Fix llava_onevision tests (#37280)

* Fix llava_onevision tests

* Trigger tests

Author: Rocketknight1

tests/models/llava_onevision/test_processor_llava_onevision.py

@@ -39,17 +39,19 @@
 class LlavaOnevisionProcessorTest(ProcessorTesterMixin, unittest.TestCase):
     processor_class = LlavaOnevisionProcessor
 
-    def setUp(self):
-        self.tmpdirname = tempfile.mkdtemp()
+    @classmethod
+    def setUpClass(cls):
+        cls.tmpdirname = tempfile.mkdtemp()
+        cls.addClassCleanup(lambda tempdir=cls.tmpdirname: shutil.rmtree(tempdir))
         image_processor = LlavaOnevisionImageProcessor()
         video_processor = LlavaOnevisionVideoProcessor()
         tokenizer = Qwen2TokenizerFast.from_pretrained("Qwen/Qwen2-0.5B-Instruct")
-        processor_kwargs = self.prepare_processor_dict()
+        processor_kwargs = cls.prepare_processor_dict()
 
         processor = LlavaOnevisionProcessor(
             video_processor=video_processor, image_processor=image_processor, tokenizer=tokenizer, **processor_kwargs
         )
-        processor.save_pretrained(self.tmpdirname)
+        processor.save_pretrained(cls.tmpdirname)
 
     def get_tokenizer(self, **kwargs):
         return AutoProcessor.from_pretrained(self.tmpdirname, **kwargs).tokenizer
@@ -60,7 +62,8 @@ def get_image_processor(self, **kwargs):
     def get_video_processor(self, **kwargs):
         return AutoProcessor.from_pretrained(self.tmpdirname, **kwargs).video_processor
 
-    def prepare_processor_dict(self):
+    @staticmethod
+    def prepare_processor_dict():
         return {
             "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + ' '}}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>' }}{% endfor %}{# Render all video then #}{% for content in message['content'] | selectattr('type', 'equalto', 'video') %}{{ '<video>' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ '\n' + content['text'] }}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ '\n' + content['text'] }}{% endgeneration %}{% endfor %}{% endif %}{{'<|im_end|>'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
             "num_image_tokens": 6,
@@ -88,9 +91,6 @@ def test_chat_template_is_saved(self):
         processor_dict = self.prepare_processor_dict()
         self.assertTrue(processor_loaded.chat_template == processor_dict.get("chat_template", None))
 
-    def tearDown(self):
-        shutil.rmtree(self.tmpdirname)
-
     def test_chat_template(self):
         processor = AutoProcessor.from_pretrained("llava-hf/llava-onevision-qwen2-7b-ov-hf")
         expected_prompt = "<|im_start|>user <image>\nWhat is shown in this image?<|im_end|><|im_start|>assistant\n"