ROCm/jax

Pull Requests Commits

Merge pull request #26426 from NKlug:nklug-linalg-solve-docu

Google-ML-Automation committed 320 days ago

6a638ac8

[pallas:triton] Updated :export_pallas_test following the changes in Pallas Triton lowering

superbobry committed 320 days ago

700298fd

[TPU][Pallas][XLA] Add BUILD time codegen tool that turns a pallas kernel into a parameterized kernel loader header that can be utilized anywhere in C++

Google-ML-Automation committed 320 days ago

6bedabd3

Run GPU CI on PRs destined for QA branches (#228)

charleshofer committed 320 days ago

Verified c501484b

Remove conflicting param for ci_build

charleshofer committed 320 days ago

4e403d29

Merge branch 'rocm-main' into ci-upstream-sync-110_1

charleshofer committed 320 days ago

31c1f254

[pallas] fix typo

jenriver committed 320 days ago

e6f38b7f

Merge pull request #26386 from gnecula:debug_info_jaxpr_5

Google-ML-Automation committed 320 days ago

4f979b44

Some lax.linalg housekeeping.

dfm committed 320 days ago

154e4506

Merge pull request #26384 from gnecula:debug_info_jaxpr_4

Google-ML-Automation committed 320 days ago

1a8d5377

Merge pull request #26411 from jakevdp:jnp-window-functions

Google-ML-Automation committed 320 days ago

260a879b

Bump oldest supported libtpu to match the compatibility window (12 weeks)

apaszke committed 320 days ago

c24a3a77

Add a missing jaxlib version check in ragged_collective_test

apaszke committed 320 days ago

26d8e112

Merge pull request #26436 from hawkinsp:minver2

Google-ML-Automation committed 320 days ago

91c9bbfa

Bump the minimum Mac OS X version for x86 builds to 11.0.

hawkinsp committed 320 days ago

f6ca6866

Update XLA dependency to use revision

Google-ML-Automation committed 320 days ago

1402220e

#sdy enable `test_mem_kind_donation_pinned_host` for Shardy.

bartchr808 committed 320 days ago

6ed4c29c

[Pallas] Add pipeline mode to pltpu

Google-ML-Automation committed 320 days ago

6740165e

Use an PartitionSpec.UNCONSTRAINED to represent unconstrained dimensions in ParsedPartitionSpec, rather than None.

hawkinsp committed 321 days ago

cf308a84

[better_errors] Continue adding debug info to Jaxprs (step 7)

gnecula committed 321 days ago

817b3e57

Update XLA dependency to use revision

Google-ML-Automation committed 321 days ago

8401d9b7

Document behavior of linalg.solve in case the system matrix is singular

NKlug committed 322 days ago

af794721

Update XLA dependency to use revision

Google-ML-Automation committed 322 days ago

b3df4d58

[better_errors] Continue adding debug info to Jaxprs (step 4)

gnecula committed 322 days ago

1e813e16

Merge pull request #26407 from jakevdp:printoptions-doc

Google-ML-Automation committed 322 days ago

28903574

Merge pull request #26406 from jakevdp:lax-docs

Google-ML-Automation committed 323 days ago

a7c20e7f

Don't call `get_cur_mesh_sharding` if sharding-in-types mode is not enabled

yashk2810 committed 323 days ago

21e1be33

Merge pull request #221 from ROCm/ci-upstream-sync-106_1

github-actions[bot] committed 323 days ago

Verified 9fbc1c1e

Speed up NamedSharding construction.

hawkinsp committed 323 days ago

f21b0f03

Increase the absolute error tolerance to fix flaky tests.

Rifur13 committed 323 days ago

cd075375

Newer Older