strisunshinewentingwang/pytorch

Pull Requests Commits

Unify libtorch_python_cuda_core_sources filelists between CMakeList, fbcode and bazel (#40554)

scintiller committed 4 years ago

461014d5

Use CPU Allocator for reading from zip container

dreiss committed 4 years ago

7369dc8d

Disallow passing functions that don't return Tensors to vmap (#40518)

zou3519 committed 4 years ago

c362138f

Add batching rule for Tensor.permute (#40517)

zou3519 committed 4 years ago

43757ea9

Add batching rule for unsqueeze, squeeze, and transpose (#40455)

zou3519 committed 4 years ago

7038579c

doc string fix for torch.cuda.set_rng_state_all (#40544)

xwang233 committed 4 years ago

88ea51c0

[quant] Fix fuse linear pass (#40549)

jerryzh168 committed 4 years ago

e440c370

caffe2 | Fix building with `-Wrange-loop-analysis` on

mattjgalloway committed 4 years ago

eae1ed99

Allow ReflectionPad to accept 0-dim batch sizes. (#39231)

v0dro committed 4 years ago

cf8a9b50

Adjust CUDA memory leak test (#40504)

Ilia Cherniavskii committed 4 years ago

82e9318a

Revert D22208758: [pytorch][PR] Report error when ATEN_THEADING is OMP and USE_OPENMP is turned off.

pbelevich committed 4 years ago

85b87df5

move __range_length and __derive_index to lite interpreter (#40533)

jiatongzhou committed 4 years ago

06debf63

Fix backup solution (#40515)

seemethere committed 4 years ago

adcd755e

Add missing file to BUILD.bazel (#40536)

malfet committed 4 years ago

e12f73ee

Use tree-based sum for floats to avoid numerical instability (#39516)

peterbell10 committed 4 years ago

3dcc3297

Release GIL during DDP construction. (#40495)

pritamdamania committed 4 years ago

ea06db94

Update FP16 to FP16:4dfe081cf6bcd15db339cf2680b9281b8451eeb3. (#40526)

Ashkan Aliabadi committed 4 years ago

71edd7f1

Add C++-only `int dim` overloads to `std`-related operations (#40451)

peterbell10 committed 4 years ago

16f276ce

Update cpuinfo to cpuinfo:63b254577ed77a8004a9be6ac707f3dccc4e1fd9. (#40516)

Ashkan Aliabadi committed 4 years ago

a208a272

Unify `torch/csrc/cuda/shared/cudnn.cpp` include path (#40525)

malfet committed 4 years ago

c120fdc0

Update FXdiv to FXdiv:b408327ac2a15ec3e43352421954f5b1967701d1. (#40520)

Ashkan Aliabadi committed 4 years ago

cef35e33

Update psimd to psimd:072586a71b55b7f8c584153d223e95687148a900. (#40522)

Ashkan Aliabadi committed 4 years ago

4a0ba62d

[jit] allow dict to be mixed between tracing and scripting (#39601)

wanchaol committed 4 years ago

3e09268c

[jit] fix dictConstruct order issue (#40424)

wanchaol committed 4 years ago

787e1c4c

Update docs feature classifications (#39966)

Jessica Lin committed 4 years ago

2e6e8d55

Migrate equal from the TH to Aten (CPU) (#33286)

xcnick committed 4 years ago

72f2c479

Skip test_mem_leak on Windows (#40486)

peterjc123 committed 4 years ago

4d549077

Add finishAndThrow function to ProcessGroup::Work, and use with Gloo (#40405)

osalpekar committed 4 years ago

0c923eea

[NCCL Docs] Adding Comments for Work-level Finish in ProcessGroup (#40404)

osalpekar committed 4 years ago

3e2d2fc8

[NCCL] Explicitly Abort NCCL Communicators on Process Group Destruction (#40241)

osalpekar committed 4 years ago

527ab134

Newer Older