strisunshinewentingwang/pytorch

Pull Requests Commits

Warn on modification of OptimizedModule.forward (#95673)

wconstab committed 1 year ago

dc10ab15

Warn on dynamo OptimizedModule.forward() (#95672)

wconstab committed 1 year ago

6bdef7a5

Add support for Inductor + symbolic shapes + training (#93059)

ezyang committed 1 year ago

20dfce59

[jit] Add c++ stacktraces for jit::ErrorReport (#94842)

davidberard98 committed 1 year ago

70029214

Run tests in USE_PYTEST_LIST through run_tests (#95659)

clee2000 committed 1 year ago

e3c5c369

Rephrase zero_grad docs (#95643)

janeyx99 committed 1 year ago

e5b9d987

Build Triton in Docker image (#95233)

huydhn committed 1 year ago

ba43d908

Fix trymerge changed files count (#95720)

huydhn committed 1 year ago

b55d0d2a

Enable thp(transparent huge pages) for buffer sizes >=2MB (#93888)

snadampal committed 1 year ago

2cc845eb

[ao][fx] Enable observed -> quantized float for static quantized MultiheadAttention (#95636)

kev-zheng committed 1 year ago

f1dbfe2f

Clean up unused `fill_` sample inputs (#95117)

qqaatw committed 1 year ago

fafb4109

Add missing f-string specifiers (#95707)

ezyang committed 1 year ago

835122c8

Add standalone torch._inductor.compile() API (#95594)

jansel committed 1 year ago

e13b8041

[quant][pt2e] Add support for dynamic quantization with symmetric quant for input (#94854)

JacobSzwejbka committed 1 year ago

fc324d34

[dynamo] avoid truncation of python pointers (#95619)

cyyever committed 1 year ago

f8ad64d5

[dtensor][BE] remove redundant tests (#94838)

wanchaol committed 1 year ago

1e15a272

[dtensor] support creating DTensor in submesh (#95458)

wanchaol committed 1 year ago

2a1cb964

[dtensor] refactor get_coordiniate (#95457)

wanchaol committed 1 year ago

261eb46d

[dtensor] use tracing for metadata prop (#95456)

wanchaol committed 1 year ago

bb9a05b1

Enabling FlashAttention for SDPA when given NestedTensor (#95438)

drisspg committed 1 year ago

80614783

Update skip message to reflect why test is being skipped (#95127)

Michael Gschwind committed 1 year ago

57f2c588

MHA torch.jit.script fix for in_proj_weight = None (#95653)

Michael Gschwind committed 1 year ago

4fada6eb

Add dynamo graph break stats to CI (#95635)

wconstab committed 1 year ago

1a727126

[MPS] Add pow.Scalar (#95201)

qqaatw committed 1 year ago

f33180fb

Add prelu into Autocast CPU whitelist (#95366)

yanbing-j committed 1 year ago

71ad1005

Reland #94719 - Update ideep to add primitive cache for ARM (#95688)

yanbing-j committed 1 year ago

b87229f1

[MTA] Skip size-0 tensors in `multi_tensor_apply` (#94655)

awgu committed 1 year ago

05943712

[MPS] Add copysign op. (#95552)

qqaatw committed 1 year ago

9e16f128

[MPS] Fix type casting copy with storage offset (#95573)

qqaatw committed 1 year ago

b7c2a651

[pt] add share_memory_ to aten TensorBase (#95557)

xunnanxu committed 1 year ago

7c66333c

Newer Older