strisunshinewentingwang/pytorch

Pull Requests Commits

Trace module calls

James Reed committed 4 years ago

309b28ee

API for finding a common ancestor block for a pair of nodes

James Reed committed 4 years ago

0f4b226a

Consistently use TORCH_CUDA_API for all files that live in cuda targets. (#29158)

ezyang committed 4 years ago

adb7df71

Delete THP_CORE macro; partially replace with THP_BUILD_MAIN_LIB (#29143)

ezyang committed 4 years ago

a5d356cb

Clean up named tensor `propagate_names` API (#29239)

zou3519 committed 4 years ago

f227530c

Fixed export for random (#28470)

neginraoof committed 4 years ago

364e525f

skip broken custom op test (#29334)

suo committed 4 years ago

8ed84a91

update op bench readme (#29289)

mingzhe09088 committed 4 years ago

7d01d5ef

move cuda abs to Aten (#25857)

XiaobingSuper committed 4 years ago

e51d937e

Skips test_equiv_recurrent (#29255)

Mike Ruberry committed 4 years ago

74b2d9ed

split remaining "easy" tests (#29249)

suo committed 4 years ago

cc457ca3

Add removeAttribute to `ClassType` (#28984)

jerryzh168 committed 4 years ago

f93a6e54

update gloo submodule (#29248)

rohan-varma committed 4 years ago

7069eee2

Remove CollisionChecker from typeids (#29242)

smessmer committed 4 years ago

eb46d647

Print aars content detailed size info (#28438)

IvanKobzarev committed 4 years ago

ab855d06

Revert D18171156: Merge Tensor and Variable.

ezyang committed 4 years ago

9c43b16d

Add the intra-op parallelism for equal operator (#28810)

jianyuh committed 4 years ago

6a4b51ae

explicitly provide memory format when calling to clone() at TensorFactories.cpp

ifedan committed 4 years ago

9ae6fd25

group quantized op benchmarks into a new binary (#29288)

mingzhe09088 committed 4 years ago

e4c4ff07

skip cuda test if not on GPU machines

mingzhe09088 committed 4 years ago

114e7382

add cuda to all op benchmark (#29285)

mingzhe09088 committed 4 years ago

e8645062

add execution mode to the test name (#29284)

mingzhe09088 committed 4 years ago

27115612

explicitly provide memory format when calling to clone() at SortingKthValue.cu

ifedan committed 4 years ago

50fa132b

explicitly provide memory format when calling to clone() at SpectralOps.cu

ifedan committed 4 years ago

af45801f

Fix virtualenv builds on Windows (#29273)

peterjc123 committed 4 years ago

d05da7da

explicitly provide memory format when calling to clone() at Unique.cu

ifedan committed 4 years ago

4e53f3bc

explicitly provide memory format when calling to clone() at TensorTransformations.cpp

ifedan committed 4 years ago

5e0cf055

Revert D18195868: Implementation of cosine learning rate training policy

ezyang committed 4 years ago

abe05a16

explicitly provide memory format when calling to clone() at LinearAlgebra.cpp

ifedan committed 4 years ago

689599d0

Autogenerated contiguous memory format for old *_like calls

VitalyFedyunin committed 4 years ago

81bf7364

Newer Older