strisunshinewentingwang/pytorch

Pull Requests Commits

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

lly-zero-one committed 4 years ago

20f52cdd

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

malfet committed 4 years ago

81bb19c9

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

IvanKobzarev committed 4 years ago

9a31eee1

CUDA BFloat16 addmm, addmv (#44986)

zasdfgbnm committed 4 years ago

dfb8f2d5

CUDA BFloat16 unary ops part 1 (#44813)

zasdfgbnm committed 4 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 4 years ago

1cab27d4

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

James Reed committed 4 years ago

c941dd34

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

nikitaved committed 4 years ago

9dc2bcdc

add .cache to gitignore (#45017)

suo committed 4 years ago

7118d537

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

z-a-f committed 4 years ago

1a580c10

nightly robustness fixes for linking across devices (#43771)

scopatz committed 4 years ago

7de512ce

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

suo committed 4 years ago

42af2c79

Add alias dispatch key Math. (#44354)

Ailing Zhang committed 4 years ago

92f8f75c

Update submodule gloo (#45025)

gunandrose4u committed 4 years ago

acc2a1e5

fix compile error (#45052)

gunandrose4u committed 4 years ago

a4aba1d4

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

beauby committed 4 years ago

ac8c7c4e

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

nickgg committed 4 years ago

4bbb6adf

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

Rong Rong committed 4 years ago

e2f49c84

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

gchanan committed 4 years ago

a6895d43

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

kauterry committed 4 years ago

48103655

Complex gradcheck logic (#43208)

anjali411 committed 4 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

CUDA BFloat16 layernorm (#45002)

zasdfgbnm committed 4 years ago

7ecfaef7

histogram observer: ensure buffer shape consistency (#44956)

vkuzo committed 4 years ago

2163d310

[pytorch][tensorexpr] Make gtest-style macros in tests match actual gtest signatures (#44861)

bertmaher committed 4 years ago

0714c003

[pytorch] clean up normalized_dynamic_type() hack (#44889)

ljk53 committed 4 years ago

9e5045e9

Add cusolver to build, rewrite MAGMA inverse with cusolver (#42403)

xwang233 committed 4 years ago

d75c4027

Newer Older