pytorch/xla

Pull Requests Commits

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

Remove unused variables from run_tests.sh (#7217)

bhavya01 committed 1 year ago

Verified de283c98

init glossary (#7214)

ahoblitz committed 1 year ago

Verified 66c7ee8b

Training example for llama3; misc changes to make the training work. (#7194)

qihqi committed 1 year ago

Verified f4a612cb

Add int8 per channel weight-only quantized matmul (#7201)

lsy323 committed 1 year ago

Verified 56ddd5de

bhavya01 committed 1 year ago

Verified 4a30ea7d

ask user to install tf-cpu instead for the profiler (#7212)

JackCaoG committed 1 year ago

Verified 26327a2f

let 3.8 compatiable (#7210)

zpcore committed 1 year ago

Verified 8748a2fb

Allow user to configure distributed runtime service. (#7204)

vanbasten23 committed 1 year ago

Verified 0e1f765c

Add options in benchmark script to keep model on CUDA (#7193)

vanbasten23 committed 1 year ago

Verified f95454ac

Let grad ckpt apply opt-barrier to all params and buffers (#7206)

alanwaketan committed 1 year ago

Verified aec27305

Make clip_grad_norm_ follow the input's dtype (#7205)

alanwaketan committed 1 year ago

Verified f5c00624

openxla pin update to 06/05 (#7200)

zpcore committed 1 year ago

Verified 676853f2

[Pallas] Make _histogram output int32 (#7202)

alanwaketan committed 1 year ago

Verified 218f4e54

Update images to support dark mode (#7195)

loganthomas committed 1 year ago

Verified cdb4aec6

Implement XLASymNodeImpl::gt (#7192)

vanbasten23 committed 1 year ago

Verified 73346d7a

Re-land: `upsample_bilinear`: fix output data-type. (#7168)

ysiraichi committed 1 year ago

Verified e563cfef

Output tensorid before erroring out for empty data handle (#7176)

JackCaoG committed 1 year ago

Verified 376d6454

Warn when using persistent cache with debug env vars (#7175)

jonb377 committed 1 year ago

Verified bcf46c7a

[torch_xla2] Use `torch.Generator` for random number generation in `Environment` (#7018)

will-cromar committed 1 year ago

Verified 94fa9993

Update index for torch_xla apis (#7171)

JackCaoG committed 1 year ago

Verified 8102ad42

Run some of the examples in TPU CI (#7172)

JackCaoG committed 1 year ago

Verified 0d754931

isolate the PJRT dependency between torch_xla and torch_xla2 (#7144)

zpcore committed 1 year ago

Verified 0f9bff34

`mul`: convert inputs to result type. (#7130)

ysiraichi committed 1 year ago

Verified 7938bb5d

Unify how we use torch commit in the CI. (#7164)

vanbasten23 committed 1 year ago

Verified 89832c3d

Reuse DLDeviceType. (#7163)

vanbasten23 committed 1 year ago

Verified 09cfdbf4

Newer Older