strisunshinewentingwang/pytorch

Pull Requests Commits

remove step invocation warning (#107216)

briancoutinho committed 1 year ago

dcc674de

Fix fallback FBGEMM implementation for Big Endian systems. (#96422)

AlekseiNikiforovIBM committed 1 year ago

60bb02a9

Automatically turn on dynamo in cond (#108028)

ydwu4 committed 1 year ago

49e964ca

Add UncapturedHigherOrderOpError to always raise exceptions for cond. (#108027)

ydwu4 committed 1 year ago

6f8eecfb

Enable tuple operands for cond (#108026)

ydwu4 committed 1 year ago

138e2895

Move cond to torch/_higher_order_ops/ (#108025)

ydwu4 committed 1 year ago

86889653

[2/N] fix clang-tidy warnings in torch/csrc (#107966)

cyyever committed 1 year ago

1fd4e787

[reland][quant][pt2e][xnnpack_quantizer] Add support for mul and mul_relu (#107930) (#107992)

jerryzh168 committed 1 year ago

9ae3d7ca

Serialize pytree to json string (#106116)

angelayi committed 1 year ago

a432f37e

[Quant][Inductor] add UT of dequant promotion for linear (#106935)

Xia-Weiwen committed 1 year ago

4b27e46d

[Quant][Inductor] Enable quantization linear pattern fusion inside inductor (#106934)

Xia-Weiwen committed 1 year ago

f3adbab4

[Quant][Inductor] Enable qlinear weight prepack inside inductor constant folding (#106782)

Xia-Weiwen committed 1 year ago

15ceafb5

[Quant][PT2E] Enable linear and linear-unary post-op quant recipe for x86 inductor quantizer (#106781)

Xia-Weiwen committed 1 year ago

e9b0f62a

[inductor] correctly handle resize for AOTInductor wrapper calls (#107848)

chenyang78 committed 1 year ago

2179ebde

[Quant] Add int8 linear op impl for quantization PT2E with Inductor. input is an int8 CPU tensor; weight is an int8 MdkldnnCPU tensor. (#105818)

Xia-Weiwen committed 1 year ago

a6d3da18

[vision hash update] update the pinned vision hash (#108011)

pytorchupdatebot committed 1 year ago

bad3f2db

Update writing_batching_rules.md (#108007)

james-a-watson committed 1 year ago

808e0886

[ROCm] ROCm compatible configs for triton kernels (#107584)

jataylo committed 1 year ago

a18ee0c6

[ONNX] Support `torch.compile(backend="onnxrt", options=OrtBackendOptions(...))` (#107973)

abock committed 1 year ago

15e5bd51

[Quant][PT2E]Make _fuse_conv_bn_ support graph capture by torch._dynamo.export (#107951)

leslie-fang-intel committed 1 year ago

c85c5954

[Quant][Inductor] Fix the non contiguous load with uint8 data type (#106958)

leslie-fang-intel committed 1 year ago

fdbc2ec5

[Quant][Inductor] Enable lowering of qcat (#106838)

leslie-fang-intel committed 1 year ago

9e3f3f0b

[Quant][PT2E] Add cat and avg_pool2d recipe into x86InductorQuantizer (#106836)

leslie-fang-intel committed 1 year ago

1147a28b

[quant][pt2e] Add reference representation rewrite for statically quantized linear (#107994)

jerryzh168 committed 1 year ago

15d4dedb

[export] Don't save example_inputs for now. (#107978)

zhxchen17 committed 1 year ago

162109f6

Handle 2D blocking with foreach (#107840)

mlazos committed 1 year ago

d4a99631

[cuda] remove dead CUDA code in layer_norm_kernel.cu (#107976)

valentinandrei committed 1 year ago

558a9501

[dynamo+aten] Enable embedding_bag_byte_unpack + meta kernel impl (#107937)

hl475 committed 1 year ago

9fa52834

[submodule][Quant][PT2E] Upgrade IDeep to remove redundant QConv weight scale reciprocal calculation (#107565)

leslie-fang-intel committed 1 year ago

2bddfb02

[Quant][PT2E] Enable weight scale optimization in QConv PT2E (#105996)

leslie-fang-intel committed 1 year ago

780a5a0c

Newer Older