strisunshinewentingwang/pytorch

Pull Requests Commits

Redundant schema registration Prevention for Manually Boxed Wrappers (#38588)

mjavanmard committed 4 years ago

ddfd720e

override gcc version in cuda related test (#38675)

lixinyu committed 4 years ago

5e55f080

handle grad with `stride=0` on GPU MvBackward (#38321)

kshitij12345 committed 4 years ago

fc19747d

[quant] Remove get_qparams in Observers (#38435)

jerryzh168 committed 4 years ago

86397f6b

[quant][graphmode] Move processing code to prepare_script (#38669)

jerryzh168 committed 4 years ago

d5461e7a

organize verbatim sources with subdirectories (#38688)

kostmo committed 4 years ago

91163add

Do not use "buffer" in reentrant autograd err msg (#38625)

ssnl committed 4 years ago

f184ec81

Fix memory usage increase reported in #38568 (#38674)

jjsjann123 committed 4 years ago

958313a7

[CUDA] torch.roll for complex dtypes (#38664)

anjali411 committed 4 years ago

f3048609

Profiler: Call `populate_cpu_children` inside `__str__` and fix typo (#37816)

ShawnZhong committed 4 years ago

724b2b6e

[JIT][to_backend] Move code that is not related to the user-facing API out of `jit/backends/backend.h` (#38567)

jansel committed 4 years ago

49d687f2

Docker constants refactor (#38676)

malfet committed 4 years ago

76fc9bd2

[TensorExpr] Eager reduction initialization & removal from ReduceOp (#38585)

nickgg committed 4 years ago

2f21dfb5

[quant] Remove TensorListObserver (#38584)

supriyar committed 4 years ago

97abed7c

Updating submodules

svcscm committed 4 years ago

c430b7d8

Error out of default_collate for lists of unequal size (#38492)

Donna Choi committed 4 years ago

8c07a98a

FakeLowP Readme update (#38666)

venkatacrc committed 4 years ago

d7e08b45

Make find_first_set works on x86 MSVC (#38637)

peterjc123 committed 4 years ago

378956b4

towards guard what you use (#38576)

Krovatkin committed 4 years ago

23207ae6

[Distributed Autograd] Make debugInfoMap from strings to ints (#38416)

osalpekar committed 4 years ago

5fcb2f67

RIP CUDA <9.2: circleci, aten, and caffe2 (#36846)

zasdfgbnm committed 4 years ago

5e2d8745

[TensorExpr] Use couldMoveBefore instead of couldMoveAfter checks in the fuser pass, add CPP tests. (#38592)

Mikhail Zolotukhin committed 4 years ago

b29e7f9b

Complete codegen of 'build' workflow YAML tree (#38631)

kostmo committed 4 years ago

895479e6

[PyTorch] Remove module and operator observer macros. (#38489)

xcheng16 committed 4 years ago

262f70c9

[ONNX]Update Dropout Export (#37641)

KsenijaS committed 4 years ago

5b12c29b

Vectorize non-persistent Softmax (#38557)

slayton58 committed 4 years ago

59d92e44

[JIT] Export all jit/backend headers in BUILD.bazel (#38668)

Meghan Lele committed 4 years ago

b2c06ad8

Enabled dropout removal pass in mobile optimizer. (#38254)

kimishpatel committed 4 years ago

eb224721

support complex types for tanh_backward_cpu (#37791)

kshitij12345 committed 4 years ago

09c430a2

Handling Active Call Count through Future Callback (#38589)

osalpekar committed 4 years ago

a84fd8de

Newer Older