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

🚨 🚨 Setup -> setupclass conversion (#37282)

Browse Source 
* More limited setup -> setupclass conversion

* make fixup

* Trigger tests

* Fixup UDOP

* Missed a spot

* tearDown -> tearDownClass where appropriate

* Couple more class fixes

* Fixups for UDOP and VisionTextDualEncoder

* Ignore errors when removing the tmpdir, in case it already got cleaned up somewhere

* CLIP fixes

* More correct classmethods

* Wav2Vec2Bert fixes

* More methods become static

* More class methods

* More class methods

* Revert changes for integration tests / modeling files

* Use a different tempdir for tests that actually write to it

* Remove addClassCleanup and just use teardownclass

* Remove changes in modeling files

* Cleanup get_processor_dict() for got_ocr2

* Fix regression on Wav2Vec2BERT test that was masked by this before

* Rework tests that modify the tmpdir

* make fix-copies

* revert clvp modeling test changes

* Fix CLIP processor test

* make fix-copies
This commit is contained in:
Matt
2025-04-08 17:15:37 +01:00
committed by GitHub
parent c15a7adb28
commit 4d0de5f73a
49 changed files with 740 additions and 574 deletions
Download Patch File Download Diff File Expand all files Collapse all files

74
tests/models/layoutxlm/test_processor_layoutxlm.py
View File

@@ -46,40 +46,46 @@ class LayoutXLMProcessorTest(ProcessorTesterMixin, unittest.TestCase):
rust_tokenizer_class = LayoutXLMTokenizerFast
processor_class = LayoutXLMProcessor
def setUp(self):
@classmethod
def setUpClass(cls):
image_processor_map = {
"do_resize": True,
"size": 224,
"apply_ocr": True,
}
self.tmpdirname = tempfile.mkdtemp()
self.feature_extraction_file = os.path.join(self.tmpdirname, FEATURE_EXTRACTOR_NAME)
with open(self.feature_extraction_file, "w", encoding="utf-8") as fp:
cls.tmpdirname = tempfile.mkdtemp()
cls.feature_extraction_file = os.path.join(cls.tmpdirname, FEATURE_EXTRACTOR_NAME)
with open(cls.feature_extraction_file, "w", encoding="utf-8") as fp:
fp.write(json.dumps(image_processor_map) + "\n")
# taken from `test_tokenization_layoutxlm.LayoutXLMTokenizationTest.test_save_pretrained`
self.tokenizer_pretrained_name = "hf-internal-testing/tiny-random-layoutxlm"
cls.tokenizer_pretrained_name = "hf-internal-testing/tiny-random-layoutxlm"
tokenizer = self.get_tokenizer()
image_processor = self.get_image_processor()
tokenizer = cls.get_tokenizer()
image_processor = cls.get_image_processor()
processor = LayoutXLMProcessor(tokenizer=tokenizer, image_processor=image_processor)
processor.save_pretrained(self.tmpdirname)
processor.save_pretrained(cls.tmpdirname)
def get_tokenizer(self, **kwargs) -> PreTrainedTokenizer:
return self.tokenizer_class.from_pretrained(self.tokenizer_pretrained_name, **kwargs)
@classmethod
def get_tokenizer(cls, **kwargs) -> PreTrainedTokenizer:
return cls.tokenizer_class.from_pretrained(cls.tokenizer_pretrained_name, **kwargs)
def get_rust_tokenizer(self, **kwargs) -> PreTrainedTokenizerFast:
return self.rust_tokenizer_class.from_pretrained(self.tokenizer_pretrained_name, **kwargs)
@classmethod
def get_rust_tokenizer(cls, **kwargs) -> PreTrainedTokenizerFast:
return cls.rust_tokenizer_class.from_pretrained(cls.tokenizer_pretrained_name, **kwargs)
def get_tokenizers(self, **kwargs) -> list[PreTrainedTokenizerBase]:
return [self.get_tokenizer(**kwargs), self.get_rust_tokenizer(**kwargs)]
@classmethod
def get_tokenizers(cls, **kwargs) -> list[PreTrainedTokenizerBase]:
return [cls.get_tokenizer(**kwargs), cls.get_rust_tokenizer(**kwargs)]
def get_image_processor(self, **kwargs):
return LayoutLMv2ImageProcessor.from_pretrained(self.tmpdirname, **kwargs)
@classmethod
def get_image_processor(cls, **kwargs):
return LayoutLMv2ImageProcessor.from_pretrained(cls.tmpdirname, **kwargs)
def tearDown(self):
shutil.rmtree(self.tmpdirname)
@classmethod
def tearDownClass(cls):
shutil.rmtree(cls.tmpdirname, ignore_errors=True)
def test_save_load_pretrained_default(self):
image_processor = self.get_image_processor()
@@ -87,8 +93,9 @@ class LayoutXLMProcessorTest(ProcessorTesterMixin, unittest.TestCase):
for tokenizer in tokenizers:
processor = LayoutXLMProcessor(image_processor=image_processor, tokenizer=tokenizer)
processor.save_pretrained(self.tmpdirname)
processor = LayoutXLMProcessor.from_pretrained(self.tmpdirname)
with tempfile.TemporaryDirectory() as tmpdir:
processor.save_pretrained(tmpdir)
processor = LayoutXLMProcessor.from_pretrained(tmpdir)
self.assertEqual(processor.tokenizer.get_vocab(), tokenizer.get_vocab())
self.assertIsInstance(processor.tokenizer, (LayoutXLMTokenizer, LayoutXLMTokenizerFast))
@@ -97,21 +104,22 @@ class LayoutXLMProcessorTest(ProcessorTesterMixin, unittest.TestCase):
self.assertIsInstance(processor.image_processor, LayoutLMv2ImageProcessor)
def test_save_load_pretrained_additional_features(self):
processor = LayoutXLMProcessor(image_processor=self.get_image_processor(), tokenizer=self.get_tokenizer())
processor.save_pretrained(self.tmpdirname)
with tempfile.TemporaryDirectory() as tmpdir:
processor = LayoutXLMProcessor(image_processor=self.get_image_processor(), tokenizer=self.get_tokenizer())
processor.save_pretrained(tmpdir)
# slow tokenizer
tokenizer_add_kwargs = self.get_tokenizer(bos_token="(BOS)", eos_token="(EOS)")
image_processor_add_kwargs = self.get_image_processor(do_resize=False, size=30)
# slow tokenizer
tokenizer_add_kwargs = self.get_tokenizer(bos_token="(BOS)", eos_token="(EOS)")
image_processor_add_kwargs = self.get_image_processor(do_resize=False, size=30)
processor = LayoutXLMProcessor.from_pretrained(
self.tmpdirname,
use_fast=False,
bos_token="(BOS)",
eos_token="(EOS)",
do_resize=False,
size=30,
)
processor = LayoutXLMProcessor.from_pretrained(
tmpdir,
use_fast=False,
bos_token="(BOS)",
eos_token="(EOS)",
do_resize=False,
size=30,
)
self.assertEqual(processor.tokenizer.get_vocab(), tokenizer_add_kwargs.get_vocab())
self.assertIsInstance(processor.tokenizer, LayoutXLMTokenizer)