strisunshinewentingwang/pytorch

Pull Requests Commits

[Dynamo] More robust pyop support, module properties as args (#87020)

voznesenskym committed 1 year ago

3c320a56

OpInfo: Sample input cleanup (3/n) (#86380)

peterbell10 committed 1 year ago

5d6e8315

Sync changes from `pytorch/torchdynamo` (#87013)

jansel committed 1 year ago

054a2fd6

Fixed partitioner issue with getitem and made metadata a storage more consistent (#87012)

Chillee committed 1 year ago

2c1bc216

[einsum] Fix opt_einsum defaults to be more reasonable (#86985)

janeyx99 committed 1 year ago

91c70154

Support unpacking python dictionary in torch.jit.trace() (#81623)

tangleintel committed 1 year ago

7980ed95

[RFC] Separate CPU offload activation to its own wrapper (#85459)

rohan-varma committed 1 year ago

bdefa260

[quant][docs] Formatting fixes for fx graph mode quantization README (#86914)

jerryzh168 committed 1 year ago

100113b8

[vision hash update] update the pinned vision hash (#86758)

pytorchmergebot committed 1 year ago

f6f1aefb

torchdynamo: add linear pointwise(binary) fusion kernel (#86583)

XiaobingSuper committed 1 year ago

46aaae98

torchdynamo: add convolution pointwise(binary) fusion kernel (#86582)

XiaobingSuper committed 1 year ago

5210fab6

torchdynamo: add convolution pointwise(unary) fusion kernel (#86581)

XiaobingSuper committed 1 year ago

9a7a49b2

ATen/native (1/6): Use per-operator headers (#75571)

peterbell10 committed 1 year ago

d5a7e6db

[data] add autocompletion to datapipes (#86960)

edward-io committed 1 year ago

4584d06e

[BE] Extend linter to detect DOS newlines (#86973)

malfet committed 1 year ago

3924aa75

[quant][be] Remove unused helper functions in convert.py (#86913)

jerryzh168 committed 1 year ago

b8aa1767

[quant][be] Rename qconfig_map to node_name_to_qconfig (#86861)

jerryzh168 committed 1 year ago

761ca20d

Sync changes from pytorch/torchdynamo, enable tests (#86950)

jansel committed 1 year ago

8f71e8de

Set -Werror=braced-scalar-init (#86911)

wconstab committed 1 year ago

78ef4097

[xnnpack][lite-int] preprocess (#86980)

mcr229 committed 1 year ago

155b8858

[onnx] Add support for autograd function inlining in ONNX_ATEN_FALLBACK mode (#85736)

shubhambhokare1 committed 1 year ago

7c73b456

enable optim tests on dynamo to test flaky bot (#86976)

clee2000 committed 1 year ago

d29c8c0f

[CoreML][ios_crash] Use special throw macro when encountering CoreML API errors (#86938)

mcr229 committed 1 year ago

1a7409c7

symintify all of derivatives.yaml (#86610)

bdhirsh committed 1 year ago

34c86ade

min/max support for SymInt/Floats, finish as_strided/scatter/squeeze() backward symint support (#86609)

bdhirsh committed 1 year ago

d7bbb61f

Added new test test_compare_cpu that checks if cpu and gpu results are consistent (#85011)

srossross committed 1 year ago

1bb609ad

Chore: Add 'mps' to the docs of tensor_attributes (#86585)

LukasM937 committed 1 year ago

e027740e

Remove empty_like+fill from AOT Autograd graphs for nvFuser (#86908)

IvanYashchuk committed 1 year ago

fc3afc84

[ONNX] Reland: Update training state logic to support ScriptedModule (#86745)

justinchuby committed 1 year ago

56a744bf

Sparse support for ReLU (#86749)

amjames committed 1 year ago

527ebedb

Newer Older