Restructure repo

README.md

@@ -1,5 +1,3 @@
 ---
 license: mit
-tags:
-  - custom-handler
 ---

handler.py

@@ -7,31 +7,22 @@ from io import BytesIO
 
 class EndpointHandler:
     """
-    A handler class for processing image data and generating embeddings using a specified model and processor.
+    A handler class for processing image data, generating embeddings using a specified model and processor.
 
     Attributes:
-        model (:obj:):
-            The pre-trained model used for generating embeddings.
-        processor (:obj:):
-            The pre-trained processor used to process images before model inference.
-        device (:obj:):
-            The device (CPU or CUDA) used to run model inference.
-        default_batch_size (:obj:int:):
-            The default batch size for processing images in batches.
+        model: The pre-trained model used for generating embeddings.
+        processor: The pre-trained processor used to process images before model inference.
+        device: The device (CPU or CUDA) used to run model inference.
+        default_batch_size: The default batch size for processing images in batches.
     """
 
-    def __init__(self, path: str = "", default_batch_size: int = 4):
+    def __init__(self, path: str = "./model", default_batch_size: int = 4):
         """
         Initializes the EndpointHandler with a specified model path and default batch size.
 
         Args:
-            path (:obj:`str`, optional):
-                Path to the pre-trained model and processor.
-            default_batch_size (:obj:`int`, optional):
-                Default batch size for image processing.
-
-        Return:
-            None
+            path (str): Path to the pre-trained model and processor.
+            default_batch_size (int): Default batch size for image processing.
         """
         from colpali_engine.models import ColQwen2, ColQwen2Processor
 
@@ -50,11 +41,10 @@ class EndpointHandler:
         Processes a batch of images and generates embeddings.
 
         Args:
-            images (:obj:`List[Image.Image]`):
-                List of images to process.
+            images (List[Image.Image]): List of images to process.
 
-        Return:
-            A :obj:`List[List[float]]`. A list of embeddings for each image, where each embedding is a list of floats.
+        Returns:
+            List[List[float]]: List of embeddings for each image.
         """
         batch_images = self.processor.process_images(images)
         batch_images = {k: v.to(self.device) for k, v in batch_images.items()}
@@ -69,12 +59,10 @@ class EndpointHandler:
         Processes input data containing base64-encoded images, decodes them, and generates embeddings.
 
         Args:
-            data (:obj:`Dict[str, Any]`):
-                Includes the input data and the parameters for the inference, such as "inputs" containing a list of base64-encoded images and an optional "batch_size".
+            data (Dict[str, Any]): Dictionary containing input images and optional batch size.
 
-        Return:
-            A :obj:`dict`. The object returned should be a dict like {"embeddings": [[0.6331314444541931, 0.8802216053009033, ..., -0.7866355180740356]]} containing:
-                - "embeddings": A list of lists, where each inner list is a set of floats corresponding to the embeddings of each image.
+        Returns:
+            Dict[str, Any]: Dictionary containing generated embeddings or error messages.
         """
         images_data = data.get("inputs", [])
         batch_size = data.get("batch_size", self.default_batch_size)