strisunshinewentingwang/pytorch

Pull Requests Commits

[LTC] Update quickstart documentation

davidberard98 committed 2 years ago

61638447

Add sync after each overhead computation step to try to flush GPU ops; should also try fixing CPU/GPU clocks. not sure this helps..

wconstab committed 2 years ago

e6abf446

Add codegen support for avg_pool2d + backward (#69739)

wconstab committed 2 years ago

Verified 98d2dbc1

Ensure different output filenames for test/train results (#69784)

wconstab committed 2 years ago

Verified 7ec7a37f

Move internal ops under ts_backend

desertfire committed 2 years ago

f3229d64

Upstream several internal ops used in LazyTensor, Part 2

desertfire committed 2 years ago

d6224f24

Upstream several internal ops used in LazyTensor

desertfire committed 2 years ago

dd3e15a3

Clean up lazy_bench so it runs working models only by default (#69749)

wconstab committed 2 years ago

Verified 1b04924e

Upstream utils in computation_client, Part 2

desertfire committed 2 years ago

e62b77f8

Upstream utils in computation_client

desertfire committed 2 years ago

cd685f90

[LTC] Turn on LTC on TorchBench on-demand PR CI (#64863)

alanwaketan committed 2 years ago

Verified 1d8c2937

Remove unused header files in lazy_tensors/core/platform/

desertfire committed 2 years ago

4fdf8387

Use non-blocking copy for creation of lazy tensors in TS backend impl (#69397)

wconstab committed 2 years ago

Verified bd7db8bf

Add no-op execution mode to support profiling lazy trace overhead (#69513)

wconstab committed 2 years ago

Verified 85b8957b

Codegen pow.Tensor_Scalar and pow.Tensor_Tensor

desertfire committed 2 years ago

303dea85

Adopt upstreamed ltc_ops.h

desertfire committed 2 years ago

2de9902f

Adopt upstreamed tensor_util.h

desertfire committed 2 years ago

3e5fd4e4

Adopt upstreamed ir_dump_util.h

desertfire committed 2 years ago

c443e95f

Adopt upstreamed helpers.h

desertfire committed 2 years ago

ce0f3ef5

Adopt upstreamed LazyView and view ops IR

desertfire committed 2 years ago

41a4c16e

Improve efficiency of shape hash by not using tostring (#69404)

wconstab committed 2 years ago

Verified 767ce365

Deprecate CreateFrom method in LazyTensor

desertfire committed 2 years ago

e532862d

t push origin lazy_tensor_staging Merge branch 'binbao/merge2' into lazy_tensor_staging

desertfire committed 2 years ago

6046e049

Adopt updatestream changes after merging master

desertfire committed 2 years ago

51df95cd

Merge branch 'master' into binbao/merge2

desertfire committed 2 years ago

75f87bad

[LTC] Adopt the upstreamed c10::Device <=> BackendDevice helpers (#69411)

alanwaketan committed 2 years ago

Verified 52df08f8

[LTC] Adopt the upstreamed util.h (#69410)

alanwaketan committed 2 years ago

Verified a860996e

[Core ML] Avoid recompiling models when the OS version is not changed (#69438)

xta0 committed 2 years ago

3202028e

Revert D32780415: [Static Runtime] Move implementation details from impl.h into internal.h

d1jang committed 2 years ago

c97dc928

Revert D32743881: [Core ML] Avoid recompiling models when the OS version is not changed

suo committed 2 years ago

29a45f00

Older