strisunshinewentingwang/pytorch

Pull Requests Commits

[Inductor] Build FX Linear + Permute Vertical Fusion in Inductor (#89118)

jiawenliu64 committed 1 year ago

52701227

Revert "Rewrite assert statement with torch._assert under config (#88246)"

pytorchmergebot committed 1 year ago

9d28775c

[dynamo] Support if cond on NNModuleVariable (#89095)

anijain2305 committed 1 year ago

9d2f5a27

[dtensor] PART 8: move tensor parallel api and tests to core distributed (#88180)

wanchaol committed 1 year ago

f20b3f2e

[dtensor] PART 7: move remaining DTensor tests to core distributed (#88179)

wanchaol committed 1 year ago

0230e52b

[dtensor] PART 6: move DTensor op tests to core distributed (#88551)

wanchaol committed 1 year ago

550a019f

[dtensor] PART 5: move DTensor basic tests to core distributed (#88178)

wanchaol committed 1 year ago

527c5bdb

[dtensor] PART 4: move remaining DTensor ops to core distributed (#88550)

wanchaol committed 1 year ago

1b884763

[dtensor] PART 3: move most DTensor ops to core distributed (#88177)

wanchaol committed 1 year ago

2dcf0978

[dtensor] PART 2: move DTensor abstraction and APIs to core distributed (#88176)

wanchaol committed 1 year ago

4b945967

[dtensor] PART 1: move DeviceMesh and placement to core distributed (#88549)

wanchaol committed 1 year ago

370fc5cb

Upload CSV test reports from inductor (#89112)

huydhn committed 1 year ago

59ba15f3

[Inductor] Support Shape Padding for aten.mm in Inductor (#89086)

jiawenliu64 committed 1 year ago

7e66d1d6

Add an option to include actual license terms to the output (#85624)

kmaehashi committed 1 year ago

e2f06487

Easier to understand event_dim computation (#81396)

johannespitz committed 1 year ago

8ebbd5a8

Symintify numel(), infer_size, prims.elementwise_meta (#88956)

SherlockNoMad committed 1 year ago

ce2f8700

Create native function for determining which implementation of SDP to call (#89029)

drisspg committed 1 year ago

b291c121

[FSDP] Test `named_parameters()` in forward (`use_orig_params=True`) (#89066)

awgu committed 1 year ago

397f1006

Increase slow grad check timeout (#89079)

huydhn committed 1 year ago

46ba0150

Revert "[Inductor] Build FX Linear + Permute Vertical Fusion in Inductor (#88859)"

pytorchmergebot committed 1 year ago

9f0b2c73

Add int64_t, SymInt overloads for all binary operators in C++ (#89063)

ezyang committed 1 year ago

d96dd8ff

Move ConvParams methods directly on struct (#89062)

ezyang committed 1 year ago

43164211

Hide ConvParams struct from ConvUtils.h (#89059)

ezyang committed 1 year ago

49f0be07

Fix and Re-enable test_quantize_fx_lite_script_module.py (#88897)

salilsdesai committed 1 year ago

19cacecf

PyDispatcher integration with functorch (#88785)

zou3519 committed 1 year ago

3bc32799

[functorch] add switch to enable autograd.Function (#88784)

zou3519 committed 1 year ago

2268a321

Revert "Enable correct supported activities for kineto on rocm (#88207)"

pytorchmergebot committed 1 year ago

0ce22574

allow loading model from a path in torchbench (#89028)

shunting314 committed 1 year ago

a1343394

Don't iterate over graph when adding graph input (#89084)

mlazos committed 1 year ago

60ffeb98

Rebase and re-land thread PG (#88795)

yhcharles committed 1 year ago

ee05f47b

Newer Older