strisunshinewentingwang/pytorch

Pull Requests Commits

Improve warnings to actually point at user code (#39143)

albanD committed 4 years ago

d6715e63

[TensorPipe] Use PrefixStore to avoid conflicting keys (#39185)

lw committed 4 years ago

d1212e58

[TensorPipe] Bind to hostname's IP address instead of localhost (#39184)

lw committed 4 years ago

99f6df3c

[TensorPipe] Don't use separate heap allocation for metrics (#39183)

lw committed 4 years ago

3ac0ec3d

[TensorPipe] Ignore expected errors (#39182)

lw committed 4 years ago

587d453b

Update TensorPipe submodule (#39189)

lw committed 4 years ago

debb7ba6

[JIT] Make new zip serialization for torch save/load significantly (~70%) faster (#38379)

voznesenskym committed 4 years ago

fce01a9b

[PyTorch] Fix operator perf observer index issue.

xcheng16 committed 4 years ago

b08a4aaf

Change __CUDACC__ and __HIPCC__ to __CUDA_ARCH__ and __HIP_ARCH__ in NumericUtils.h (#39213)

pbelevich committed 4 years ago

d0650af2

[caffe2] Fix the correctness check for GivenTensorFill operator

Dmytro Dzhulgakov committed 4 years ago

23318532

Remove sumall from TH, THC, THCUNN (#39042)

zasdfgbnm committed 4 years ago

2f497573

Regenerate config.yml (#39215)

malfet committed 4 years ago

ca6579bd

[Reland] add xenial + cuda 9.2 + gcc 5.4 CI test (#39036)

lixinyu committed 4 years ago

a04fb2ab

Fix argmin/max bug (#38946)

ShawnZhong committed 4 years ago

f7a8851e

fused convbn: respect the strict argument when loading from state dict (#39205)

vkuzo committed 4 years ago

527ee63b

Migrate AT_DISPATCH_FLOATING_AND_COMPLEX_TYPES_AND1 to c10::complex (#39045)

zasdfgbnm committed 4 years ago

98a755bc

test_bottleneck_cuda works on ROCm 3.3 (#38249)

jithunnair-amd committed 4 years ago

41363b29

Add timeout to TestBottleneck (#39191)

malfet committed 4 years ago

0e8c65f7

fix asserts in cuda code (#39047)

ngimel committed 4 years ago

9c19a129

[dper][pruning] add histogram op (#38514)

hwangjeff committed 4 years ago

0b9d5370

[vulkan] jni build support USE_VULKAN (#39188)

IvanKobzarev committed 4 years ago

928e99b9

Moves angle/abs test to test_torch (#39154)

Mike Ruberry committed 4 years ago

ee3bd104

.cirlceci: Remove setup job (#39081)

seemethere committed 4 years ago

7b343cc3

short-circuit pow for complex 1 and 0 exponents (#39117)

nairbv committed 4 years ago

feaf7208

Stops cross-device data movement in tensor iterator (#38998)

Mike Ruberry committed 4 years ago

5e975cf8

Fixup: rename BatchedTensorKey to Batched (#38798)

zou3519 committed 4 years ago

d26f7f09

Make collect_env more robust on Windows (#39136)

peterjc123 committed 4 years ago

e029d678

Revert D21748644: [pytorch][PR] Fix index overflow in ConvTranspose3d

rohan-varma committed 4 years ago

5267b17a

implement dynamic bucket order in DDP (#35137)

zhaojuanmao committed 4 years ago

b98948e6

Revert D21729544: add overload name for op eq.str

rohan-varma committed 4 years ago

7e1cc2da

Older