strisunshinewentingwang/pytorch

Pull Requests Commits

[vulkan] support dimensions negative indexing (#45068)

IvanKobzarev committed 3 years ago

dfc88d4f

[vulkan] reshape op to use infer_size to expand -1 (#45104)

IvanKobzarev committed 3 years ago

5621ba87

[WIP] Add vec256 test to linux CI (#44912)

lixinyu committed 3 years ago

8968030f

Vectorize int8_t on CPU (#44759)

xuhdev committed 3 years ago

4b3046ed

Change typo 'momemtum' to 'momentum' (#45045)

ylsung committed 3 years ago

f77ba0e4

[hpc]optimize the torch.cat cuda kernel (#44833)

lly-zero-one committed 3 years ago

20f52cdd

[JIT] Prohibit subscripted assignments for tuple types (#44929)

malfet committed 3 years ago

81bb19c9

[vulkan] Remove duplication of op registration and clean unused vars (#44932)

IvanKobzarev committed 3 years ago

9a31eee1

CUDA BFloat16 addmm, addmv (#44986)

zasdfgbnm committed 3 years ago

dfb8f2d5

CUDA BFloat16 unary ops part 1 (#44813)

zasdfgbnm committed 3 years ago

581a3644

Add a torch.hub.load_local() function that can load models from any local directory with a hubconf.py (#44204)

AdeelH committed 3 years ago

1cab27d4

[FX] s/get_param/get_attr/ (#45000)

James Reed committed 3 years ago

c941dd34

Introducing (Const)StridedRandomAccessor + CompositeRandomAccessor + migrate `sort` to ATen (CPU) (#39744)

nikitaved committed 3 years ago

9dc2bcdc

add .cache to gitignore (#45017)

suo committed 3 years ago

7118d537

Adding test to quantized copy for 'from float' (#43681)

z-a-f committed 3 years ago

1a580c10

nightly robustness fixes for linking across devices (#43771)

scopatz committed 3 years ago

7de512ce

[jit] gtest-ify test_alias_analysis.cpp (#45018)

suo committed 3 years ago

42af2c79

Add alias dispatch key Math. (#44354)

Ailing Zhang committed 3 years ago

92f8f75c

Update submodule gloo (#45025)

gunandrose4u committed 3 years ago

acc2a1e5

fix compile error (#45052)

gunandrose4u committed 3 years ago

a4aba1d4

Make Channel API accept buffer structs rather than raw pointers. (#45014)

beauby committed 3 years ago

ac8c7c4e

[NNC] fix SyncThreads insertion and reenable CudaSharedMem test (#44909)

nickgg committed 3 years ago

4bbb6adf

skip im2col & vol2col in cpu/cuda convolution methods (#44600)

Rong Rong committed 3 years ago

e2f49c84

Turn on gradgrad check for BCELoss Criterion Tests. (#44894)

gchanan committed 3 years ago

a6895d43

Enabled torch.testing._internal.jit_utils.* typechecking. (#44985)

kauterry committed 3 years ago

48103655

Complex gradcheck logic (#43208)

anjali411 committed 3 years ago

9f67176b

Add one dimensional FFTs to torch.fft namespace (#43011)

peterbell10 committed 4 years ago

da7863f4

For logical tests, use the dtypes decorator (#42483)

xuhdev committed 4 years ago

49db7b59

Adds multiply and divide aliases (#44463)

Mike Ruberry committed 4 years ago

60709ad1

CUDA BFloat Pooling (#44836)

zasdfgbnm committed 4 years ago

faef89c8

Newer Older