Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
Select Git revision
  • cupyactx
  • main default protected
2 results
Created with Raphaël 2.2.024Nov1514131214Oct116Sep45Aug429Jul15985127Jun724May2139Apr13Mar116Feb18Oct15817Sep14Aug30Jul2625232025Jun1926May985225Apr23422Mar19Feb11Jan25Dec241514Nov13631Oct26143227Sep26222019109Aug751Jul29Jun2726191411109820May122119Apr7517Mar163115Feb23Jan1613121127Dec26241625Nov241776229Oct272422201918171511524Sep2220430Aug29262416121121Jul2017161476130Jun2928272524222120181716151413121110111098742129May2827262221205mypy: ignore pytato errorspyproject: switch to hatchlingruff: fix re errorsfeat: improve dataclass containerMerge branch 'main' into cupyactxadd device<->host transfer mappers (#282)mypy: fix errors from pytato typing improvementsmypy: fix errors from pytato typing improvementsruff: fix zip strict argumentruff: fix type import errorspyproject: bump min python to 3.10PyOpenCL array context: don't fail if no transformfix: new mypy 1.12 errorsdocs: update link to __instancecheck__fix: add type ignore for mypy bugsimplify CI install slightlychange CI cupy integrationremove a few spurious changesCI fixesmake cupy import optionalruffrestore some testsupdate with current numpy actxMerge branch 'main' into cupyactxForbid setting `force_device_scalars=False` (#278)Add minimum, fmin (#281)Numpy actx: cache execuctorwith_container_arithmetic: Rename arguments to signal who broadcasts across whoNumpy actx: warn (not error) on no user-provided transformsFix a typo in the pytato actxFix ruff C409 failuresouter: disallow non-object numpy arraysSwitch to __array_ufunc__ in tests as a way to avoid numpy broadcastingRework dataclass array container arithmeticClarify that actx.array_types allows ABCsImprove, type, fix array_equal across all array contextsContainer serialization: iterable -> sequence, plus type aliasesDon't expect unflatten failure from numpy array for numpy actxSkip numpy conversion tests when using the numpy actxSkip tagging test for numpy actx