strisunshinewentingwang/pytorch

Pull Requests Commits

Simplify forward / backward AD for linalg.eigh and add checks (#70528)

lezcano committed 2 years ago

97585ae1

Correct forward AD for linalg.eig and add checks (#70527)

lezcano committed 2 years ago

061be8d6

Add retry to disabled tests file download (#71030)

janeyx99 committed 2 years ago

e1aea9b9

fix TensorLikePair origination (#70304)

pmeier committed 2 years ago

928ca95f

add a equality comparison helper for assert_close internals (#69750)

pmeier committed 2 years ago

49a5b33a

add explanation of quantized comparison strategy in assert_close (#68911)

pmeier committed 2 years ago

b0a10a70

change sparse COO comparison strategy in assert_close (#68728)

pmeier committed 2 years ago

802dd2b7

make meta tensor data access error message for expressive in assert_close (#68802)

pmeier committed 2 years ago

8d05174d

improve documentation of comparison internals (#68977)

pmeier committed 2 years ago

b652887a

Remove deprecated cuDNN convolution ops (#71128)

jbschlosser committed 2 years ago

523d4489

[AutoAccept][Codemod][FBSourceClangFormatLinter] Daily `arc lint --take CLANGFORMAT`

CodemodService FBSourceClangFormatLinterBot committed 2 years ago

93b2399c

Fix for tensor in list return added to wildcard set (#71170)

Elias Ellison committed 2 years ago

4a8d4cde

Remove precise tuple construct flag (#71121)

Elias Ellison committed 2 years ago

9bccb313

add optional refining (#69776)

Elias Ellison committed 2 years ago

47ad6628

Parse symbolic shapes (#69775)

Elias Ellison committed 2 years ago

772b3e92

Fix mis-specified device arg name (#69645)

Elias Ellison committed 2 years ago

97e8dcba

[jit][edge] Use dynamic type instead of union types for schema parsers. (#70509)

zhxchen17 committed 2 years ago

9465c242

Sparse CSR: Add `torch.randn_like` (#68083)

IvanYashchuk committed 2 years ago

40121456

fx quant: fix test_fx_acc_tracer::test_quantized_batch_norm2d (#71175)

vkuzo committed 2 years ago

831c129e

Performance and memory improvements to batched torch.linalg.solve (#69752)

IvanYashchuk committed 2 years ago

410e91ad

[Profiler] Add glue layer to reduce the use of `#ifdef USE_KINETO` in the profiler code. (#69798)

Taylor Robie committed 2 years ago

786f9460

Enable nested default hooks (#70932)

Varal7 committed 2 years ago

a3b7dd7b

delete ecr_gc_docker job (#71178)

suo committed 2 years ago

433cf44b

[jit][edge] Migrate base types to DynamicType on mobile. (#70233)

zhxchen17 committed 2 years ago

e7634f83

Fixed docs for forward_ad.make_dual (#71159)

vfdev-5 committed 2 years ago

ecb6defa

Speed up quantized upsampling for channels last (#70903)

Timon Künzle committed 2 years ago

2c8cb8a9

Adding python 3.10 binary workflows (#71132)

atalman committed 2 years ago

edf15ebb

Make Kineto + distributed a warning rather than an error (#71120)

Taylor Robie committed 2 years ago

7d6535ca

Drop more unused variables (#71123)

r-barnes committed 2 years ago

45b0bafb

Drop unused variables and add some const (#71106)

r-barnes committed 2 years ago

6c03f8d9

Newer Older