pytorch/xla

Pull Requests Commits

Increase max in flight operation to accommodate eager mode

JackCaoG committed 1 year ago

029ab8fb

Modularize zero step function and make it customizable (#7233)

karakusc committed 1 year ago

Verified 51f8d3e6

Bump JAX version (#7258)

vanbasten23 committed 1 year ago

Verified 1e3e6b83

Update the XLA pin (#7254)

vanbasten23 committed 1 year ago

Verified a2edc217

Supporting passing ML framework name and version through the PjRt Client Create options in PyTorch/xla. (#7249)

aman2930 committed 1 year ago

Verified 81f657a2

add torch_xla.experimental.compile for eager mode (#7246)

JackCaoG committed 1 year ago

Verified 90168e8e

Revert "Remove stack dump for compiler failure which can happen often" (#7226)

JackCaoG committed 1 year ago

Verified f4665a73

Move nightly version to 2.5.0 (#7245)

JackCaoG committed 1 year ago

Verified 283abdab

Register custom call with CUDA plugin (#7219)

bhavya01 committed 1 year ago

Verified 86d8967c

Change #include <c10/util/Optional.h> to #include <optional> (#7241)

cyyever committed 1 year ago

Verified c65e87b3

Refactor a bit so that the ops/ is a standalone module (#7250)

qihqi committed 1 year ago

Verified 8cd6ad43

Updating ml_framework_name and version to pytorch/xla2 in jax_configs. (#7211)

aman2930 committed 1 year ago

Verified 3f54fa21

Enable eager mode for PyTorch/XLA (#7234)

JackCaoG committed 1 year ago

Verified 4d0b94fd

Use dest_offsets directly in LoadPlanner (#7155)

jonb377 committed 1 year ago

Verified 7be1d3d0

Update gpu.md to run ResNet example on GPU (#7222)

ahoblitz committed 1 year ago

Verified 71c25e65

Add build trigger for 2.4.0-rc1 release (#7236)

bhavya01 committed 1 year ago

Verified 7bedfe6e

force to string output (#7239)

zpcore committed 1 year ago

Verified 02559649

Disable test_mp_early_exit.py temporarily (#7229)

ManfeiBai committed 1 year ago

Verified 258f9a71

Add python3.11 doc for XLA v2.3 (#7166)

Mon-ius committed 1 year ago

Verified cd8ac11b

Fix __getitem__ (#7237)

qihqi committed 1 year ago

Verified ad126b8b

[SPMD] Support reduce-scatter in manual sharding (#7231)

alanwaketan committed 1 year ago

Verified 70d2e9ea

Support int4 weight in quantized matmul/linear (#7235)

lsy323 committed 1 year ago

Verified ac371fb8

[ZeRO-1] Sync features from r2.1_aws_neuron branch (#7132)

hgt312 committed 1 year ago

Verified 551a76cd

temporarily skip the pjrt test (#7230)

zpcore committed 1 year ago

Verified d0fb59e7

Use std::optional (#7228)

cyyever committed 1 year ago

Verified 2c71e9b3

torchbench CI (#7162)

zpcore committed 1 year ago

Verified 291764f6

Fix wrong env flag for using auto sharding in documentation (#7225)

lausannel committed 1 year ago

Verified 28c5e141

Remove stack dump for compiler failure which can happen often (#7207)

jeffhataws committed 1 year ago

Verified 5338a3b0

Copy examples dir for TPU CI (#7199)

JackCaoG committed 1 year ago

Verified 1ac8968d

Add stub plugins for XPU and Neuron (#7209)

will-cromar committed 1 year ago

Verified f9851b36

Older