strisunshinewentingwang/pytorch

Pull Requests Commits

Add OpInfo tests for torch.{dot, vdot, bmm, mv}

anjali411 committed 3 years ago

63fd50ff

Add lint for unqualified `noqa` (#56272)

samestep committed 3 years ago

e3900d2b

[iOS GPU] Move the definition of `fp16_t` to MetalUtils.h (#54521)

xta0 committed 3 years ago

7bcf95bb

Support factory kwargs in torch.nn modules (#54508)

jbschlosser committed 3 years ago

40483acc

[NNC] added more python bindings for loopnest (#56213)

huiguoo committed 3 years ago

ca6e5c7f

Hide warnings for deprecated quantization APIs (#56291)

davidriazati@fb.com committed 3 years ago

d1b6383d

unified GlooStore and c10d store API (#56222)

marksaroufim committed 3 years ago

48aaea33

[Mobile GPU] Ban mutations in JIT passes (#56070)

xta0 committed 3 years ago

5748cc0d

Enable AutoGradMode in InferenceMode. (#56107)

Ailing Zhang committed 3 years ago

98162cb0

Remove the unused maximum and minimum functions in vec256_base (#56313)

xuhdev committed 3 years ago

8881f504

Sort glob of files to ensure it is deterministic (#55850)

ezyang committed 3 years ago

6409d344

Lazily initialize alias db in remove_mutation opt (#55949)

Peng Wu committed 3 years ago

838d3079

Increase default rendezvous timeout to 15 minutes

aivanou committed 3 years ago

98ac6f7c

Support int32 indices in torch.repeat_interleave (#55102)

Brandon Lin committed 3 years ago

d806b061

Added OpInfos of add & mm (#55915)

imaginary-person committed 3 years ago

b6b2fc7e

Add OpInfo for (nan)quantile (#55548)

heitorschueroff committed 3 years ago

92991d95

Revert D27600457: [pytorch][PR] Support factory kwargs in torch.nn modules

samestep committed 3 years ago

d05e7c16

[special] OpInfo `i0e`: fix missing check (#56232)

kshitij12345 committed 3 years ago

7d175591

Support factory kwargs in torch.nn modules (#54508)

jbschlosser committed 3 years ago

1077f872

Make tensordot resize output tensor's size if out= argument is specified & make it safely cast & copy output (#56286)

imaginary-person committed 3 years ago

7513455c

add tests for torch.testing (#54784)

pmeier committed 3 years ago

0e106fce

Updated internal code for orgqr function (#56247)

IvanYashchuk committed 3 years ago

2219286d

[NNC] Make normalization transformation in-place (#56158)

navahgar committed 3 years ago

b387f7ca

[pytorch][PR] Automated submodule update: tensorpipe (#56348)

Lucas Hosseini committed 3 years ago

22d4d9f4

[CUDA graphs] Allows DeviceCachingAllocator to capture cross-stream memory use (#55860)

mcarilli committed 3 years ago

ffdecc1a

Porting logcumsumexp tests to OpInfo (#56135)

iramazanli committed 3 years ago

3e42da09

[c10d] Remove deprecated use of torch.LongTensor, torch.ByteTensor (#55861)

rohan-varma committed 3 years ago

ce05b7a3

Short circuits DistributedDataParallel._recursive_to's copy and stream syncs if input is already on the right device (#55624)

mcarilli committed 3 years ago

a24b1724

[NNC] Fuse loops that have the same bounds as expressions (#55997)

navahgar committed 3 years ago

29c5cb79

OpInfo porting for logsumexp operator (#55520)

iramazanli committed 3 years ago

b0e0841f

Older