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

28
tests/models/instructblip/test_modeling_instructblip.py
View File

@@ -30,6 +30,8 @@ from transformers import (
InstructBlipVisionConfig,
)
from transformers.testing_utils import (
Expectations,
cleanup,
require_accelerate,
require_bitsandbytes,
require_torch,
@@ -722,6 +724,9 @@ def prepare_img():
@require_torch
@slow
class InstructBlipModelIntegrationTest(unittest.TestCase):
def tearDown(self):
cleanup(torch_device, gc_collect=False)
@require_bitsandbytes
@require_accelerate
def test_inference_vicuna_7b(self):
@@ -739,13 +744,24 @@ class InstructBlipModelIntegrationTest(unittest.TestCase):
outputs = model.generate(**inputs, max_new_tokens=30)
generated_text = processor.batch_decode(outputs, skip_special_tokens=True)[0].strip()
expected_outputs = [32001] * 32 + [2, 1724, 338, 22910, 1048, 445, 1967, 29973, 450, 22910, 9565, 310, 445, 1967, 338, 393, 263, 767, 338, 13977, 292, 22095, 373, 278, 1250, 310, 263, 13328, 20134, 29963, 1550, 19500, 373, 263, 19587, 4272, 11952, 29889] # fmt: off
expected_outputs = Expectations(
{
("xpu", 3): [32001] * 32 + [2, 1724, 338, 22910, 1048, 445, 1967, 29973, 450, 22910, 9565, 310, 445, 1967, 338, 393, 263, 767, 338, 13977, 292, 22095, 373, 278, 1250, 310, 263, 13328, 20134, 29963, 1550, 19500, 1623, 263, 19587, 4272, 11952, 29889],
("cuda", None): [32001] * 32 + [2, 1724, 338, 22910, 1048, 445, 1967, 29973, 450, 22910, 9565, 310, 445, 1967, 338, 393, 263, 767, 338, 13977, 292, 22095, 373, 278, 1250, 310, 263, 13328, 20134, 29963, 1550, 19500, 373, 263, 19587, 4272, 11952, 29889],
}
) # fmt: off
expected_output = expected_outputs.get_expectation()
self.assertEqual(outputs[0].tolist(), expected_outputs)
self.assertEqual(
generated_text,
"What is unusual about this image? The unusual aspect of this image is that a man is ironing clothes on the back of a yellow SUV while driving on a busy city street.",
)
expected_texts = Expectations(
{
("xpu", 3): "What is unusual about this image? The unusual aspect of this image is that a man is ironing clothes on the back of a yellow SUV while driving down a busy city street.",
("cuda", None): "What is unusual about this image? The unusual aspect of this image is that a man is ironing clothes on the back of a yellow SUV while driving on a busy city street.",
}
) # fmt: off
expected_text = expected_texts.get_expectation()
self.assertEqual(outputs[0].tolist(), expected_output)
self.assertEqual(generated_text, expected_text)
def test_inference_flant5_xl(self):
processor = InstructBlipProcessor.from_pretrained("Salesforce/instructblip-flan-t5-xl")