add tests for video llava

src/transformers/models/video_llava/processing_video_llava.py

@@ -136,6 +136,8 @@ def __call__(
             tokenizer_init_kwargs=self.tokenizer.init_kwargs,
             **kwargs,
         )
+        # Temporary fix for "paddding_side" in init_kwargs
+        _ = output_kwargs["text_kwargs"].pop("padding_side", None)
 
         data = {}
         if images is not None or videos is not None:

tests/models/video_llava/test_processor_video_llava.py

@@ -0,0 +1,17 @@
+import tempfile
+import unittest
+
+from transformers.models.video_llava.processing_video_llava import VideoLlavaProcessor
+
+from ...test_processing_common import ProcessorTesterMixin
+
+
+class VideoLlavaProcessorTest(ProcessorTesterMixin, unittest.TestCase):
+    from_pretrained_id = "LanguageBind/Video-LLaVA-7B-hf"
+    processor_class = VideoLlavaProcessor
+    images_data_arg_name = "pixel_values_images"
+
+    def setUp(self):
+        self.tmpdirname = tempfile.mkdtemp()
+        processor = self.processor_class.from_pretrained(self.from_pretrained_id)
+        processor.save_pretrained(self.tmpdirname)

tests/test_processing_common.py

@@ -242,8 +242,7 @@ def test_unstructured_kwargs_batched(self):
         )
 
         self.assertEqual(inputs[self.images_data_arg_name].shape[2], 214)
-
-        self.assertEqual(len(inputs[self.text_data_arg_name][0]), 6)
+        self.assertEqual(len(inputs[self.text_data_arg_name][0]), len(inputs[self.text_data_arg_name][1]))
 
     @require_torch
     @require_vision