strisunshinewentingwang/pytorch

Pull Requests Commits

[ROCm] remove use of HIP_PLATFORM (#47241)

jeffdaily committed 3 years ago

9e58c85d

Moving test order to rebalance test1 and test2 times (#47290)

janeyx99 committed 3 years ago

579cfc66

Delete CUDA build rules from MacOS build (#47277)

malfet committed 3 years ago

5c8896f8

Fix return-type-is-always-copy warning (#47279)

malfet committed 3 years ago

c05ee86e

Format error message for unmatched signature between _out and base functions (#47087)

ejguan committed 3 years ago

a341a432

[GPU] Enable optimize_for_metal in fbcode (#47102)

xta0 committed 3 years ago

73e121de

[GPU] Add an attribute to the torchscript model exported by metal (#47174)

xta0 committed 3 years ago

ad3a3bd0

[quant][graphmode][fx] Add test for non quantized embedding and embeddingbag (#47092)

jerryzh168 committed 3 years ago

0ead9d54

[TensorExpr] Support LLVM versions 8 through 12 (#47033)

asuhan committed 3 years ago

4df7eefa

expose Timer docs to PyTorch website. (#46880)

Taylor Robie committed 3 years ago

ac8a8185

Add NestedTensor specific dispatch key to PyTorch (#44668)

Christian Puhrsch committed 3 years ago

09a52676

add node by node cost function (#47009)

scottxu0730 committed 3 years ago

1fe273d7

Fix bug in toComplexWithDefault (#43841)

zasdfgbnm committed 3 years ago

084b7112

Back out "[Gradient Compression] Surface C++ comm hooks to Python API as built-in comm hooks" (#47234)

Yi Wang committed 3 years ago

b1b77148

[vulkan_api][ops] Mm, Pool, Upsample (#47063)

IvanKobzarev committed 3 years ago

2cff3bba

quantize_tensor_per_channel ARM implementation (#46018)

ajliu committed 3 years ago

b0e954ff

[jit] fix traced training attribute (#47211)

zdevito committed 3 years ago

ecfa7a27

Add benchmark for per channel tensor quantization (#46017)

ajliu committed 3 years ago

27f4a78b

For torch::jit::module's attr method to moble::module (#47059)

kimishpatel committed 3 years ago

82b74bd9

[PT] optional -> c10::optional (#47144)

Hao Lu committed 3 years ago

b6685d38

[quant][graphmode][fx][fix] Linear work with float_qparam_dynamic_qconfig (#47068)

jerryzh168 committed 3 years ago

be2e3dd2

Add scalar.conj() and update backward formulas for add and sub (#46596)

anjali411 committed 3 years ago

cedeee2c

Port CPU Trace from TH to ATen (#47126)

zou3519 committed 3 years ago

86151da1

Add test for trace (#47125)

zou3519 committed 3 years ago

8054ae3e

Enable inlining into reductions (#47020)

navahgar committed 3 years ago

f58842c2

Add RAII DetectAnomalyGuard (#47164)

t-vi committed 3 years ago

b5a1be02

Remove travis-python references as well as some unnecessary dependencies (#47209)

janeyx99 committed 3 years ago

ebf36ad3

Make "Run flake8" step always succeed again (#47236)

samestep committed 3 years ago

42b6f967

Add `inputs` argument to `autograd.backward()` (#46855)

soulitzer committed 3 years ago

f5073b0c

Fix max_pool1d on discontiguous tensor (#47065)

heitorschueroff committed 3 years ago

18470f68

Newer Older