strisunshinewentingwang/pytorch

Pull Requests Commits

turn functionalization on in aot_autograd inference (#92857)

bdhirsh committed 1 year ago

ceb0f157

Add rnn.unpad_sequence and rnn.unpack_sequence to documentation (#94316)

mikaylagawarecki committed 1 year ago

5ce1fad7

Update gradcheck docs to mention non-differentiability (#94618)

soulitzer committed 1 year ago

701412a4

Pin setup-buildx-action version. Fix Docker build (#94734)

atalman committed 1 year ago

a064ce19

ao migration: remove package test as this behavior is tested by other things (#94422)

vkuzo committed 1 year ago

216f88d0

ao migration: delete unused test class (#94420)

vkuzo committed 1 year ago

f6adbf4d

Logcumsumexp for CUDA (build-time optimized) (#94310)

mfkasim1 committed 1 year ago

2acac8a8

Update Triton hash (#94249)

malfet committed 1 year ago

4869929f

Revert "Dynamo Export use fake tensor (#94276)"

pytorchmergebot committed 1 year ago

e61d5b95

Revert "[quant] Add quantize and dequantize operators to decomposition table (#93312)"

pytorchmergebot committed 1 year ago

641dc0b8

[Executorch][Quant] Add Choose_qparams_symmetric (#94685)

JacobSzwejbka committed 1 year ago

26289010

Tweak config for mode=max-autotune/reduce-overhead (#94659)

jansel committed 1 year ago

ab261ff5

Fix NVML visible device parsing (#92315)

malfet committed 1 year ago

e7e51b3a

Checkout torchbench with only needed models (#94578)

weiwangmeta committed 1 year ago

6fadd5e9

[MPS] Add sort and argSort Op. (#94697)

kulinseth committed 1 year ago

18587cb3

[BE] [3/3] Rewrite `super()` calls in test (#94592)

XuehaiPan committed 1 year ago

046e88a2

[MPS] Add Python Module Bindings for the MPS backend (#94417)

razarmehr committed 1 year ago

bdd8f518

Update Cutlass to v2.11 (#94188)

Skylion007 committed 1 year ago

a0f9abdc

fix some MKL detection issues of CMake (#94402)

cyyever committed 1 year ago

7ef46d40

[inductor] Persistent reductions (#92267)

jansel committed 1 year ago

a8fdfb4b

[FSDP] Avoid printing incorrect warning for _get_param_to_fqns (#94494)

fegin committed 1 year ago

eb81e7ec

[FSDP][state_dict] Return tensors instead of FlatParameters to avoid pickling errors (#94637)

fegin committed 1 year ago

963d8f54

Issue-88098: extract utils from check labels (#94597)

Goldspear committed 1 year ago

2c76838d

inductor: fix customer op _convolution_pointwise_.binary functional error at AOTAutograd (#94581)

XiaobingSuper committed 1 year ago

d04fd6b8

[MPS] Add repeat_interleave to MPS (#88649)

jazzysoggy committed 1 year ago

fe0c7fbc

[MPS] Add scatter gather kernels (support up to 5 dimensions) (#94663)

DenisVieriu97 committed 1 year ago

b794fd19

[functorch] Add support on CUDA keys for control flow ops. (#94465)

zhxchen17 committed 1 year ago

e3c4cea6

[vision hash update] update the pinned vision hash (#94557)

pytorchmergebot committed 1 year ago

989fb7c9

[BE] Apply almost all remaining flake8-comprehension checks (#94676)

Skylion007 committed 1 year ago

67d97909

[MPS] Add support for TopK k>16 (#94639)

kulinseth committed 1 year ago

54c0f376

Newer Older