vision-agent 0.2.205__py3-none-any.whl → 0.2.207__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- vision_agent/agent/vision_agent_coder_v2.py +2 -2
 - vision_agent/tools/__init__.py +1 -0
 - vision_agent/tools/planner_tools.py +5 -5
 - vision_agent/tools/tools.py +7 -0
 - {vision_agent-0.2.205.dist-info → vision_agent-0.2.207.dist-info}/METADATA +1 -1
 - {vision_agent-0.2.205.dist-info → vision_agent-0.2.207.dist-info}/RECORD +8 -8
 - {vision_agent-0.2.205.dist-info → vision_agent-0.2.207.dist-info}/LICENSE +0 -0
 - {vision_agent-0.2.205.dist-info → vision_agent-0.2.207.dist-info}/WHEEL +0 -0
 
| 
         @@ -34,7 +34,7 @@ from vision_agent.utils.execute import ( 
     | 
|
| 
       34 
34 
     | 
    
         
             
                CodeInterpreterFactory,
         
     | 
| 
       35 
35 
     | 
    
         
             
                Execution,
         
     | 
| 
       36 
36 
     | 
    
         
             
            )
         
     | 
| 
       37 
     | 
    
         
            -
            from vision_agent.utils.sim import Sim 
     | 
| 
      
 37 
     | 
    
         
            +
            from vision_agent.utils.sim import Sim
         
     | 
| 
       38 
38 
     | 
    
         | 
| 
       39 
39 
     | 
    
         
             
            _CONSOLE = Console()
         
     | 
| 
       40 
40 
     | 
    
         | 
| 
         @@ -316,7 +316,7 @@ class VisionAgentCoderV2(AgentCoder): 
     | 
|
| 
       316 
316 
     | 
    
         
             
                        elif isinstance(tool_recommender, Sim):
         
     | 
| 
       317 
317 
     | 
    
         
             
                            self.tool_recommender = tool_recommender
         
     | 
| 
       318 
318 
     | 
    
         
             
                    else:
         
     | 
| 
       319 
     | 
    
         
            -
                        self.tool_recommender =  
     | 
| 
      
 319 
     | 
    
         
            +
                        self.tool_recommender = T.get_tool_recommender()
         
     | 
| 
       320 
320 
     | 
    
         | 
| 
       321 
321 
     | 
    
         
             
                    self.verbose = verbose
         
     | 
| 
       322 
322 
     | 
    
         
             
                    self.code_sandbox_runtime = code_sandbox_runtime
         
     | 
    
        vision_agent/tools/__init__.py
    CHANGED
    
    
| 
         @@ -32,10 +32,8 @@ from vision_agent.utils.execute import ( 
     | 
|
| 
       32 
32 
     | 
    
         
             
                MimeType,
         
     | 
| 
       33 
33 
     | 
    
         
             
            )
         
     | 
| 
       34 
34 
     | 
    
         
             
            from vision_agent.utils.image_utils import convert_to_b64
         
     | 
| 
       35 
     | 
    
         
            -
            from vision_agent.utils.sim import load_cached_sim
         
     | 
| 
       36 
35 
     | 
    
         | 
| 
       37 
36 
     | 
    
         
             
            TOOL_FUNCTIONS = {tool.__name__: tool for tool in T.TOOLS}
         
     | 
| 
       38 
     | 
    
         
            -
            TOOL_RECOMMENDER = load_cached_sim(T.TOOLS_DF)
         
     | 
| 
       39 
37 
     | 
    
         | 
| 
       40 
38 
     | 
    
         
             
            _LOGGER = logging.getLogger(__name__)
         
     | 
| 
       41 
39 
     | 
    
         
             
            EXAMPLES = f"\n{TEST_TOOLS_EXAMPLE1}\n{TEST_TOOLS_EXAMPLE2}\n"
         
     | 
| 
         @@ -52,7 +50,7 @@ def format_tool_output(tool_thoughts: str, tool_docstring: str) -> str: 
     | 
|
| 
       52 
50 
     | 
    
         | 
| 
       53 
51 
     | 
    
         | 
| 
       54 
52 
     | 
    
         
             
            def extract_tool_info(
         
     | 
| 
       55 
     | 
    
         
            -
                tool_choice_context: Dict[str, Any]
         
     | 
| 
      
 53 
     | 
    
         
            +
                tool_choice_context: Dict[str, Any],
         
     | 
| 
       56 
54 
     | 
    
         
             
            ) -> Tuple[Optional[Callable], str, str, str]:
         
     | 
| 
       57 
55 
     | 
    
         
             
                tool_thoughts = tool_choice_context.get("thoughts", "")
         
     | 
| 
       58 
56 
     | 
    
         
             
                tool_docstring = ""
         
     | 
| 
         @@ -124,7 +122,7 @@ def run_tool_testing( 
     | 
|
| 
       124 
122 
     | 
    
         
             
                        f"I need models from the {category.strip()} category of tools. {task}"
         
     | 
| 
       125 
123 
     | 
    
         
             
                    )
         
     | 
| 
       126 
124 
     | 
    
         | 
| 
       127 
     | 
    
         
            -
                tool_docs =  
     | 
| 
      
 125 
     | 
    
         
            +
                tool_docs = T.get_tool_recommender().top_k(category, k=10, thresh=0.2)
         
     | 
| 
       128 
126 
     | 
    
         
             
                if exclude_tools is not None and len(exclude_tools) > 0:
         
     | 
| 
       129 
127 
     | 
    
         
             
                    cleaned_tool_docs = []
         
     | 
| 
       130 
128 
     | 
    
         
             
                    for tool_doc in tool_docs:
         
     | 
| 
         @@ -246,7 +244,9 @@ def get_tool_for_task( 
     | 
|
| 
       246 
244 
     | 
    
         
             
                            context=f"<code>\n{code}\n</code>\n<tool_output>\n{tool_output_str}\n</tool_output>",
         
     | 
| 
       247 
245 
     | 
    
         
             
                            previous_attempts=error_message,
         
     | 
| 
       248 
246 
     | 
    
         
             
                        )
         
     | 
| 
       249 
     | 
    
         
            -
                        tool_choice_context_dict = extract_json( 
     | 
| 
      
 247 
     | 
    
         
            +
                        tool_choice_context_dict = extract_json(
         
     | 
| 
      
 248 
     | 
    
         
            +
                            lmm.generate(prompt, media=image_paths)  # type: ignore
         
     | 
| 
      
 249 
     | 
    
         
            +
                        )
         
     | 
| 
       250 
250 
     | 
    
         
             
                        tool, tool_thoughts, tool_docstring, error_message = extract_tool_info(
         
     | 
| 
       251 
251 
     | 
    
         
             
                            tool_choice_context_dict
         
     | 
| 
       252 
252 
     | 
    
         
             
                        )
         
     | 
    
        vision_agent/tools/tools.py
    CHANGED
    
    | 
         @@ -4,6 +4,7 @@ import logging 
     | 
|
| 
       4 
4 
     | 
    
         
             
            import os
         
     | 
| 
       5 
5 
     | 
    
         
             
            import tempfile
         
     | 
| 
       6 
6 
     | 
    
         
             
            import urllib.request
         
     | 
| 
      
 7 
     | 
    
         
            +
            from functools import lru_cache
         
     | 
| 
       7 
8 
     | 
    
         
             
            from importlib import resources
         
     | 
| 
       8 
9 
     | 
    
         
             
            from pathlib import Path
         
     | 
| 
       9 
10 
     | 
    
         
             
            from typing import Any, Dict, List, Optional, Tuple, Union, cast
         
     | 
| 
         @@ -45,6 +46,7 @@ from vision_agent.utils.image_utils import ( 
     | 
|
| 
       45 
46 
     | 
    
         
             
                rle_decode,
         
     | 
| 
       46 
47 
     | 
    
         
             
                rle_decode_array,
         
     | 
| 
       47 
48 
     | 
    
         
             
            )
         
     | 
| 
      
 49 
     | 
    
         
            +
            from vision_agent.utils.sim import Sim, load_cached_sim
         
     | 
| 
       48 
50 
     | 
    
         
             
            from vision_agent.utils.video import (
         
     | 
| 
       49 
51 
     | 
    
         
             
                extract_frames_from_video,
         
     | 
| 
       50 
52 
     | 
    
         
             
                frames_to_bytes,
         
     | 
| 
         @@ -80,6 +82,11 @@ _OCR_URL = "https://app.landing.ai/ocr/v1/detect-text" 
     | 
|
| 
       80 
82 
     | 
    
         
             
            _LOGGER = logging.getLogger(__name__)
         
     | 
| 
       81 
83 
     | 
    
         | 
| 
       82 
84 
     | 
    
         | 
| 
      
 85 
     | 
    
         
            +
            @lru_cache(maxsize=1)
         
     | 
| 
      
 86 
     | 
    
         
            +
            def get_tool_recommender() -> Sim:
         
     | 
| 
      
 87 
     | 
    
         
            +
                return load_cached_sim(TOOLS_DF)
         
     | 
| 
      
 88 
     | 
    
         
            +
             
     | 
| 
      
 89 
     | 
    
         
            +
             
     | 
| 
       83 
90 
     | 
    
         
             
            def grounding_dino(
         
     | 
| 
       84 
91 
     | 
    
         
             
                prompt: str,
         
     | 
| 
       85 
92 
     | 
    
         
             
                image: np.ndarray,
         
     | 
| 
         @@ -10,7 +10,7 @@ vision_agent/agent/vision_agent.py,sha256=I75bEU-os9Lf9OSICKfvQ_H_ftg-zOwgTwWnu4 
     | 
|
| 
       10 
10 
     | 
    
         
             
            vision_agent/agent/vision_agent_coder.py,sha256=ANwUuCO4JpTYJs4s6ynSRFcdjZFUVuSoSfcqp8ZQDDQ,27451
         
     | 
| 
       11 
11 
     | 
    
         
             
            vision_agent/agent/vision_agent_coder_prompts.py,sha256=gPLVXQMNSzYnQYpNm0wlH_5FPkOTaFDV24bqzK3jQ40,12221
         
     | 
| 
       12 
12 
     | 
    
         
             
            vision_agent/agent/vision_agent_coder_prompts_v2.py,sha256=9v5HwbNidSzYUEFl6ZMniWWOmyLITM_moWLtKVaTen8,4845
         
     | 
| 
       13 
     | 
    
         
            -
            vision_agent/agent/vision_agent_coder_v2.py,sha256= 
     | 
| 
      
 13 
     | 
    
         
            +
            vision_agent/agent/vision_agent_coder_v2.py,sha256=WKYPJAliupxnF2TP5jZlinqxnID37xnYSDNGMwoFKwU,16092
         
     | 
| 
       14 
14 
     | 
    
         
             
            vision_agent/agent/vision_agent_planner.py,sha256=KWMA7XemcSmc_jn-MwdWz9wnKDtj-sYQ9tINi70_OoU,18583
         
     | 
| 
       15 
15 
     | 
    
         
             
            vision_agent/agent/vision_agent_planner_prompts.py,sha256=Y3jz9HRf8fz9NLUseN7cTgZqewP0RazxR7vw1sPhcn0,6691
         
     | 
| 
       16 
16 
     | 
    
         
             
            vision_agent/agent/vision_agent_planner_prompts_v2.py,sha256=Tzon3h5iZdHJglesk8GVS-2myNf5-fhf7HUbkpZWHQk,33143
         
     | 
| 
         @@ -26,12 +26,12 @@ vision_agent/fonts/default_font_ch_en.ttf,sha256=1YM0Z3XqLDjSNbF7ihQFSAIUdjF9m1r 
     | 
|
| 
       26 
26 
     | 
    
         
             
            vision_agent/lmm/__init__.py,sha256=jyY1sJb_tYKg5-Wzs3p1lvwFkc-aUNZfMcLy3TOC4Zg,100
         
     | 
| 
       27 
27 
     | 
    
         
             
            vision_agent/lmm/lmm.py,sha256=x_nIyDNDZwq4-pfjnJTmcyyJZ2_B7TjkA5jZp88YVO8,17103
         
     | 
| 
       28 
28 
     | 
    
         
             
            vision_agent/lmm/types.py,sha256=ZEXR_ptBL0ZwDMTDYkgxUCmSZFmBYPQd2jreNzr_8UY,221
         
     | 
| 
       29 
     | 
    
         
            -
            vision_agent/tools/__init__.py,sha256= 
     | 
| 
      
 29 
     | 
    
         
            +
            vision_agent/tools/__init__.py,sha256=R6Ua9j1ragY-omt1OLivCwwFY_x6-nn-Ic9vTTvzUss,2904
         
     | 
| 
       30 
30 
     | 
    
         
             
            vision_agent/tools/meta_tools.py,sha256=TPeS7QWnc_PmmU_ndiDT03dXbQ5yDSP33E7U8cSj7Ls,28660
         
     | 
| 
       31 
     | 
    
         
            -
            vision_agent/tools/planner_tools.py,sha256= 
     | 
| 
      
 31 
     | 
    
         
            +
            vision_agent/tools/planner_tools.py,sha256=gzCCdruzAmVXSgEVRjFdc9qLhe8rlJ-O4Mi7NnI4LsA,13364
         
     | 
| 
       32 
32 
     | 
    
         
             
            vision_agent/tools/prompts.py,sha256=V1z4YJLXZuUl_iZ5rY0M5hHc_2tmMEUKr0WocXKGt4E,1430
         
     | 
| 
       33 
33 
     | 
    
         
             
            vision_agent/tools/tool_utils.py,sha256=AT7rMcpKwZgIErfgfSvHS0gmtvd8KMHJoHnu5aMlgO0,10259
         
     | 
| 
       34 
     | 
    
         
            -
            vision_agent/tools/tools.py,sha256= 
     | 
| 
      
 34 
     | 
    
         
            +
            vision_agent/tools/tools.py,sha256=bq5a7ZVgPDz7sxnwJ0VoZtXAUo7ACChCBXUjFE3iETI,87760
         
     | 
| 
       35 
35 
     | 
    
         
             
            vision_agent/tools/tools_types.py,sha256=8hYf2OZhI58gvf65KGaeGkt4EQ56nwLFqIQDPHioOBc,2339
         
     | 
| 
       36 
36 
     | 
    
         
             
            vision_agent/utils/__init__.py,sha256=7fMgbZiEwbNS0fBOS_hJI5PuEYBblw36zLi_UjUzvj4,244
         
     | 
| 
       37 
37 
     | 
    
         
             
            vision_agent/utils/exceptions.py,sha256=booSPSuoULF7OXRr_YbC4dtKt6gM_HyiFQHBuaW86C4,2052
         
     | 
| 
         @@ -40,7 +40,7 @@ vision_agent/utils/image_utils.py,sha256=rRWcxKggPXIRXIY_XT9rZt30ECDRq8zq7FDeXRD 
     | 
|
| 
       40 
40 
     | 
    
         
             
            vision_agent/utils/sim.py,sha256=NZc9QGD6BTY5O29NVbHH7oxDePL_QMnylT1lYcDUn1Y,7437
         
     | 
| 
       41 
41 
     | 
    
         
             
            vision_agent/utils/type_defs.py,sha256=BE12s3JNQy36QvauXHjwyeffVh5enfcvd4vTzSwvEZI,1384
         
     | 
| 
       42 
42 
     | 
    
         
             
            vision_agent/utils/video.py,sha256=tRcGp4vEnaDycigL1hBO9k0FBPtDH35fCQciVr9GqYI,6013
         
     | 
| 
       43 
     | 
    
         
            -
            vision_agent-0.2. 
     | 
| 
       44 
     | 
    
         
            -
            vision_agent-0.2. 
     | 
| 
       45 
     | 
    
         
            -
            vision_agent-0.2. 
     | 
| 
       46 
     | 
    
         
            -
            vision_agent-0.2. 
     | 
| 
      
 43 
     | 
    
         
            +
            vision_agent-0.2.207.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
         
     | 
| 
      
 44 
     | 
    
         
            +
            vision_agent-0.2.207.dist-info/METADATA,sha256=jkbYpz8kjqd2ijaSzduHZBDs0CU67a7kZ2gPF4Oq72w,19026
         
     | 
| 
      
 45 
     | 
    
         
            +
            vision_agent-0.2.207.dist-info/WHEEL,sha256=7Z8_27uaHI_UZAc4Uox4PpBhQ9Y5_modZXWMxtUi4NU,88
         
     | 
| 
      
 46 
     | 
    
         
            +
            vision_agent-0.2.207.dist-info/RECORD,,
         
     | 
| 
         
            File without changes
         
     | 
| 
         
            File without changes
         
     |