strisunshinewentingwang/pytorch

Pull Requests Commits

ezyang committed 4 years ago

6ac681c7

rewrite C++ __torch_function__ handling to work with TensorList operands (#34725)

ngoldbaum committed 4 years ago

0a1e0fb8

Pull upstream select_compute_arch from cmake for Ampere (#41133)

zasdfgbnm committed 4 years ago

8940a4e6

Cuda Support for Learnable Fake Quantize Per Channel (GPU) (#41262)

Paul Shao committed 4 years ago

c62550e3

Cuda Support for Learnable Fake Quantize Per Tensor (GPU) (#41127)

Paul Shao committed 4 years ago

4367a733

Adding epsilon input argument to the Logit Op

venkatacrc committed 4 years ago

225289ab

Revert D22480638: [pytorch][PR] Add non-deterministic alert to CUDA operations that use `atomicAdd()`

mrshenli committed 4 years ago

954c2600

[quant][pyper] Add embedding_bag weight quantize and dequantize ops (#41293)

supriyar committed 4 years ago

008ab27b

DDP Communication Hook Main Structure (#40848)

sinannasir committed 4 years ago

d5ae4a07

[cmake] Use PROJECT_SOURCE_DIR instead of CMAKE_* (#41387)

powderluv committed 4 years ago

c86699d4

Fix flaky test_stream_event_nogil due to missing event sync (#41398)

Flamefire committed 4 years ago

563b60b8

Add non-deterministic alert to CUDA operations that use `atomicAdd()` (#40056)

kurtamohler committed 4 years ago

6ff306b8

Add CUDA to pooling benchmark configs (#41438)

peterbell10 committed 4 years ago

dddac948

Krovatkin/reenable test tensorexpr (#41445)

Krovatkin committed 4 years ago

3971777e

Add optimizer_for_mobile doc into python api root doc (#41211)

xcheng16 committed 4 years ago

04320a47

Revert D22517785: [pytorch][PR] Enable TF32 support for cuBLAS

mrshenli committed 4 years ago

3a63a939

Revert D22543215: Adjust bound_shape_inferencer to take 4 inputs for FCs

mrshenli committed 4 years ago

8548a21c

Shape inference for SparseToDense in ExpertCombiner

Dinesh Govindaraj committed 4 years ago

f153b35b

Adjust bound_shape_inferencer to take 4 inputs for FCs (#41452)

csummersea committed 4 years ago

86a2bdc3

Port index_select to ATen (CUDA) (#39946)

Baranowski committed 4 years ago

14f19ab8

Revert D22516606: [pytorch][PR] Temporary fix for determinant bug on CPU

Mike Ruberry committed 4 years ago

9552ec78

SparseAdagrad/RowWiseSparseAdagrad mean fusion on CPU & GPU and dedup version for RowWiseSparse mean fusion on GPU

Yan Xie committed 4 years ago

921d2a16

Export replaceAllUsesAfterNodeWith for PythonAPI (#41414)

taewookoh committed 4 years ago

44b9306d

[adaptive_]max_pool{1,2,3}d: handle edge case when input is filled with -inf (#40665)

Baranowski committed 4 years ago

20f3051f

Temporary fix for determinant bug on CPU (#35136)

vishwakftw committed 4 years ago

fcd6d910

vectorize rounding ops (#41439)

Hector Yuen committed 4 years ago

f074994a

remove template arguments of layernorm

Hector Yuen committed 4 years ago

96f124e6

Change BCELoss size mismatch warning into an error (#41426)

kurtamohler committed 4 years ago

0b73ea0e

Fix flaky profiler and test_callback_simple RPC tests (#41287)

rohan-varma committed 4 years ago

fd032902

[JIT] Fix dead stores in JIT (#41202)

Meghan Lele committed 4 years ago

0d4a110c

Older