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

[processor] clean up mulitmodal tests (#37362)

Browse Source 
* clkea up mulitmodal processor tests

* fixup

* fix tests

* fix one last test

* forgot
This commit is contained in:
Raushan Turganbay
2025-04-11 13:32:19 +02:00
committed by GitHub
parent 3c39c07939
commit a563999a02
30 changed files with 304 additions and 817 deletions
Download Patch File Download Diff File Expand all files Collapse all files

136
tests/models/idefics/test_processor_idefics.py
View File

@@ -219,139 +219,3 @@ class IdeficsProcessorTest(ProcessorTesterMixin, unittest.TestCase):
# For now the processor supports only ['pixel_values', 'input_ids', 'attention_mask']
self.assertSetEqual(set(inputs.keys()), set(self.input_keys))
# Override the following tests as Idefics image processor does not accept do_rescale and rescale_factor
@require_torch
@require_vision
def test_image_processor_defaults_preserved_by_image_kwargs(self):
if "image_processor" not in self.processor_class.attributes:
self.skipTest(f"image_processor attribute not present in {self.processor_class}")
image_processor = self.get_component("image_processor", image_size=234)
tokenizer = self.get_component("tokenizer", max_length=117)
processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
self.skip_processor_without_typed_kwargs(processor)
input_str = self.prepare_text_inputs()
image_input = self.prepare_image_inputs()
inputs = processor(text=input_str, images=image_input)
self.assertEqual(len(inputs["pixel_values"][0][0][0]), 234)
@require_torch
@require_vision
def test_kwargs_overrides_default_image_processor_kwargs(self):
if "image_processor" not in self.processor_class.attributes:
self.skipTest(f"image_processor attribute not present in {self.processor_class}")
image_processor = self.get_component("image_processor", image_size=234)
tokenizer = self.get_component("tokenizer", max_length=117)
processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
self.skip_processor_without_typed_kwargs(processor)
input_str = self.prepare_text_inputs()
image_input = self.prepare_image_inputs()
inputs = processor(text=input_str, images=image_input, image_size=224)
self.assertEqual(len(inputs["pixel_values"][0][0][0]), 224)
@require_torch
@require_vision
def test_unstructured_kwargs(self):
if "image_processor" not in self.processor_class.attributes:
self.skipTest(f"image_processor attribute not present in {self.processor_class}")
image_processor = self.get_component("image_processor")
tokenizer = self.get_component("tokenizer")
processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
self.skip_processor_without_typed_kwargs(processor)
input_str = self.prepare_text_inputs()
image_input = self.prepare_image_inputs()
inputs = processor(
text=input_str,
images=image_input,
return_tensors="pt",
image_size=214,
padding="max_length",
max_length=76,
)
self.assertEqual(inputs["pixel_values"].shape[3], 214)
self.assertEqual(len(inputs["input_ids"][0]), 76)
@require_torch
@require_vision
def test_unstructured_kwargs_batched(self):
if "image_processor" not in self.processor_class.attributes:
self.skipTest(f"image_processor attribute not present in {self.processor_class}")
image_processor = self.get_component("image_processor")
tokenizer = self.get_component("tokenizer")
processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
self.skip_processor_without_typed_kwargs(processor)
input_str = self.prepare_text_inputs(batch_size=2)
image_input = self.prepare_image_inputs(batch_size=2)
inputs = processor(
text=input_str,
images=image_input,
return_tensors="pt",
image_size=214,
padding="longest",
max_length=76,
)
self.assertEqual(inputs["pixel_values"].shape[3], 214)
self.assertEqual(len(inputs["input_ids"][0]), 8)
@require_torch
@require_vision
def test_structured_kwargs_nested(self):
if "image_processor" not in self.processor_class.attributes:
self.skipTest(f"image_processor attribute not present in {self.processor_class}")
image_processor = self.get_component("image_processor")
tokenizer = self.get_component("tokenizer")
processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
self.skip_processor_without_typed_kwargs(processor)
input_str = self.prepare_text_inputs()
image_input = self.prepare_image_inputs()
# Define the kwargs for each modality
all_kwargs = {
"common_kwargs": {"return_tensors": "pt"},
"images_kwargs": {"image_size": 214},
"text_kwargs": {"padding": "max_length", "max_length": 76},
}
inputs = processor(text=input_str, images=image_input, **all_kwargs)
self.skip_processor_without_typed_kwargs(processor)
self.assertEqual(inputs["pixel_values"].shape[3], 214)
self.assertEqual(len(inputs["input_ids"][0]), 76)
@require_torch
@require_vision
def test_structured_kwargs_nested_from_dict(self):
if "image_processor" not in self.processor_class.attributes:
self.skipTest(f"image_processor attribute not present in {self.processor_class}")
image_processor = self.get_component("image_processor")
tokenizer = self.get_component("tokenizer")
processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
self.skip_processor_without_typed_kwargs(processor)
input_str = self.prepare_text_inputs()
image_input = self.prepare_image_inputs()
# Define the kwargs for each modality
all_kwargs = {
"common_kwargs": {"return_tensors": "pt"},
"images_kwargs": {"image_size": 214},
"text_kwargs": {"padding": "max_length", "max_length": 76},
}
inputs = processor(text=input_str, images=image_input, **all_kwargs)
self.assertEqual(inputs["pixel_values"].shape[3], 214)
self.assertEqual(len(inputs["input_ids"][0]), 76)