that 0.7.0 made wrong, and finish the prose pass
andremsouza
committed
a0ceea7