vision-agent 0.2.188__tar.gz → 0.2.190__tar.gz

{vision_agent-0.2.188 → vision_agent-0.2.190}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vision-agent
-Version: 0.2.188
+Version: 0.2.190
 Summary: Toolset for Vision Agent
 Author: Landing AI
 Author-email: dev@landing.ai

{vision_agent-0.2.188 → vision_agent-0.2.190}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "vision-agent"
-version = "0.2.188"
+version = "0.2.190"
 description = "Toolset for Vision Agent"
 authors = ["Landing AI <dev@landing.ai>"]
 readme = "README.md"

{vision_agent-0.2.188 → vision_agent-0.2.190}/vision_agent/tools/__init__.py

@@ -69,6 +69,7 @@ from .tools import (
     qwen2_vl_video_vqa,
     video_temporal_localization,
     flux_image_inpainting,
+    siglip_classification,
 )
 
 __new_tools__ = [

{vision_agent-0.2.188 → vision_agent-0.2.190}/vision_agent/tools/tools.py

@@ -1845,6 +1845,48 @@ def flux_image_inpainting(
     return output_image
 
 
+def siglip_classification(image: np.ndarray, labels: List[str]) -> Dict[str, Any]:
+    """'siglip_classification' is a tool that can classify an image or a cropped detection given a list
+    of input labels or tags. It returns the same list of the input labels along with
+    their probability scores based on image content.
+
+    Parameters:
+        image (np.ndarray): The image to classify or tag
+        labels (List[str]): The list of labels or tags that is associated with the image
+
+    Returns:
+        Dict[str, Any]: A dictionary containing the labels and scores. One dictionary
+            contains a list of given labels and other a list of scores.
+
+    Example
+    -------
+        >>> siglip_classification(image, ['dog', 'cat', 'bird'])
+        {"labels": ["dog", "cat", "bird"], "scores": [0.68, 0.30, 0.02]},
+    """
+
+    if image.shape[0] < 1 or image.shape[1] < 1:
+        return {"labels": [], "scores": []}
+
+    image_file = numpy_to_bytes(image)
+
+    files = [("image", image_file)]
+
+    payload = {
+        "model": "siglip",
+        "labels": labels,
+    }
+
+    response: dict[str, Any] = send_inference_request(
+        payload=payload,
+        endpoint_name="classification",
+        files=files,
+        v2=True,
+        metadata_payload={"function_name": "siglip_classification"},
+    )
+
+    return response
+
+
 # Utility and visualization functions
 
 
@@ -2362,6 +2404,7 @@ FUNCTION_TOOLS = [
     qwen2_vl_images_vqa,
     qwen2_vl_video_vqa,
     video_temporal_localization,
+    flux_image_inpainting,
 ]
 
 UTIL_TOOLS = [