strisunshinewentingwang/pytorch

Pull Requests Commits

delete torch::deploy from pytorch code (but non internal)

PaliC committed 1 year ago

0b748e73

Add mask_type=2 to masked_softmax for when mask.size() == input.size() (#85915)

george-qi committed 1 year ago

85258ec1

Fix rand_like nvprim meta function. (#85882)

kevinstephano committed 1 year ago

6004c65a

Update _torch_docs.py (#85924)

vfdev-5 committed 1 year ago

103a21f4

assert lambda >= 0 in poisson distribution cuda kernel (#85906)

yuguo68 committed 1 year ago

c036fb3e

Fix typo under docs directory and RELEASE.md (#85896)

kiszk committed 1 year ago

bc57306b

assert weights being 1-d tensor in bincount (#85881)

yuguo68 committed 1 year ago

11224f34

Revert "Improve make_tensor performance for float and complex types (#85473)"

pytorchmergebot committed 1 year ago

6db3539e

Align functorch docs with PyTorch's (#85856)

zou3519 committed 1 year ago

50000f3c

[functorch] Update notebooks for latest release (#85855)

zou3519 committed 1 year ago

19c7a6b5

[functorch] Update install instructions in docs (#85854)

zou3519 committed 1 year ago

48b3582e

Remove codesigning from ios circleci workflows (#85630)

jmdetloff committed 1 year ago

06868004

Fix bug in PythonFallBack (#85795)

tugsbayasgalan committed 1 year ago

a9183c0f

Fix `check_compiler_ok_for_platform` on non-English locales (#85891)

Flamefire committed 1 year ago

fe87ae69

Re-introduce the functorch docs build (#85838)

zou3519 committed 1 year ago

0449cf0c

Pass group ranks and options to third party distributed backends (#73164)

esaliya committed 1 year ago

941d7a31

(bsr/csr) x dense mm (#85551)

nikitaved committed 1 year ago

e15a48de

Use `int64_t` for nll_loss with cuda inputs (#85395)

crcrpar committed 1 year ago

ef0baba2

resubmit: "resubmit: [mta] APEX style Fused Adam (#81705) (#85507)" (#85739)

crcrpar committed 1 year ago

5f26df03

Update debug flag for vulkan (#85715)

kimishpatel committed 1 year ago

44eefb13

Add vulkan qualifier to the kernel name (#85714)

kimishpatel committed 1 year ago

ad3bea58

[Pytorch][benchmark vulkan] Fix vulkan profiling (#85713)

kimishpatel committed 1 year ago

e0af12a0

Remove torch/extension.h dependency in torch/csrc/functorch/init.cpp (#85659)

zou3519 committed 1 year ago

e0170c7c

[fix] max_pool1d: shape check (#85594)

kshitij12345 committed 1 year ago

8fb470e8

catches failure on nvprim speculative lowering (#85580)

jjsjann123 committed 1 year ago

cab6ffa0

Stop cuda-10.2 binary builds (#85873)

atalman committed 1 year ago

a807f198

Add opt-einsum to CI (#85574)

janeyx99 committed 1 year ago

3cdf621f

Release GIL when doing shared memory copies on Tensors (#85389)

ezhang887 committed 1 year ago

b25a1ce2

Revert "C10D extension to enable per-thread PG (#84153)"

pytorchmergebot committed 1 year ago

6fae62b3

Separate magma installation for ROCm into its own file (#85567)

jithunnair-amd committed 1 year ago

976e2a35

Older