vision-agent 0.2.216__py3-none-any.whl → 0.2.217__py3-none-any.whl

vision_agent/tools/__init__.py

@@ -33,6 +33,7 @@ from .tools import (
     depth_anything_v2,
     detr_segmentation,
     document_extraction,
+    document_qa,
     extract_frames_and_timestamps,
     florence2_ocr,
     florence2_phrase_grounding,

vision_agent/tools/tools.py

@@ -2174,6 +2174,77 @@ def document_extraction(image: np.ndarray) -> Dict[str, Any]:
     return data
 
 
+def document_qa(
+    prompt: str,
+    image: np.ndarray,
+) -> str:
+    """'document_qa' is a tool that can answer any questions about arbitrary
+    images of documents or presentations. It answers by analyzing the contextual document data
+    and then using a model to answer specific questions. It returns text as an answer to the question.
+
+    Parameters:
+        prompt (str): The question to be answered about the document image
+        image (np.ndarray): The document image to analyze
+
+    Returns:
+        str: The answer to the question based on the document's context.
+
+    Example
+    -------
+        >>> document_qa(image, question)
+        'The answer to the question ...'
+    """
+
+    image_file = numpy_to_bytes(image)
+
+    files = [("image", image_file)]
+
+    payload = {
+        "model": "document-analysis",
+    }
+
+    data: dict[str, Any] = send_inference_request(
+        payload=payload,
+        endpoint_name="document-analysis",
+        files=files,
+        v2=True,
+        metadata_payload={"function_name": "document_qa"},
+    )
+
+    def normalize(data: Any) -> Dict[str, Any]:
+        if isinstance(data, Dict):
+            if "bbox" in data:
+                data["bbox"] = normalize_bbox(data["bbox"], image.shape[:2])
+            for key in data:
+                data[key] = normalize(data[key])
+        elif isinstance(data, List):
+            for i in range(len(data)):
+                data[i] = normalize(data[i])
+        return data  # type: ignore
+
+    data = normalize(data)
+
+    prompt = f"""
+    Document Context:
+    {data}\n
+    Question: {prompt}\n
+    Please provide a clear, concise answer using only the information from the document. If the answer is not definitively contained in the document, say "I cannot find the answer in the provided document."
+    """
+
+    lmm = AnthropicLMM()
+    llm_output = lmm.generate(prompt=prompt)
+    llm_output = cast(str, llm_output)
+
+    _display_tool_trace(
+        document_qa.__name__,
+        payload,
+        llm_output,
+        files,
+    )
+
+    return llm_output
+
+
 # Utility and visualization functions
 
 

{vision_agent-0.2.216.dist-info → vision_agent-0.2.217.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vision-agent
-Version: 0.2.216
+Version: 0.2.217
 Summary: Toolset for Vision Agent
 Author: Landing AI
 Author-email: dev@landing.ai

{vision_agent-0.2.216.dist-info → vision_agent-0.2.217.dist-info}/RECORD

@@ -26,12 +26,12 @@ vision_agent/fonts/default_font_ch_en.ttf,sha256=1YM0Z3XqLDjSNbF7ihQFSAIUdjF9m1r
 vision_agent/lmm/__init__.py,sha256=jyY1sJb_tYKg5-Wzs3p1lvwFkc-aUNZfMcLy3TOC4Zg,100
 vision_agent/lmm/lmm.py,sha256=x_nIyDNDZwq4-pfjnJTmcyyJZ2_B7TjkA5jZp88YVO8,17103
 vision_agent/lmm/types.py,sha256=ZEXR_ptBL0ZwDMTDYkgxUCmSZFmBYPQd2jreNzr_8UY,221
-vision_agent/tools/__init__.py,sha256=qzAqY2WnRLoClz3qiNtupkLtvpPlcGa5ZUCIs21WS7k,2795
+vision_agent/tools/__init__.py,sha256=Jdq34jMw_KuYwk4Wexqm4DRjuLLoL1Q8wukm0NBv1Tc,2812
 vision_agent/tools/meta_tools.py,sha256=TPeS7QWnc_PmmU_ndiDT03dXbQ5yDSP33E7U8cSj7Ls,28660
 vision_agent/tools/planner_tools.py,sha256=tU1qz_VIQM_yPKDmuxjMWu68ZlAZ7ePWI1g7zswyWhI,13540
 vision_agent/tools/prompts.py,sha256=V1z4YJLXZuUl_iZ5rY0M5hHc_2tmMEUKr0WocXKGt4E,1430
 vision_agent/tools/tool_utils.py,sha256=LAnrb_nY6PNVamqJahRN-J0cuOy4gsKvCtSuXJf0RsI,10075
-vision_agent/tools/tools.py,sha256=n6-UPaZ4XjF29_7EF5GRgx74GjiZ7HqZn4a1Aw-e4P0,94059
+vision_agent/tools/tools.py,sha256=Xcm_9EQdDCR9X5FhIm7VJaTL0qWqhnJUVTRVrRtETrA,96112
 vision_agent/tools/tools_types.py,sha256=8hYf2OZhI58gvf65KGaeGkt4EQ56nwLFqIQDPHioOBc,2339
 vision_agent/utils/__init__.py,sha256=7fMgbZiEwbNS0fBOS_hJI5PuEYBblw36zLi_UjUzvj4,244
 vision_agent/utils/exceptions.py,sha256=booSPSuoULF7OXRr_YbC4dtKt6gM_HyiFQHBuaW86C4,2052
@@ -40,7 +40,7 @@ vision_agent/utils/image_utils.py,sha256=5uoYgXa6E0-lVrXR7K2XE7fe6r_n7pvK64HYQ50
 vision_agent/utils/sim.py,sha256=f1emBQM8SmyVKSrhj0NHItnfMHSeTw-Nk2pw-0eBZ5c,7462
 vision_agent/utils/type_defs.py,sha256=BE12s3JNQy36QvauXHjwyeffVh5enfcvd4vTzSwvEZI,1384
 vision_agent/utils/video.py,sha256=e1VwKhXzzlC5LcFMyrcQYrPnpnX4wxDpnQ-76sB4jgM,6001
-vision_agent-0.2.216.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-vision_agent-0.2.216.dist-info/METADATA,sha256=B88HzV_M0A12EmhiC-968LcdospsiOUUR-aTcZFTH8A,19071
-vision_agent-0.2.216.dist-info/WHEEL,sha256=7Z8_27uaHI_UZAc4Uox4PpBhQ9Y5_modZXWMxtUi4NU,88
-vision_agent-0.2.216.dist-info/RECORD,,
+vision_agent-0.2.217.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+vision_agent-0.2.217.dist-info/METADATA,sha256=xl9AmXP9RBpC5frlASsiG7YktdIOTRuJgv8WZdRV_bA,19071
+vision_agent-0.2.217.dist-info/WHEEL,sha256=7Z8_27uaHI_UZAc4Uox4PpBhQ9Y5_modZXWMxtUi4NU,88
+vision_agent-0.2.217.dist-info/RECORD,,