feat-unify-encode-function-0622 (#19)

- feat: make the encode_texts and encode_images support single inputs (8e178a167d8f5ea41b0e6ab2a74775363abe32ae)
- feat: update the function names (a9d6eecadd0c92a8b60b12724a5e4387d16ee327)
- feat: return a single tensor when a single image is given (2dc412733ac7e28f74080fb48d0c29062156a429)

modeling_jina_embeddings_v4.py

@@ -407,9 +407,9 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
                 )
         return task
 
-    def encode_texts(
+    def encode_text(
         self,
-        texts: List[str],
+        texts: Union[str, List[str]],
         task: Optional[str] = None,
         max_length: int = 8192,
         batch_size: int = 8,
@@ -417,12 +417,12 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
         return_numpy: bool = False,
         truncate_dim: Optional[int] = None,
         prompt_name: Optional[str] = None,
-    ) -> List[torch.Tensor]:
+    ) -> Union[List[torch.Tensor], torch.Tensor]:
         """
         Encodes a list of texts into embeddings.
 
         Args:
-            texts: List of text strings to encode
+            texts: text or list of text strings to encode
             max_length: Maximum token length for text processing
             batch_size: Number of texts to process at once
             vector_type: Type of embedding vector to generate ('single_vector' or 'multi_vector')
@@ -431,7 +431,7 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
             prompt_name: Type of text being encoded ('query' or 'passage')
 
         Returns:
-            List of text embeddings as tensors or numpy arrays
+            List of text embeddings as tensors or numpy arrays when encoding multiple texts, or single text embedding as tensor when encoding a single text
         """
         prompt_name = prompt_name or "query"
         encode_kwargs = self._validate_encoding_params(
@@ -446,6 +446,9 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
             prefix=encode_kwargs.pop("prefix"),
         )
 
+        if isinstance(texts, str):
+            texts = [texts]
+
         embeddings = self._process_batches(
             data=texts,
             processor_fn=processor_fn,
@@ -456,7 +459,7 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
             **encode_kwargs,
         )
 
-        return embeddings
+        return embeddings if len(texts) > 1 else embeddings[0]
 
     def _load_images_if_needed(
         self, images: List[Union[str, Image.Image]]
@@ -472,21 +475,21 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
             loaded_images.append(image)
         return loaded_images
 
-    def encode_images(
+    def encode_image(
         self,
-        images: List[Union[str, Image.Image]],
+        images: Union[str, Image.Image, List[Union[str, Image.Image]]],
         task: Optional[str] = None,
         batch_size: int = 8,
         vector_type: Optional[str] = None,
         return_numpy: bool = False,
         truncate_dim: Optional[int] = None,
         max_pixels: Optional[int] = None,
-    ) -> List[torch.Tensor]:
+    ) -> Union[List[torch.Tensor], torch.Tensor]:
         """
-        Encodes a list of images into embeddings.
+        Encodes a list of images or a single image into embedding(s).
 
         Args:
-            images: List of PIL images, URLs, or local file paths to encode
+            images: image(s) to encode, can be PIL Image(s), URL(s), or local file path(s)
             batch_size: Number of images to process at once
             vector_type: Type of embedding vector to generate ('single_vector' or 'multi_vector')
             return_numpy: Whether to return numpy arrays instead of torch tensors
@@ -494,7 +497,7 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
             max_pixels: Maximum number of pixels to process per image
 
         Returns:
-            List of image embeddings as tensors or numpy arrays
+            List of image embeddings as tensors or numpy arrays when encoding multiple images, or single image embedding as tensor when encoding a single image
         """
         if max_pixels:
             default_max_pixels = self.processor.image_processor.max_pixels
@@ -503,6 +506,11 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
             )
         encode_kwargs = self._validate_encoding_params(vector_type, truncate_dim)
         task = self._validate_task(task)
+        
+        # Convert single image to list
+        if isinstance(images, (str, Image.Image)):
+            images = [images]
+            
         images = self._load_images_if_needed(images)
         embeddings = self._process_batches(
             data=images,
@@ -517,7 +525,7 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
         if max_pixels:
             self.processor.image_processor.max_pixels = default_max_pixels
 
-        return embeddings
+        return embeddings if len(images) > 1 else embeddings[0]
 
     @classmethod
     def from_pretrained(