strisunshinewentingwang/pytorch

Pull Requests Commits

Removing cublas static linking (#79280)

atalman committed 2 years ago

0e25a949

[CUDA graphs] Allows Adam and AdamW to be capture-safe (#77862)

mcarilli committed 2 years ago

ba27ee9e

[meta] Add meta support for fft ops (#79311)

kshitij12345 committed 2 years ago

a732bbea

[meta] diag ops, trace (#79341)

kshitij12345 committed 2 years ago

bd1a35df

[ci] fix env var propagation

suo committed 2 years ago

aa681e40

Put symint overloads on a different name

ezyang committed 2 years ago

213a8fc9

[transformer] BT enablement on fairseq - pytorch change (#79186)

Wei Wei committed 2 years ago

f23685f0

move tree tests to the start of test_profiler.py

Taylor Robie committed 2 years ago

c321dfe1

Fix `shard_module` to appropriately deal with sub process groups.

pritamdamania87 committed 2 years ago

a81be444

Revert "[JIT] Propagate profiled information to DifferentiableGraph outputs"

pytorchmergebot committed 2 years ago

5413580f

Revert "Fixes maybe_broadcast to actually broadcast only when needed (#79298)"

pytorchmergebot committed 2 years ago

66460c4a

Fixes maybe_broadcast to actually broadcast only when needed (#79298)

Mike Ruberry committed 2 years ago

1cb1c2c0

[lint] autoformat test/cpp and torch/csrc

suo committed 2 years ago

30fb2c4a

Add offsets-based reduction to segment_reduce (CPU, CUDA)

mikaylagawarecki committed 2 years ago

1ec30a66

[ci] remove IN_CI env var

suo committed 2 years ago

c978b609

[ci] fix GITHUB_ACTIONS env var checks

suo committed 2 years ago

f51d5233

[ci] do not pass GITHUB_RUN_ID explicitly to docker image

suo committed 2 years ago

a8bafd26

Update ideep for ideep conv changes -v2 (#79258)

yanbing-j committed 2 years ago

2edcc9e6

masked logasumexp/logaddexp

george-qi committed 2 years ago

164029f7

Update and improve the heuristics for linalg.lu_solve

lezcano committed 2 years ago

9fc2518a

Expose lu_factor_batched_cublas

lezcano committed 2 years ago

1880d293

Simplify and optimize linalg.solve

lezcano committed 2 years ago

54949a5a

[Static Runtime] Exception handling during fork subgraph execution (#79292)

akshayParashar1995 committed 2 years ago

65a37923

turn on -Werror=unused-variable in our Bazel CPU build

Michael Andreas Dagitses committed 2 years ago

ab2ca95d

Support multi-dimensional lengths in segment_reduce to support pytorch_scatter.segment_* functionalities (CUDA)

mikaylagawarecki committed 2 years ago

e727539c

Autogen Tags enum, and allow specifying tags while defining an op

anjali411 committed 2 years ago

38350acf

update is_same_size to work with nested tensor dispatch

drisspg committed 2 years ago

bdcee8f9

drisspg committed 2 years ago

10dfdf20

qcat: use direct memcpy when all the inputs and output share the same scale and zero_point

mingfeima committed 2 years ago

50e13012

Remove linux-xenial-py3.7-gcc7-no-ops from distributed checks (#79320)

kit1980 committed 2 years ago

77d5eb7d

Newer Older