vision-agent 0.2.48__py3-none-any.whl → 0.2.50__py3-none-any.whl

vision_agent/agent/vision_agent.py

@@ -36,11 +36,25 @@ logging.basicConfig(stream=sys.stdout)
 _LOGGER = logging.getLogger(__name__)
 _MAX_TABULATE_COL_WIDTH = 80
 _CONSOLE = Console()
-_DEFAULT_IMPORT = "\n".join(T.__new_tools__) + "\n".join(
-    [
+
+
+class DefaultImports:
+    """Container for default imports used in the code execution."""
+
+    common_imports = [
         "from typing import *",
     ]
-)
+
+    @staticmethod
+    def to_code_string() -> str:
+        return "\n".join(DefaultImports.common_imports + T.__new_tools__)
+
+    @staticmethod
+    def prepend_imports(code: str) -> str:
+        """Run this method to prepend the default imports to the code.
+        NOTE: be sure to run this method after the custom tools have been registered.
+        """
+        return DefaultImports.to_code_string() + "\n\n" + code
 
 
 def get_diff(before: str, after: str) -> str:
@@ -202,18 +216,20 @@ def write_and_test_code(
             "type": "code",
             "status": "running",
             "payload": {
-                "code": code,
+                "code": DefaultImports.prepend_imports(code),
                 "test": test,
             },
         }
     )
-    result = code_interpreter.exec_isolation(f"{_DEFAULT_IMPORT}\n{code}\n{test}")
+    result = code_interpreter.exec_isolation(
+        f"{DefaultImports.to_code_string()}\n{code}\n{test}"
+    )
     log_progress(
         {
             "type": "code",
             "status": "completed" if result.success else "failed",
             "payload": {
-                "code": code,
+                "code": DefaultImports.prepend_imports(code),
                 "test": test,
                 "result": result.to_json(),
             },
@@ -264,19 +280,21 @@ def write_and_test_code(
                 "type": "code",
                 "status": "running",
                 "payload": {
-                    "code": code,
+                    "code": DefaultImports.prepend_imports(code),
                     "test": test,
                 },
             }
         )
 
-        result = code_interpreter.exec_isolation(f"{_DEFAULT_IMPORT}\n{code}\n{test}")
+        result = code_interpreter.exec_isolation(
+            f"{DefaultImports.to_code_string()}\n{code}\n{test}"
+        )
         log_progress(
             {
                 "type": "code",
                 "status": "completed" if result.success else "failed",
                 "payload": {
-                    "code": code,
+                    "code": DefaultImports.prepend_imports(code),
                     "test": test,
                     "result": result.to_json(),
                 },
@@ -307,7 +325,14 @@ def write_and_test_code(
 def _print_code(title: str, code: str, test: Optional[str] = None) -> None:
     _CONSOLE.print(title, style=Style(bgcolor="dark_orange3", bold=True))
     _CONSOLE.print("=" * 30 + " Code " + "=" * 30)
-    _CONSOLE.print(Syntax(code, "python", theme="gruvbox-dark", line_numbers=True))
+    _CONSOLE.print(
+        Syntax(
+            DefaultImports.prepend_imports(code),
+            "python",
+            theme="gruvbox-dark",
+            line_numbers=True,
+        )
+    )
     if test:
         _CONSOLE.print("=" * 30 + " Test " + "=" * 30)
         _CONSOLE.print(Syntax(test, "python", theme="gruvbox-dark", line_numbers=True))
@@ -464,10 +489,6 @@ class VisionAgent(Agent):
                     if chat_i["role"] == "user":
                         chat_i["content"] += f" Image name {media}"
 
-            # re-grab custom tools
-            global _DEFAULT_IMPORT
-            _DEFAULT_IMPORT = "\n".join(T.__new_tools__)
-
             code = ""
             test = ""
             working_memory: List[Dict[str, str]] = []
@@ -531,38 +552,35 @@ class VisionAgent(Agent):
                 working_memory.extend(results["working_memory"])  # type: ignore
                 plan.append({"code": code, "test": test, "plan": plan_i})
 
-                if self_reflection:
-                    self.log_progress(
-                        {
-                            "type": "self_reflection",
-                            "status": "started",
-                        }
-                    )
-                    reflection = reflect(
-                        chat,
-                        FULL_TASK.format(
-                            user_request=chat[0]["content"], subtasks=plan_i_str
-                        ),
-                        code,
-                        self.planner,
-                    )
-                    if self.verbosity > 0:
-                        _LOGGER.info(f"Reflection: {reflection}")
-                    feedback = cast(str, reflection["feedback"])
-                    success = cast(bool, reflection["success"])
-                    self.log_progress(
-                        {
-                            "type": "self_reflection",
-                            "status": "completed" if success else "failed",
-                            "payload": reflection,
-                        }
-                    )
-                    working_memory.append(
-                        {"code": f"{code}\n{test}", "feedback": feedback}
-                    )
-                else:
+                if not self_reflection:
                     break
 
+                self.log_progress(
+                    {
+                        "type": "self_reflection",
+                        "status": "started",
+                    }
+                )
+                reflection = reflect(
+                    chat,
+                    FULL_TASK.format(
+                        user_request=chat[0]["content"], subtasks=plan_i_str
+                    ),
+                    code,
+                    self.planner,
+                )
+                if self.verbosity > 0:
+                    _LOGGER.info(f"Reflection: {reflection}")
+                feedback = cast(str, reflection["feedback"])
+                success = cast(bool, reflection["success"])
+                self.log_progress(
+                    {
+                        "type": "self_reflection",
+                        "status": "completed" if success else "failed",
+                        "payload": reflection,
+                    }
+                )
+                working_memory.append({"code": f"{code}\n{test}", "feedback": feedback})
                 retries += 1
 
             execution_result = cast(Execution, results["test_result"])
@@ -571,7 +589,7 @@ class VisionAgent(Agent):
                     "type": "final_code",
                     "status": "completed" if success else "failed",
                     "payload": {
-                        "code": code,
+                        "code": DefaultImports.prepend_imports(code),
                         "test": test,
                         "result": execution_result.to_json(),
                     },
@@ -586,7 +604,7 @@ class VisionAgent(Agent):
                         play_video(res.mp4)
 
             return {
-                "code": code,
+                "code": DefaultImports.prepend_imports(code),
                 "test": test,
                 "test_result": execution_result,
                 "plan": plan,

vision_agent/tools/tools.py

@@ -187,7 +187,7 @@ def extract_frames(
 
     Returns:
         List[Tuple[np.ndarray, float]]: A list of tuples containing the extracted frame
-        and the timestamp in seconds.
+        as a numpy array and the timestamp in seconds.
 
     Example
     -------
@@ -515,7 +515,7 @@ def save_json(data: Any, file_path: str) -> None:
 
 
 def load_image(image_path: str) -> np.ndarray:
-    """'load_image' is a utility function that loads an image from the given path.
+    """'load_image' is a utility function that loads an image from the given file path string.
 
     Parameters:
         image_path (str): The path to the image.
@@ -527,7 +527,9 @@ def load_image(image_path: str) -> np.ndarray:
     -------
     >>> load_image("path/to/image.jpg")
     """
-
+    # NOTE: sometimes the generated code pass in a NumPy array
+    if isinstance(image_path, np.ndarray):
+        return image_path
     image = Image.open(image_path).convert("RGB")
     return np.array(image)
 
@@ -647,12 +649,9 @@ def overlay_bounding_boxes(
         box = elt["bbox"]
         scores = elt["score"]
 
-        box = [
-            int(box[0] * width),
-            int(box[1] * height),
-            int(box[2] * width),
-            int(box[3] * height),
-        ]
+        # denormalize the box if it is normalized
+        box = denormalize_bbox(box, (height, width))
+
         draw.rectangle(box, outline=color[label], width=4)
         text = f"{label}: {scores:.2f}"
         text_box = draw.textbbox((box[0], box[1]), text=text, font=font)

{vision_agent-0.2.48.dist-info → vision_agent-0.2.50.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vision-agent
-Version: 0.2.48
+Version: 0.2.50
 Summary: Toolset for Vision Agent
 Author: Landing AI
 Author-email: dev@landing.ai

{vision_agent-0.2.48.dist-info → vision_agent-0.2.50.dist-info}/RECORD

@@ -11,7 +11,7 @@ vision_agent/agent/easytool_v2.py,sha256=CjY-sSj3abxnSq3ZHZMt-7YvRWDXEZsC6RN8FFI
 vision_agent/agent/easytool_v2_prompts.py,sha256=MZSIwovYgB-f-kdJ6btaNDVXptJn47bfOL3-Zn6NiC0,8573
 vision_agent/agent/reflexion.py,sha256=AlM5AvBJvCslXlYQdZiadq4oVHsNBm3IF_03DglTxRo,10506
 vision_agent/agent/reflexion_prompts.py,sha256=G7UAeNz_g2qCb2yN6OaIC7bQVUkda4m3z42EG8wAyfE,9342
-vision_agent/agent/vision_agent.py,sha256=X_LF2wRXVYAr8xMuJs3Omi8n06uVgLNgtF25sidKtfM,20424
+vision_agent/agent/vision_agent.py,sha256=0EqpLyyzpRGmT7fhS2XvLeUlktgCXTE5k1KGMQ8z3_s,20963
 vision_agent/agent/vision_agent_prompts.py,sha256=hgnTlaYp2HMBHLi3e4faPb-DI5jQL9jfhKq9jyEUEgY,8370
 vision_agent/fonts/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 vision_agent/fonts/default_font_ch_en.ttf,sha256=1YM0Z3XqLDjSNbF7ihQFSAIUdjF9m1rtHiNC_6QosTE,1594400
@@ -23,14 +23,14 @@ vision_agent/tools/__init__.py,sha256=Sng6dChynJJCYWjraXXM0tep_VPdnYl3L9vb0HMy_P
 vision_agent/tools/easytool_tools.py,sha256=pZc5dQlYINlV4nYbbzsDi3-wauA-fCeD2iGmJUMoUfE,47373
 vision_agent/tools/prompts.py,sha256=V1z4YJLXZuUl_iZ5rY0M5hHc_2tmMEUKr0WocXKGt4E,1430
 vision_agent/tools/tool_utils.py,sha256=wzRacbUpqk9hhfX_Y08rL8qP0XCN2w-8IZoYLi3Upn4,869
-vision_agent/tools/tools.py,sha256=Vpn2SxtjEcnztovat6qMiH52gFsDHo3ikEPrAT4e5yc,26639
+vision_agent/tools/tools.py,sha256=Sc6tAYbH03TbrPKAT8XIj1YZIwhd9j2k4ia8iKHhxzM,26743
 vision_agent/utils/__init__.py,sha256=Ce4yPhoWanRsnTy3X7YzZNBYYRJsrJeT7N59WUf8GZM,209
 vision_agent/utils/execute.py,sha256=GqoAodxtwTPBr1nujPTsWiZO2rBGvWVXTe8lgxY4d_g,20603
 vision_agent/utils/image_utils.py,sha256=_cdiS5YrLzqkq_ZgFUO897m5M4_SCIThwUy4lOklfB8,7700
 vision_agent/utils/sim.py,sha256=oUZ-6eu8Io-UNt9GXJ0XRKtP-Wc0sPWVzYGVpB2yDFk,3001
 vision_agent/utils/type_defs.py,sha256=BlI8ywWHAplC7kYWLvt4AOdnKpEW3qWEFm-GEOSkrFQ,1792
 vision_agent/utils/video.py,sha256=BJ9fomy2giAl038JThQP1WQZ-u4J4J_nsZB7QEWvlcQ,8767
-vision_agent-0.2.48.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-vision_agent-0.2.48.dist-info/METADATA,sha256=sJSWNAHN2-JMNb5hi4iA-HTzKNskLioIse9sdrMDuy4,6817
-vision_agent-0.2.48.dist-info/WHEEL,sha256=7Z8_27uaHI_UZAc4Uox4PpBhQ9Y5_modZXWMxtUi4NU,88
-vision_agent-0.2.48.dist-info/RECORD,,
+vision_agent-0.2.50.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+vision_agent-0.2.50.dist-info/METADATA,sha256=nLyeSFYnn4Bv_RyKzrP5iqnCRRkwCZT_d3euN1zgBOA,6817
+vision_agent-0.2.50.dist-info/WHEEL,sha256=7Z8_27uaHI_UZAc4Uox4PpBhQ9Y5_modZXWMxtUi4NU,88
+vision_agent-0.2.50.dist-info/RECORD,,