strisunshinewentingwang/pytorch

Pull Requests Commits

[PyTorch] Make TensorImpl::sizes() customizable and disable it for NestedTensorImpl (#73817)

swolchok committed 4 years ago

90be8fa2

Update circleci pipelines to support both master and main branches.

musebc committed 4 years ago

4646caed

Create method to map JIT module to (source, constant) and back. (#74119)

qihqi committed 4 years ago

ded82ad7

Revert D34868005: [torch::deploy] remove asserts from deploy

malfet committed 4 years ago

cc5f8aea

Improve numerical stability of `torch.distributions.wishart.Wishart` (#72993)

nonconvexopt committed 4 years ago

c77857cb

Install ffmpeg=4.4.1 because torchvision doesn't compile on ffmpeg-5

xuzhao9 committed 4 years ago

039b62cf

[torch::deploy] remove asserts from deploy (#73456)

PaliC committed 4 years ago

48b16540

[Static Runtime] Use composite op for TE fusion (#74126)

navahgar committed 4 years ago

b0cab0f4

[Static Runtime] Fix a bug that `aten::full` reuses a tensor that does not match requested one (#73990)

d1jang committed 4 years ago

381c0c08

check if owner is pytorch in jobs

Catherine Lee committed 4 years ago

7ab482f2

[Reducer] small fix (#74127)

rohan-varma committed 4 years ago

3b421af8

[ddp] parameter verification (#74113)

rohan-varma committed 4 years ago

017be349

Parametrize remaining tests in TestAutogradFunctional to use logging_tensor

soulitzer committed 4 years ago

3f140c5b

[BE Hackathon][DataPipe] Automatically generate datapipe.pyi via CMake (#73991)

NivekT committed 4 years ago

ff3688f0

[DataPipe] Separating DataPipes from Dataset into different files (#73396)

NivekT committed 4 years ago

eec994fc

[PyTorch] Avoid schema parsing in lightweight dispatch (#74069)

larryliu0820 committed 4 years ago

33998763

[fix] kaiser_window : meta for window_length > 1 (#73733)

kshitij12345 committed 4 years ago

8bc647b3

[Easy] Remove erroneous comment (#74195)

rohan-varma committed 4 years ago

38b12732

Fix `test_reduce_add_coalesced ` failure (#74027)

IvanYashchuk committed 4 years ago

4bb5e6e8

Make `torch.nn` importable on Python-3.7.0

malfet committed 4 years ago

21a82fb5

Revert D34846005: [quant] Fix implementation for `output_quantized_idxs` in convert

Natalia Gimelshein committed 4 years ago

1e64c8a8

[FSDP] summon offload to CPU (#73904)

rohan-varma committed 4 years ago

c2aba0f1

[FSDP] Option to summon on rank 0 only (#73903)

rohan-varma committed 4 years ago

a5d47201

[Quant][fx] Reenable serialization test after convert refactor (#74204)

andrewor14 committed 4 years ago

9bb47f00

[quant] Fix implementation for `output_quantized_idxs` in convert (#74140)

jerryzh168 committed 4 years ago

a7f9fb99

Remove inaccurate confusing signal box from README.md

janeyx99 committed 4 years ago

bb82a700

Add onednn quant backend (#74137)

Xia-Weiwen committed 4 years ago

060f1b82

Mention milestones as way of tracking issues/PRs

malfet committed 4 years ago

deae5950

[FX] Disable buffer tracing test due to SEV remediation

James Reed committed 4 years ago

b68f2277

[PyTorchEdge] Start writing magic to flatbuffer output (#74084)

dbort committed 4 years ago

6c18a995

Newer Older