pytorch/xla

Pull Requests Commits

latency_hiding_scheduler pin update

alanwaketan committed 2 years ago

d66dbaef

clear pending ir should also clear the cc op tokens (#5385)

JackCaoG committed 2 years ago

Verified 2676a200

Add gpu doc for how to build PyTorch/XLA from source with GPU support. (#5384)

vanbasten23 committed 2 years ago

Verified c0a432b9

Rectify test_zero1.py once optim.load_state_dict doesn't guarantee immutability (#5382)

janeyx99 committed 2 years ago

Verified ca5eab87

[Dynamo|TPU] Tweak `atol` and `rtol` for `test_simple_model_with_different_input_shape` on TPU (#5373)

ManfeiBai committed 2 years ago

Verified fbec7c10

Remove the mention of XRT_TPU_CONFIG in the CONTRIBUTING.md (#5379)

JackCaoG committed 2 years ago

Verified f1a29639

Revert "Fix ReplicateShardedData for int type (#5374)" (#5380)

JackCaoG committed 2 years ago

Verified 42ca03ed

Update dynamo.md (#5378)

wonjoo-wj committed 2 years ago

Verified d209207c

Fix ReplicateShardedData for int type (#5374)

JackCaoG committed 2 years ago

Verified 7fb7dfe3

[BE] use self.assertEquals instead of str equality in test_zero1.py (#5367)

janeyx99 committed 2 years ago

Verified 498bce41

[Dynamo|TPU] Skip `DynamoInferenceBasicTest.test_resnet18` on TPU (#5361)

ManfeiBai committed 2 years ago

Verified 4f4978c9

Don't rewrite index hints in global save planning (#5348)

jonb377 committed 2 years ago

Verified 34b64b0c

Add a script for running stablehlo tests. (#5360)

qihqi committed 2 years ago

Verified dbed83c6

[Dynamo|TPU] Skip`DynamoTrainingBasicTest.test_resnet18` on TPU (#5362)

ManfeiBai committed 2 years ago

Verified 671bfa7e

[Dynamo|TPU] Tweak `atol` and `rtol` for `test_dynamo.py` (#5363)

ManfeiBai committed 2 years ago

Verified 8ed67bbc

Revert "[BE] use self.assertEquals instead of str equality in test_zero1.py (#5364)" (#5366)

JackCaoG committed 2 years ago

Verified 3a817404

[BE] use self.assertEquals instead of str equality in test_zero1.py (#5364)

janeyx99 committed 2 years ago

Verified 8ada3332

Check the actual device instead of query env var for virtual device (#5352)

JackCaoG committed 2 years ago

Verified d09cc690

Add python test for SPMD+Runtime Python API (#5349)

JackCaoG committed 2 years ago

Verified e70fb298

fix amp dtype setting for GPU (#5337)

baoleai committed 2 years ago

Verified 22ba4af2

Fix CPU/GPU test coverage (#5355)

lsy323 committed 2 years ago

Verified 443fad11

remove unused patches (#5353)

lsy323 committed 2 years ago

Verified 1fa34bf5

Add py3.10 whl path to doc, refactor whl table (#5354)

lsy323 committed 2 years ago

Verified b95554b6

add fs linker flag (#5347)

lsy323 committed 2 years ago

Verified 5673ac4a

Make all-reduce a no-op when world size is 1 (#5342)

will-cromar committed 2 years ago

Verified 61816b0b

Also dump output sharding on HLO file (#5339)

JackCaoG committed 2 years ago

Verified efb1e6fd

Check out the release branch instead of origin/master in ansible (#5344)

vanbasten23 committed 2 years ago

Verified 68d7fe80

OpenXLA pin update 07052023 (#5304)

lsy323 committed 2 years ago

Verified 8ffbd1a4

Update Python device API for SPMD (#5129)

JackCaoG committed 2 years ago

Verified f996701a

Sharding should be per output of IR Node, instead of per IR Node (#5330)

JackCaoG committed 2 years ago

Verified 901d154a

Older