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

59
tests/models/wav2vec2_bert/test_processor_wav2vec2_bert.py
View File

@@ -32,11 +32,12 @@ class Wav2Vec2BertProcessorTest(ProcessorTesterMixin, unittest.TestCase):
processor_class = Wav2Vec2BertProcessor
text_input_name = "labels"
def setUp(self):
@classmethod
def setUpClass(cls):
vocab = "<pad> <s> </s> <unk> | E T A O N I H S R D L U M W C F G Y P B V K ' X J Q Z".split(" ")
vocab_tokens = dict(zip(vocab, range(len(vocab))))
self.add_kwargs_tokens_map = {
cls.add_kwargs_tokens_map = {
"pad_token": "<pad>",
"unk_token": "<unk>",
"bos_token": "<s>",
@@ -50,28 +51,30 @@ class Wav2Vec2BertProcessorTest(ProcessorTesterMixin, unittest.TestCase):
"do_normalize": True,
}
self.tmpdirname = tempfile.mkdtemp()
self.vocab_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["vocab_file"])
self.feature_extraction_file = os.path.join(self.tmpdirname, FEATURE_EXTRACTOR_NAME)
with open(self.vocab_file, "w", encoding="utf-8") as fp:
cls.tmpdirname = tempfile.mkdtemp()
cls.vocab_file = os.path.join(cls.tmpdirname, VOCAB_FILES_NAMES["vocab_file"])
cls.feature_extraction_file = os.path.join(cls.tmpdirname, FEATURE_EXTRACTOR_NAME)
with open(cls.vocab_file, "w", encoding="utf-8") as fp:
fp.write(json.dumps(vocab_tokens) + "\n")
with open(self.feature_extraction_file, "w", encoding="utf-8") as fp:
with open(cls.feature_extraction_file, "w", encoding="utf-8") as fp:
fp.write(json.dumps(feature_extractor_map) + "\n")
tokenizer = self.get_tokenizer()
tokenizer.save_pretrained(self.tmpdirname)
tokenizer = cls.get_tokenizer()
tokenizer.save_pretrained(cls.tmpdirname)
def get_tokenizer(self, **kwargs_init):
kwargs = self.add_kwargs_tokens_map.copy()
@classmethod
def get_tokenizer(cls, **kwargs_init):
kwargs = cls.add_kwargs_tokens_map.copy()
kwargs.update(kwargs_init)
return Wav2Vec2CTCTokenizer.from_pretrained(self.tmpdirname, **kwargs)
return Wav2Vec2CTCTokenizer.from_pretrained(cls.tmpdirname, **kwargs)
def get_feature_extractor(self, **kwargs):
return SeamlessM4TFeatureExtractor.from_pretrained(self.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):
tokenizer = self.get_tokenizer()
@@ -79,8 +82,9 @@ class Wav2Vec2BertProcessorTest(ProcessorTesterMixin, unittest.TestCase):
processor = Wav2Vec2BertProcessor(tokenizer=tokenizer, feature_extractor=feature_extractor)
processor.save_pretrained(self.tmpdirname)
processor = Wav2Vec2BertProcessor.from_pretrained(self.tmpdirname)
with tempfile.TemporaryDirectory() as tmpdir:
processor.save_pretrained(tmpdir)
processor = Wav2Vec2BertProcessor.from_pretrained(tmpdir)
self.assertEqual(processor.tokenizer.get_vocab(), tokenizer.get_vocab())
self.assertIsInstance(processor.tokenizer, Wav2Vec2CTCTokenizer)
@@ -89,17 +93,22 @@ class Wav2Vec2BertProcessorTest(ProcessorTesterMixin, unittest.TestCase):
self.assertIsInstance(processor.feature_extractor, SeamlessM4TFeatureExtractor)
def test_save_load_pretrained_additional_features(self):
processor = Wav2Vec2BertProcessor(
tokenizer=self.get_tokenizer(), feature_extractor=self.get_feature_extractor()
)
processor.save_pretrained(self.tmpdirname)
with tempfile.TemporaryDirectory() as tmpdir:
processor = Wav2Vec2BertProcessor(
tokenizer=self.get_tokenizer(), feature_extractor=self.get_feature_extractor()
)
processor.save_pretrained(tmpdir)
tokenizer_add_kwargs = self.get_tokenizer(bos_token="(BOS)", eos_token="(EOS)")
feature_extractor_add_kwargs = self.get_feature_extractor(do_normalize=False, padding_value=1.0)
tokenizer_add_kwargs = Wav2Vec2CTCTokenizer.from_pretrained(
tmpdir, **(self.add_kwargs_tokens_map | {"bos_token": "(BOS)", "eos_token": "(EOS)"})
)
feature_extractor_add_kwargs = SeamlessM4TFeatureExtractor.from_pretrained(
tmpdir, do_normalize=False, padding_value=1.0
)
processor = Wav2Vec2BertProcessor.from_pretrained(
self.tmpdirname, bos_token="(BOS)", eos_token="(EOS)", do_normalize=False, padding_value=1.0
)
processor = Wav2Vec2BertProcessor.from_pretrained(
tmpdir, bos_token="(BOS)", eos_token="(EOS)", do_normalize=False, padding_value=1.0
)
self.assertEqual(processor.tokenizer.get_vocab(), tokenizer_add_kwargs.get_vocab())
self.assertIsInstance(processor.tokenizer, Wav2Vec2CTCTokenizer)