SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

HPU support (#36424)

Browse Source 
* test

* fix

* fix

* skip some and run some first

* test fsdp

* fix

* patches for generate

* test distributed

* copy

* don't test distributed loss for hpu

* require fp16 and run first

* changes from marc's PR fixing zero3

* better alternative

* return True when fp16 support on gaudi without creating bridge

* fix

* fix tested dtype in deepspeed inference test

* test

* fix

* test

* fix

* skip

* require fp16

* run first fsdp

* Apply suggestions from code review

* address comments

* address comments and refactor test

* reduce precison

* avoid doing gaudi1 specific stuff in the genreation loop

* document test_gradient_accumulation_loss_alignment_with_model_loss test a bit more
This commit is contained in:
Ilyas Moutawwakil
2025-03-12 09:08:12 +01:00
committed by GitHub
parent 50d3530aa0
commit 89f6956015
19 changed files with 337 additions and 139 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/test_modeling_common.py
View File

@@ -2770,7 +2770,7 @@ class ModelTesterMixin:
elif param_device in ["mps"]:
self.assertEqual(param.device, torch.device("mps"))
else:
# when loaded with device_map, `param_device` are integer values for cuda/xpu/npu/mlu
# when loaded with device_map, `param_device` are integer values for cuda/xpu/hpu/npu/mlu
self.assertEqual(param.device, torch.device(f"{torch_device}:{param_device}"))
@require_accelerate