strisunshinewentingwang/pytorch

Pull Requests Commits

quant: add `extra_repr` to HistogramObserver (#84760)

vkuzo committed 2 years ago

1dabb51a

flash_attention integration (#81434)

drisspg committed 2 years ago

0fc02dbb

Revert "Add __all__ for a few distributed modules plus a little typing (#84119)"

pytorchmergebot committed 2 years ago

219ff261

OpInfo: Prevent clamp sample inputs from sharing tensors (#84696)

peterbell10 committed 2 years ago

2614079f

[coreml][bug] coreml gpu flag not set (#84725)

mcr229 committed 2 years ago

5c0c8f2c

[QNNPack] Update GoogleTest SHA256 Hash (#84754)

salilsdesai committed 2 years ago

fe47e614

[Profiler] Make `RecordQueue` manage the lifetime of `PythonTracer`. (#83964)

Taylor Robie committed 2 years ago

daffff99

[Profiler][Trivial] Move `PythonTracerBase` to `torch/csrc/profiler/orchestration` (#83895)

Taylor Robie committed 2 years ago

32853870

test: adding uniform (#84292)

srossross committed 2 years ago

e8b95018

Make xnnpack based convs thread safe (#84602)

kimishpatel committed 2 years ago

a3855cc6

Use exclude_zero in i0e sample inputs function (#84453)

peterbell10 committed 2 years ago

2c4eaddb

Use presence of _symint in kernel name to generate symint sig or not (#84579)

seemethere committed 2 years ago

93aef3a0

[Mobile] Fix The Build For Model Tracer (#84755)

dhruvbird committed 2 years ago

18a31cc0

Revert "Convert NoopPyInterpreterVTable into a Meyer singleton (#84656)"

pytorchmergebot committed 2 years ago

52224139

Removing .github/scale-config.yml, now this repo is using the config in test-infra (#84753)

jeanschmidt committed 2 years ago

ac364f8b

[FSDP] Optimizer states may be on CPU, copy them to GPU before gathering (#84708)

fegin committed 2 years ago

28c830ac

Missed one CHECK_NOTNULL in #82032's find-replace (#84720)

cchan committed 2 years ago

0fd8f6b9

Add user facing documentation for CSAN (#84689)

sypneiwski committed 2 years ago

d12f3524

remove c10_defs.bzl and embed its logic directly where it is used (#84595)

Michael Andreas Dagitses committed 2 years ago

a8198a09

[quant][docs] Additonal fixes for quantize_fx docs (#84587)

jerryzh168 committed 2 years ago

214a6500

Set up aten/src/ATen/functorch directory; move some files there (#84648)

zou3519 committed 2 years ago

0a89bdf9

Add CSAN support for CPU synchronizations (#84428)

sypneiwski committed 2 years ago

8e57ce63

[functorch] Simplify BatchedTensorImpl (#84642)

zou3519 committed 2 years ago

7702ca49

[Mobile] Use -ffunction-sections and -fdata-sections for Mobile builds (#84704)

dhruvbird committed 2 years ago

0d46bfac

[Mobile] Update build_mobile.sh to allow lite interpreter and tracing based builds (#84647)

dhruvbird committed 2 years ago

747f27a9

[DataPipe] Fix mishandling of exception message when error is not iterable (#84676)

NivekT committed 2 years ago

27e5299e

[functorch] delete functorch/csrc/Constants.h (#84639)

zou3519 committed 2 years ago

df3377fb

ROCm support for test_lazy_init (#84333)

jataylo committed 2 years ago

09bcc006

Add synchronize hooks (#84427)

sypneiwski committed 2 years ago

67d6f716

Redo how custom/python_custom methods on TensorImpl work (#84641)

ezyang committed 2 years ago

591b75bf

Newer Older