strisunshinewentingwang/pytorch

Pull Requests Commits

move test_abs to device-generic tests (#36465)

ngimel committed 4 years ago

f3f640d4

[improve logging] add the param information when logging the optimizer engine (#36558)

Xing Wang committed 4 years ago

4b3e3d82

doc note on deterministic/non-deterministic gradient for min/max/median (#36481)

nikitaved committed 4 years ago

d3cf9452

DataLoader: properly diagnose exceeding file descriptor limit (#34768)

Baranowski committed 4 years ago

69e3ee2d

Revert D20147487: Refactor jit::Operator to more clearly distinguish the two possible states

philippv committed 4 years ago

ed2d1cb2

[caffe2] Add support for std::shared_ptr<std::vector<TensorList>> in PackRecordsOp and UnPackRecordsOp (#36550)

Hao Lu committed 4 years ago

fb70b4fb

Make dim, numel, element_size into prim ops (#36551)

smessmer committed 4 years ago

018c3420

Expunge TensorId from all DispatchKey names. (#36240)

ezyang committed 4 years ago

dd64e738

Update internal invariants in the world of manuallyBoxedKernel (#36388)

ezyang committed 4 years ago

8f501f30

[ROCm] Add debug flag (#36521)

iotamudelta committed 4 years ago

076d46f8

Lock optimizations for DistAutogradContainer. (#36529)

pritamdamania committed 4 years ago

6e7eaabf

Revert D20936595: Make DispatchKeyExtractor forget about TensorOptions

Karl Ostmo committed 4 years ago

411ccce2

[jit] tracer flag to guard risky behaivors (#36277)

wanchaol committed 4 years ago

999d7f6a

[TensorExpr] Postpone insertion of Alloc/Free statements in computeAt. (#36526)

Mikhail Zolotukhin committed 4 years ago

d5ba39c2

[caffe2] Enable copying for caffe2::Tensor (#36468)

Hao Lu committed 4 years ago

4d1ccafb

Fixing SyncBN dgrad (#36382)

jjsjann123 committed 4 years ago

289d52c1

Make DispatchKeyExtractor forget about TensorOptions (#36290)

Basil Hosmer committed 4 years ago

70b826a8

Refactor jit::Operator to more clearly distinguish the two possible states (#33905)

smessmer committed 4 years ago

36b273ab

Fix either::map naming (#33904)

smessmer committed 4 years ago

9fcb4ab3

Make FakeLowP tests work (#36525)

yinghai committed 4 years ago

eb00bac2

Fix a segfault in DeviceThreadHandlePool and PoolWindow (#36416)

VivekPanyam committed 4 years ago

8544591f

Output more debugging information for reduce kernel (#35946)

zasdfgbnm committed 4 years ago

c7631716

qnnpack hardswish - pytorch op integration (#36320)

vkuzo committed 4 years ago

1e227171

qnnpack hardswish - LUTs (#36252)

vkuzo committed 4 years ago

0964b662

[PyTorch Numeric Suite] Add weight compare API (#36186)

Haixin Liu committed 4 years ago

455d4aab

Fix linter warning: replace f-strings with str.format for Py2 compat (#35492)

Sudarshan Raghunathan committed 4 years ago

739351fa

Fix clang format (#36544)

malfet committed 4 years ago

501d9f33

Fix signed integer overflow in rng_test.h (#36421)

pbelevich committed 4 years ago

0b7e8323

Make patterns in test_unmatched_annotations more flexible (#36422)

malfet committed 4 years ago

fd008bd1

[TensorBoard] fix #36471 (#36495)

lanpa committed 4 years ago

1f40bddf

Newer Older