pytorch/xla

Pull Requests Commits

remove the dependency on sklearn.

vanbasten23 committed 2 years ago

49f071a9

keep extending c10::SymNodeImpl.

vanbasten23 committed 2 years ago

e60c9800

vanbasten23 committed 2 years ago

6094bf65

Fix the dynamic shape test and add it to the CI.

vanbasten23 committed 2 years ago

519a5e10

[LTC] Adopt getBackend()->GetComputationDataFromNode() (#4191)

alanwaketan committed 2 years ago

Verified dd9b67ff

Fix multi-gpu metrics report (#4187)

ymwangg committed 2 years ago

Verified 44362e6d

[LTC] Use getBackend()->ExecuteComputation() (#4186)

alanwaketan committed 2 years ago

Verified ee65ff15

[LTC] Use torch::lazy::Util::ComputePostOrder (#4178)

alanwaketan committed 2 years ago

Verified ae35e1e2

unblock dynamo integration, get autograd ops to work with python tensor subclasses (#4165)

bdhirsh committed 2 years ago

Verified 4a0aaefb

[WIP] Partially Codegen Atan2 (#4184)

ManfeiBai committed 2 years ago

Verified b5496122

Fix check_tensor_need_materialization (#4185)

JackCaoG committed 2 years ago

Verified b1f167e6

[LTC] Make some torch::lazy::Node* to const T (#4182)

alanwaketan committed 2 years ago

Verified 0115cc00

Additional dynamo test cases (#4175)

JackCaoG committed 2 years ago

Verified bc2694ff

Make ExecuteComputationOptions as default (#4177)

alanwaketan committed 2 years ago

Verified a01b9999

[LTC] Employ more Computation (#4176)

alanwaketan committed 2 years ago

Verified 834e3efd

Remove upsample_*_backward.vec ops (#4087)

wonjoolee95 committed 2 years ago

Verified f4dd9906

[LTC] Use upstream DeviceDataInfo (#4172)

alanwaketan committed 2 years ago

Verified 54f24e25

[SPMD] Mesh to support custom device order. (#4162)

yeounoh committed 2 years ago

Verified b096c5cf

[LTC] Remove XLAIrBuilder::MakeView (#4154)

alanwaketan committed 2 years ago

Verified f7c0775b

Companion PR for 'SymIntify _copy functionalization kernels (and _copy_out too)' (#4164)

ezyang committed 2 years ago

Verified 79cbdb60

Add some basic Dynamo test (#4159)

JackCaoG committed 2 years ago

Verified 515d53e9

Enable `padded` option for even tensor sharding (#4156)

yeounoh committed 2 years ago

Verified 5ff192ea

Update codegen to use xla::Shape (#4111)

wonjoolee95 committed 2 years ago

Verified 587a6b6d

Refactor argparser to be consistent with torch.distributed (#4149)

muellerzr committed 2 years ago

Verified bde1bc6a

Add short_metrics (#4148)

JackCaoG committed 2 years ago

Verified 1f370695

update docs (#4151)

JackCaoG committed 2 years ago

Verified 9ae3cbbb

[SPMD] Test sharding_spec with clear_sharding, deepcopy (#4144)

yeounoh committed 2 years ago

Verified 590dee5d

Fix clear counter (#4143)

JackCaoG committed 2 years ago

Verified 6675ada5

Update release 1.13 in README file. (#4140)

vanbasten23 committed 2 years ago

Verified 615aa68a

enable nn/convolution tests (#4120)

kshitij12345 committed 2 years ago

Verified ad84ae8a

Older