try to past test

Former-commit-id: 3b6bfae0e5fe795a70d530b2765f27d95c5862f8

src/llamafactory/data/mm_plugin.py

@@ -296,11 +296,11 @@ class LlavaPlugin(BasePlugin):
 class LlavaNextPlugin(BasePlugin):
     @override
     def process_messages(
-            self,
-            messages: Sequence[Dict[str, str]],
-            images: Sequence["ImageInput"],
-            videos: Sequence["VideoInput"],
-            processor: Optional["ProcessorMixin"],
+        self,
+        messages: Sequence[Dict[str, str]],
+        images: Sequence["ImageInput"],
+        videos: Sequence["VideoInput"],
+        processor: Optional["ProcessorMixin"],
     ) -> List[Dict[str, str]]:
         self._validate_input(images, videos)
         num_image_tokens = 0
@@ -318,13 +318,13 @@ class LlavaNextPlugin(BasePlugin):
 
     @override
     def get_mm_inputs(
-            self,
-            images: Sequence["ImageInput"],
-            videos: Sequence["VideoInput"],
-            imglens: Sequence[int],
-            vidlens: Sequence[int],
-            seqlens: Sequence[int],
-            processor: Optional["ProcessorMixin"],
+        self,
+        images: Sequence["ImageInput"],
+        videos: Sequence["VideoInput"],
+        imglens: Sequence[int],
+        vidlens: Sequence[int],
+        seqlens: Sequence[int],
+        processor: Optional["ProcessorMixin"],
     ) -> Dict[str, Union[List[int], "torch.Tensor"]]:
         self._validate_input(images, videos)
         return _get_mm_inputs(images, videos, processor)
@@ -379,6 +379,7 @@ class LlavaNextVideoPlugin(BasePlugin):
             res.update(video_res)
         return res
 
+
 class PaliGemmaPlugin(BasePlugin):
     @override
     def process_messages(

src/llamafactory/model/loader.py

@@ -120,9 +120,12 @@ def load_config(model_args: "ModelArguments") -> "PretrainedConfig":
     """
     init_kwargs = _get_init_kwargs(model_args)
     if "LLaVA-NeXT-Video" in model_args.model_name_or_path:
-        from transformers import PretrainedConfig, LlavaNextVideoConfig, CLIPVisionConfig, LlamaConfig
+        from transformers import CLIPVisionConfig, LlamaConfig, LlavaNextVideoConfig, PretrainedConfig
+
         official_config = PretrainedConfig.from_pretrained(model_args.model_name_or_path, **init_kwargs)
-        config = LlavaNextVideoConfig(CLIPVisionConfig(**official_config.vision_config), LlamaConfig(**official_config.text_config))
+        config = LlavaNextVideoConfig(
+            CLIPVisionConfig(**official_config.vision_config), LlamaConfig(**official_config.text_config)
+        )
         setattr(config, "visual_inputs", True)
         return config
     return AutoConfig.from_pretrained(model_args.model_name_or_path, **init_kwargs)
@@ -163,6 +166,7 @@ def load_model(
                 load_class = AutoModelForCausalLM
                 if "llava_next_video" == getattr(config, "model_type"):
                     from transformers import LlavaNextVideoForConditionalGeneration
+
                     load_class = LlavaNextVideoForConditionalGeneration
 
             if model_args.train_from_scratch:

src/llamafactory/model/model_utils/visual.py

@@ -108,7 +108,13 @@ def configure_visual_model(config: "PretrainedConfig") -> None:
     Patches VLMs before loading them.
     """
     model_type = getattr(config, "model_type", None)
-    if model_type in ["llava", "llava_next", "video_llava", "idefics2", "llava_next_video"]:  # required for ds zero3 and valuehead models
+    if model_type in [
+        "llava",
+        "llava_next",
+        "video_llava",
+        "idefics2",
+        "llava_next_video",
+    ]:  # required for ds zero3 and valuehead models
         setattr(config, "hidden_size", getattr(config.text_config, "hidden_size", None))
 
     if getattr(config, "is_yi_vl_derived_model", None):