chenyu 1692087db5 _one_hot_along_dim input needs to be int (#9179) ... * _one_hot_along_dim input needs to be int indexing and onehot compare with arange, and non-int dtype is likely a bug		2025-02-20 09:00:43 -05:00
..
codegen	rename can_pad arg to cache [pr] (#9170)	2025-02-19 12:24:59 +01:00
engine	hotfix: add output buffer back to kernel parents + comment [pr] (#9174)	2025-02-19 14:22:01 +01:00
nn	Revert "match torch rmsnorm implementation (#6799)" (#9052)	2025-02-13 14:42:45 +08:00
renderer	Only add enable f16 directive if ShaderF16 is supported (#9163)	2025-02-19 17:20:03 +08:00
runtime	cuda hooking (#9180)	2025-02-20 19:20:01 +08:00
shape	symbolic gets its own file [pr] (#9132)	2025-02-17 18:55:21 +08:00
viz	correctly step through bottom_up_rewrites in viz [pr] (#9176)	2025-02-19 19:20:57 +01:00
__init__.py	UOp.variable (#7010)	2024-10-12 18:20:44 +08:00
device.py	fix win32 CPUProgram missing cache flush (#9171)	2025-02-19 21:38:51 +08:00
dtype.py	add truncate_bf16 (#9078)	2025-02-15 07:59:09 +08:00
gradient.py	fix pow backward when base is 0 (#9075)	2025-02-13 21:06:01 -05:00
helpers.py	Use proper units when printing beam time (#9103)	2025-02-17 23:41:38 +08:00
ops.py	rename can_pad arg to cache [pr] (#9170)	2025-02-19 12:24:59 +01:00
py.typed	add a single py.typed (#6083)	2024-08-14 17:31:46 -07:00
spec.py	UNIQUE UOp for buffer instead of arg (#9156)	2025-02-18 16:59:59 +08:00
tensor.py	_one_hot_along_dim input needs to be int (#9179)	2025-02-20 09:00:43 -05:00