strisunshinewentingwang/pytorch

Pull Requests Commits

[tgif ts][multiforward] allow codegen to generate different func name (#111446)

842974287 committed 333 days ago

f5714b1c

[MPS] Skip virtualized devices (#111576)

malfet committed 333 days ago

ca5f6f7a

[ez] Remove unused code in upload_test_stats (#111504)

clee2000 committed 333 days ago

0617f7fa

[Autograd] Track when mutations are for triton kernels (#111500)

oulgen committed 333 days ago

4e310fd8

Allow SymInt to specialize to FLOAT (#111219)

ezyang committed 333 days ago

971f67c9

Force specialization on INT_LIST (#111216)

ezyang committed 333 days ago

40c44c23

[vmap] symintify : is_same_size and split_with_sizes (#111491)

kshitij12345 committed 333 days ago

aa3243bc

[tp] fix torch compile regression (#111521)

wanchaol committed 333 days ago

03e28bde

[DOCS][CUDA] Update TF32 docs for sm90 (#111337)

eqy committed 333 days ago

894b9957

Fix: perverse input's NaN values to prevent undefined behavior for `matrix_exp` function (#111539)

scruel committed 333 days ago

503f44fb

Allow optimizer state conversion to accommodate optimizers that have no tensor state (e.g. SGD) (#111501)

speediedan committed 333 days ago

90e2117a

[cuda] Preserve operations order between vectorized and non-vectorized in ln grad input (#111488)

valentinandrei committed 333 days ago

5ce2ab84

[caffe2] replace numpy.object with object (#111494)

Igor Sugak committed 333 days ago

b2b5f137

[ONNX] Benchmark to store test data along exported model (#111095)

BowenBao committed 333 days ago

e3463fe4

Introduce is_big_gpu condition for test_max_autotune (#111467)

jataylo committed 333 days ago

71d7173a

[BE] Clean up trymerge code handling broken trunk failures (#111520)

huydhn committed 333 days ago

4ec777e9

Mark more decomp tests as slow (#111524)

malfet committed 333 days ago

4f0cf1e1

[ProcessGroupNCCL] Avoid recording stream for synchronous ops (#111431)

kwen2501 committed 333 days ago

18cc8a92

Bump urllib3 from 2.0.6 to 2.0.7 in /tools/build/bazel (#111435)

dependabot[bot] committed 333 days ago

Verified a7883ee4

Fix redundant asserts (#111445)

tugsbayasgalan committed 333 days ago

547a116f

More NT subclass op support for SAM (#111253)

jbschlosser committed 333 days ago

ba2ba962

[Reland] Add a workflow to release Android binaries (#110976)

huydhn committed 333 days ago

53c1dca6

Update the magma to version 2.7.2 (#111442)

pruthvistony committed 333 days ago

a771fde8

[ci] Move step to get workflow job id before test step in linux (#111483)

clee2000 committed 333 days ago

102fbd40

Revert " [1/N] Apply clang-tidy to c10 cuda files (#111137)"

pytorchmergebot committed 333 days ago

9c7391ea

Add ciflow/rocm label to run ROCm jobs (#111394)

huydhn committed 333 days ago

7fabb73d

Skip `test_quick_core_backward_baddbmm_cuda_float64` (#111493)

malfet committed 333 days ago

16cb3bdd

[CODEMOD][caffe2] replace numpy.bool with bool (#111432)

Igor Sugak committed 334 days ago

93e5065b

[ROCm] Bump kineto submodule commit to clear kineto cache to avoid memory leaks (#110849)

ramcherukuri committed 334 days ago

fa995626

int4 mm kernel enhancement (#111460)

yanboliang committed 334 days ago

256a5ff4

Older