chenyuandGitHub 8a8aed23d2 test dtypes of return values of cumsum, argmax/min, multinomial (#2933) ...

* test dtypes of return values of cumsum, argmax/min, multinomial

cumsum behaves like sum, and functions that return an index return in dtypes.default_int

* because webgpu is different

2023-12-25 11:33:17 -05:00

..

codegen

green linearizer asserts for ops (#2800)

2023-12-25 10:41:54 -05:00

features

add CMPEQ (#2931)

2023-12-25 00:15:55 -05:00

nn

remove realize from optimizer (#2880)

2023-12-20 16:42:41 -08:00

renderer

add CMPEQ (#2931)

2023-12-25 00:15:55 -05:00

runtime

add CMPEQ (#2931)

2023-12-25 00:15:55 -05:00

shape

apply the multitensor optimizations in lazy.py (#2901)

2023-12-21 13:55:49 -08:00

__init__.py

move device to device.py (#2466)

2023-11-27 11:34:37 -08:00

device.py

Fix double exception in __del__() when __init__() raises exception. (#2738)

2023-12-13 15:46:11 -08:00

graph.py

catch cycles in print_tree (#2891)

2023-12-21 18:40:37 -08:00

helpers.py

move graph to runtime, check line count with sz.py (#2842)

2023-12-18 20:30:06 -08:00

jit.py

new lazy, benchmark (#2878)

2023-12-20 14:33:21 -08:00

lazy.py

add CMPEQ (#2931)

2023-12-25 00:15:55 -05:00

mlops.py

add CMPEQ (#2931)

2023-12-25 00:15:55 -05:00

ops.py

add CMPEQ (#2931)

2023-12-25 00:15:55 -05:00

realize.py

get_lazyops() -> lazyops (#2884)

2023-12-20 18:04:49 -08:00

tensor.py

test dtypes of return values of cumsum, argmax/min, multinomial (#2933)

2023-12-25 11:33:17 -05:00