tinygrad/tinygrad
chenyu ca68037f26
lazy basic setitem to unrealized Tensor (#14756)
undo the view and make it a mask, this fuses the setitem with any pending compute too.

one behavior change is that for target not backed by a buffer (const and arange), rangeify makes output contiguous under the hood.
this is stricter better than raise and ask user to call contiguous, as that would no longer be fuse-able.
2026-02-14 20:27:03 -05:00
..
apps recursive resolve assign dependency (#14688) 2026-02-11 17:41:05 -05:00
codegen remove llvm requirement from amd (#14717) 2026-02-13 10:50:12 +08:00
engine fix write-after-read tracking (#14754) 2026-02-14 17:23:05 -05:00
mixin return types for all math.py function (#14413) 2026-01-28 20:10:11 -05:00
nn Revert "use zip_extract and tar_extract in torch load (#14734)" (#14745) 2026-02-14 13:24:01 +08:00
renderer clean up amd/elf.py (#14741) 2026-02-14 12:09:05 +08:00
runtime fix devices for copies (#14747) 2026-02-14 17:39:41 +03:00
schedule lazy basic setitem to unrealized Tensor (#14756) 2026-02-14 20:27:03 -05:00
uop remove llvm requirement from amd (#14717) 2026-02-13 10:50:12 +08:00
viz viz: wave color is locally scoped (#14728) 2026-02-13 18:22:20 +09:00
__init__.py move files into uop dir (#10399) 2025-05-18 11:38:28 -07:00
device.py everything has dtype.long now (#14661) 2026-02-10 15:08:50 +08:00
dtype.py support bitcast in sym_infer (#14708) 2026-02-12 10:21:05 -05:00
gradient.py remove CUSTOM_KERNEL / directly construct it (#14604) 2026-02-08 18:43:33 +08:00
helpers.py fix limit_bufs to not limit index (#14751) 2026-02-14 16:00:03 -05:00
py.typed add a single py.typed (#6083) 2024-08-14 17:31:46 -07:00
tensor.py lazy basic setitem to unrealized Tensor (#14756) 2026-02-14 20:27:03 -05:00