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

fix UT failures on XPU w/ stock PyTorch 2.7 & 2.8 (#39116)

Browse Source 
* fix UT failures on XPU w/ stock PyTorch 2.7 & 2.8

Signed-off-by: YAO Matrix <matrix.yao@intel.com>

* zamba2

Signed-off-by: YAO Matrix <matrix.yao@intel.com>

* xx

Signed-off-by: YAO Matrix <matrix.yao@intel.com>

* internvl

Signed-off-by: YAO Matrix <matrix.yao@intel.com>

* tp cases

Signed-off-by: YAO Matrix <matrix.yao@intel.com>

---------

Signed-off-by: YAO Matrix <matrix.yao@intel.com>
This commit is contained in:
Yao Matrix
2025-06-30 17:49:03 +08:00
committed by GitHub
parent ccf2ca162e
commit 2100ee6545
8 changed files with 119 additions and 51 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tests/quantization/bnb/test_4bit.py
View File

@@ -520,14 +520,14 @@ class Pipeline4BitTest(Base4bitTest):
@require_torch_multi_accelerator
@apply_skip_if_not_implemented
class Bnb4bitTestMultiGpu(Base4bitTest):
class Bnb4bitTestMultiAccelerator(Base4bitTest):
def setUp(self):
super().setUp()
def test_multi_gpu_loading(self):
def test_multi_accelerator_loading(self):
r"""
This tests that the model has been loaded and can be used correctly on a multi-GPU setup.
Let's just try to load a model on 2 GPUs and see if it works. The model we test has ~2GB of total, 3GB should suffice
This tests that the model has been loaded and can be used correctly on a multi-accelerator setup.
Let's just try to load a model on 2 accelerators and see if it works. The model we test has ~2GB of total, 3GB should suffice
"""
device_map = {
"transformer.word_embeddings": 0,