vision-agent 0.2.73__tar.gz → 0.2.75__tar.gz

{vision_agent-0.2.73 → vision_agent-0.2.75}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vision-agent
-Version: 0.2.73
+Version: 0.2.75
 Summary: Toolset for Vision Agent
 Author: Landing AI
 Author-email: dev@landing.ai
@@ -182,8 +182,8 @@ you. For example:
 
 ```python
 >>> import vision_agent as va
->>> llm = va.llm.OpenAILMM()
->>> detector = llm.generate_detector("Can you build a jar detector for me?")
+>>> lmm = va.lmm.OpenAILMM()
+>>> detector = lmm.generate_detector("Can you build a jar detector for me?")
 >>> detector(va.tools.load_image("jar.jpg"))
 [{"labels": ["jar",],
   "scores": [0.99],

{vision_agent-0.2.73 → vision_agent-0.2.75}/README.md

@@ -145,8 +145,8 @@ you. For example:
 
 ```python
 >>> import vision_agent as va
->>> llm = va.llm.OpenAILMM()
->>> detector = llm.generate_detector("Can you build a jar detector for me?")
+>>> lmm = va.lmm.OpenAILMM()
+>>> detector = lmm.generate_detector("Can you build a jar detector for me?")
 >>> detector(va.tools.load_image("jar.jpg"))
 [{"labels": ["jar",],
   "scores": [0.99],

{vision_agent-0.2.73 → vision_agent-0.2.75}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "vision-agent"
-version = "0.2.73"
+version = "0.2.75"
 description = "Toolset for Vision Agent"
 authors = ["Landing AI <dev@landing.ai>"]
 readme = "README.md"

{vision_agent-0.2.73 → vision_agent-0.2.75}/vision_agent/agent/vision_agent.py

@@ -437,7 +437,7 @@ def retrieve_tools(
         {
             "type": "tools",
             "status": "completed",
-            "payload": {v["description"]: v for v in tool_list}.values(),
+            "payload": list({v["description"]: v for v in tool_list}.values()),
         }
     )
 

{vision_agent-0.2.73 → vision_agent-0.2.75}/vision_agent/lmm/lmm.py

@@ -164,6 +164,7 @@ class OpenAILMM(LMM):
                 {"role": "system", "content": SYSTEM_PROMPT},
                 {"role": "user", "content": prompt},
             ],
+            response_format={"type": "json_object"},
         )
 
         try:
@@ -179,7 +180,7 @@ class OpenAILMM(LMM):
         return lambda x: T.clip(x, params["prompt"])
 
     def generate_detector(self, question: str) -> Callable:
-        api_doc = T.get_tool_documentation([T.grounding_dino])
+        api_doc = T.get_tool_documentation([T.owl_v2])
         prompt = CHOOSE_PARAMS.format(api_doc=api_doc, question=question)
         response = self.client.chat.completions.create(
             model=self.model_name,
@@ -187,6 +188,7 @@ class OpenAILMM(LMM):
                 {"role": "system", "content": SYSTEM_PROMPT},
                 {"role": "user", "content": prompt},
             ],
+            response_format={"type": "json_object"},
         )
 
         try:
@@ -199,7 +201,7 @@ class OpenAILMM(LMM):
             )
             raise ValueError("Failed to decode response")
 
-        return lambda x: T.grounding_dino(params["prompt"], x)
+        return lambda x: T.owl_v2(params["prompt"], x)
 
     def generate_segmentor(self, question: str) -> Callable:
         api_doc = T.get_tool_documentation([T.grounding_sam])
@@ -210,6 +212,7 @@ class OpenAILMM(LMM):
                 {"role": "system", "content": SYSTEM_PROMPT},
                 {"role": "user", "content": prompt},
             ],
+            response_format={"type": "json_object"},
         )
 
         try: