strisunshinewentingwang/pytorch

Pull Requests Commits

[dynamo][distributed] handle _rank_not_in_group, _get_or_create_default_group (#119628)

jansel committed 216 days ago

f2cf0768

[pytree] add function `tree_iter` (#120155)

XuehaiPan committed 217 days ago

372d078f

[nit][DTensor][Test] Update test name to reflect the actual test (#118960)

wz337 committed 217 days ago

61a3a762

add register_device_op_overrides (#119268)

wangjiangben-hw committed 217 days ago

2864a7e1

[executorch hash update] update the pinned executorch hash (#120165)

pytorchupdatebot committed 217 days ago

70bc3b3b

[inductor] Always allow 64 bit in next_power_of_2 (#120164)

jansel committed 217 days ago

d74bdd50

[cuDNN] Bump cuDNN frontend submodule to 1.1.1 (#120137)

eqy committed 217 days ago

de15781a

[dynamo] Use EQUALS_MATCH guard for mod.training (#120147)

anijain2305 committed 217 days ago

b642a18e

[dynamo][refactor] Use existing helper functions for CLOSURE_MATCH (#120145)

anijain2305 committed 217 days ago

0b11b0ed

enhance next_power_of_2 function (#120153)

wangjiangben-hw committed 217 days ago

0c972c7c

[dynamo] Refactor reconstruct() not to return anything (#120150)

jansel committed 217 days ago

2fea4752

[dynamo][refactor] Use TYPE_MATCH instead of manually constructing guard (#120140)

anijain2305 committed 217 days ago

757fc663

[dynamo][guards] Use EQUALS_MATCH for NAME_MATCH (#120132)

anijain2305 committed 217 days ago

48d96c08

Remove unused c10/util/C++17.h inclusion and outdated checks (#120149)

cyyever committed 217 days ago

a9953a5e

[inductor] allow padding mm/bmm/addmm in the presence of dynamic dims (#120073)

chenyang78 committed 218 days ago

fac598c4

Fix skip for test_set_nccl_pg_timeout (#120130)

tinglvv committed 218 days ago

2f8a80ec

[inductor] Add torch.cond support to JIT Inductor (#119759)

aakhundov committed 218 days ago

badf84bd

[c10d] remove one line of verbose log (#120138)

shuqiangzhang committed 218 days ago

30000aa3

[AOTI] Fix a typo (#120094)

desertfire committed 218 days ago

fa0e3956

[executorch hash update] update the pinned executorch hash (#120134)

pytorchupdatebot committed 218 days ago

0a7471e0

[export] turn on replace_set_grad_with_hop_pass in pre_dispatch (#119915)

ydwu4 committed 218 days ago

ac2ba788

[export] manuually create test cases for split and inline (#119914)

ydwu4 committed 218 days ago

73763026

[export] make node_inline_ also inline the get_item calls (#119913)

ydwu4 committed 218 days ago

8d81e61f

[export] add replace_set_grad_with_hop_pass (#119810)

ydwu4 committed 218 days ago

812f05d7

[export] add node_inline_ to prepare replacing set_grad_enabled with hop (#119736)

ydwu4 committed 218 days ago

4769e691

[export] add sequential_split to prepare replacing set_grad_enabled with hop (#119732)

ydwu4 committed 218 days ago

068659dd

tiny improvement to the cprofile wrapper (#120100)

shunting314 committed 218 days ago

becfda00

[inductor] logging meta data for inductor generated triton kernel (#120048)

shunting314 committed 218 days ago

36e118b8

Add quantized gelu (#119935)

andre-meta committed 218 days ago

24968ff0

[Memory Snapshot] Add CUDAAllocatorConfig details into snapshot metadata (#119404)

aaronenyeshi committed 218 days ago

7973ac58

Newer Older