vision-agent 0.2.96__tar.gz → 0.2.98__tar.gz

{vision_agent-0.2.96 → vision_agent-0.2.98}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vision-agent
-Version: 0.2.96
+Version: 0.2.98
 Summary: Toolset for Vision Agent
 Author: Landing AI
 Author-email: dev@landing.ai
@@ -11,7 +11,7 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Requires-Dist: anthropic (>=0.31.0,<0.32.0)
 Requires-Dist: e2b (>=0.17.1,<0.18.0)
-Requires-Dist: e2b-code-interpreter (==0.0.11a17)
+Requires-Dist: e2b-code-interpreter (==0.0.11a27)
 Requires-Dist: ipykernel (>=6.29.4,<7.0.0)
 Requires-Dist: langsmith (>=0.1.58,<0.2.0)
 Requires-Dist: moviepy (>=1.0.0,<2.0.0)

{vision_agent-0.2.96 → vision_agent-0.2.98}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "vision-agent"
-version = "0.2.96"
+version = "0.2.98"
 description = "Toolset for Vision Agent"
 authors = ["Landing AI <dev@landing.ai>"]
 readme = "README.md"
@@ -35,7 +35,7 @@ rich = "^13.7.1"
 langsmith = "^0.1.58"
 ipykernel = "^6.29.4"
 e2b = "^0.17.1"
-e2b-code-interpreter = "0.0.11a17"
+e2b-code-interpreter = "0.0.11a27"
 tenacity = "^8.3.0"
 pillow-heif = "^0.16.0"
 pytube = "15.0.0"

{vision_agent-0.2.96 → vision_agent-0.2.98}/vision_agent/agent/vision_agent_coder.py

@@ -93,7 +93,7 @@ def format_plans(plans: Dict[str, Any]) -> str:
 
 
 def extract_image(
-    media: Optional[Sequence[Union[str, Path]]]
+    media: Optional[Sequence[Union[str, Path]]],
 ) -> Optional[Sequence[Union[str, Path]]]:
     if media is None:
         return None
@@ -186,7 +186,8 @@ def pick_plan(
                 if tool_output.success
                 else "Code execution failed"
             ),
-            "payload": tool_output.to_json(),
+            "code": DefaultImports.prepend_imports(code),
+            # "payload": tool_output.to_json(),
             "status": "completed" if tool_output.success else "failed",
         }
     )
@@ -211,6 +212,9 @@ def pick_plan(
             }
         )
         code = extract_code(model(prompt))
+        tool_output = code_interpreter.exec_isolation(
+            DefaultImports.prepend_imports(code)
+        )
         log_progress(
             {
                 "type": "log",
@@ -220,13 +224,10 @@ def pick_plan(
                     else "Code execution failed"
                 ),
                 "code": DefaultImports.prepend_imports(code),
-                "payload": tool_output.to_json(),
+                # "payload": tool_output.to_json(),
                 "status": "completed" if tool_output.success else "failed",
             }
         )
-        tool_output = code_interpreter.exec_isolation(
-            DefaultImports.prepend_imports(code)
-        )
         tool_output_str = ""
         if len(tool_output.logs.stdout) > 0:
             tool_output_str = tool_output.logs.stdout[0]

{vision_agent-0.2.96 → vision_agent-0.2.98}/vision_agent/tools/tools.py

@@ -9,6 +9,7 @@ from typing import Any, Dict, List, Optional, Tuple, Union, cast
 import cv2
 import numpy as np
 import requests
+from moviepy.editor import ImageSequenceClip
 from PIL import Image, ImageDraw, ImageFont
 from pillow_heif import register_heif_opener  # type: ignore
 from pytube import YouTube  # type: ignore
@@ -1063,20 +1064,15 @@ def save_video(
         _LOGGER.warning(f"Invalid fps value: {fps}. Setting fps to 4 (default value).")
         fps = 4
 
-    if not output_video_path:
-        output_video_path = tempfile.NamedTemporaryFile(
-            suffix=".mp4", delete=False
-        ).name
-
-    height, width, layers = frames[0].shape if frames else (0, 0, 0)
-    fourcc = cv2.VideoWriter_fourcc(*"mp4v")  # type: ignore
-    video = cv2.VideoWriter(output_video_path, fourcc, fps, (width, height))
-    for frame in frames:
-        video.write(cv2.cvtColor(frame, cv2.COLOR_RGB2BGR))
-    video.release()
-
-    _save_video_to_result(output_video_path)
-    return output_video_path
+    with ImageSequenceClip(frames, fps=fps) as video:
+        if output_video_path:
+            f = open(output_video_path, "wb")
+        else:
+            f = tempfile.NamedTemporaryFile(suffix=".mp4", delete=False)  # type: ignore
+        video.write_videofile(f.name, codec="libx264")
+        f.close()
+        _save_video_to_result(f.name)
+        return f.name
 
 
 def _save_video_to_result(video_uri: str) -> None: