strisunshinewentingwang/pytorch

Pull Requests Commits

[ao][sparsity] add option for tensor_fqn to sparsity API

HDCharles committed 2 years ago

644c3cfa

automate hash updates - add new lint to appease lint (#79832)

clee2000 committed 2 years ago

584e21bf

[quant] Refactor nonuniform quantization mapping functions

asl3 committed 2 years ago

228e082c

Fix `SWALR` doc string (#79836)

janosh committed 2 years ago

660d9dde

finalized GHA for promoting to viable/strict (#79649)

swang392 committed 2 years ago

e2a2ecac

Small typo fix in nn.utils.parametrizations (#79854)

zazyzaya committed 2 years ago

0fa1920d

[GH1] Make pull required for merge (#79704)

janeyx99 committed 2 years ago

de842cac

[primTorch] Implements refs for gcd, lcm and remainder (#78747)

syed-ahmed committed 2 years ago

802efc90

Automated submodule update: kineto (#79662)

facebook-github-bot committed 2 years ago

9be97ea4

[Profiler] Expose extra fields to Python

Taylor Robie committed 2 years ago

332e43ed

Add opset16 onnx support for `torch.scatter_add` (#79103)

qqaatw committed 2 years ago

24a04671

[FSDP] Fix a small bug of pre_backward_hook params prefetch (#78851)

LinjianMa committed 2 years ago

5ca9253f

Back out "Back out "[const_fold] Set requires_grad based on the folded tensor; add device_for_folding option"" (#79696)

jfix71 committed 2 years ago

d4bbebbb

[static runtime] Add JIT prim ops: aten::cpu, aten::list, aten::numel, aten::__range_length (#79111)

huiguoo committed 2 years ago

0545c85f

Windows nvtx for CUDA 11.6 require _wgetenv decl (#79798)

George Gensure committed 2 years ago

7d17e3b8

Revert "Fix BytesWarning in torch.load() (#74813)"

pytorchmergebot committed 2 years ago

e10cbe38

[ONNX] Refactor heavy memory usage tests

BowenBao committed 2 years ago

73f6601c

MPS: Add amax and amin Ops with tests (#79682)

abhudev committed 2 years ago

074dc746

Revert "Create a new Ubunutu 22.04 (jammy) build for platform010 (#77591)"

pytorchmergebot committed 2 years ago

dee43798

Do not bind `c10::irange` output by reference (#79815)

malfet committed 2 years ago

399b3dca

Fix BytesWarning in torch.load() (#74813)

aphedges committed 2 years ago

6c2e8119

Exploit symmetry in comparison operators to reduce no. of kernels

peterbell10 committed 2 years ago

706fa49c

disable modes during constructor

samdow committed 2 years ago

24243659

[primTorch] refs: nn.functional.{hard|soft}shrink (#78427)

kshitij12345 committed 2 years ago

cd3a0471

Revert "Back out "[const_fold] Set requires_grad based on the folded tensor; add device_for_folding option" (#79655)"

pytorchmergebot committed 2 years ago

e627dd54

[shard] make state_dict hook be consistent

wanchaol committed 2 years ago

bef2fecb

Add test for torchscripting nn.TransformerEncoder, including fast path (#79796)

erichan1 committed 2 years ago

06274d7a

[AMP] Use generic autocast in example, specify dtype (#79579)

eqy committed 2 years ago

eff74ed7

Create a new Ubunutu 22.04 (jammy) build for platform010 (#77591)

mehtanirav committed 2 years ago

71d82917

started using mode-based tracing

Chillee committed 2 years ago

f5d7e5a1

Newer Older