ag2 0.9.9__py3-none-any.whl → 0.10.0__py3-none-any.whl

autogen/agentchat/contrib/llava_agent.py

@@ -6,7 +6,6 @@
 # SPDX-License-Identifier: MIT
 import json
 import logging
-import warnings
 from typing import Any
 
 import requests
@@ -85,18 +84,7 @@ class LLaVAAgent(MultimodalConversableAgent):
         retry = 10
         while len(out) == 0 and retry > 0:
             # image names will be inferred automatically from llava_call
-            if "max_new_tokens" in self.llm_config:
-                warnings.warn(
-                    (
-                        "`max_new_tokens` is deprecated in `llm_config` for llava agents. "
-                        "Use `max_tokens` instead. "
-                        "Scheduled for removal in 0.10.0 version."
-                    ),
-                    DeprecationWarning,
-                )
-                max_tokens = self.llm_config["max_new_tokens"]
-            else:
-                max_tokens = self.llm_config.get("max_tokens")
+            max_tokens = self.llm_config.get("max_tokens")
 
             out = llava_call_binary(
                 prompt=prompt,

autogen/agentchat/contrib/rag/mongodb_query_engine.py

@@ -17,7 +17,7 @@ with optional_import_block():
     from llama_index.core import SimpleDirectoryReader, StorageContext, VectorStoreIndex
     from llama_index.core.embeddings import BaseEmbedding
     from llama_index.core.schema import Document as LlamaDocument
-    from llama_index.llms.langchain.base import LLM
+    from llama_index.llms.langchain.base import LLM  # type: ignore[attr-defined]
     from llama_index.llms.openai import OpenAI
     from llama_index.vector_stores.mongodb import MongoDBAtlasVectorSearch
     from pymongo import MongoClient
@@ -114,7 +114,7 @@ class MongoDBQueryEngine:
         logger.info("Vector database created.")
         self.vector_search_engine = MongoDBAtlasVectorSearch(
             mongodb_client=self.vector_db.client,  # type: ignore[union-attr]
-            db_name=self.database_name,
+            db_name=self.database_name,  # type: ignore[arg-type]
             collection_name=self.collection_name,
         )
         logger.info("Vector search engine created.")

autogen/agentchat/contrib/rag/query_engine.py

@@ -29,18 +29,18 @@ class RAGQueryEngine(Protocol):
         """Initialize the database with the input documents or records.
 
         This method initializes database with the input documents or records.
-        Usually, it takes the following steps:
-        1. connecting to a database.
-        2. insert records
-        3. build indexes etc.
+        Usually, it takes the following steps:\n
+        1. connecting to a database.\n
+        2. insert records.\n
+        3. build indexes etc.\n
 
-        Args:
-            new_doc_dir (Optional[Union[Path, str]]): A directory containing documents to be ingested.
-            new_doc_paths_or_urls (Optional[Sequence[Union[Path, str]]]): A list of paths or URLs to documents to be ingested.
-            *args: Any additional arguments
-            **kwargs: Any additional keyword arguments
-        Returns:
-            bool: True if initialization is successful, False otherwise
+        Args:\n
+            new_doc_dir (Optional[Union[Path, str]]): A directory containing documents to be ingested.\n
+            new_doc_paths_or_urls (Optional[Sequence[Union[Path, str]]]): A list of paths or URLs to documents to be ingested.\n
+            *args: Any additional arguments\n
+            **kwargs: Any additional keyword arguments\n
+        Returns:\n
+            bool: True if initialization is successful, False otherwise\n
         """
         ...
 

autogen/agentchat/contrib/retrieve_assistant_agent.py

@@ -7,10 +7,13 @@
 import warnings
 from typing import Any
 
+from typing_extensions import deprecated
+
 from ..agent import Agent
 from ..assistant_agent import AssistantAgent
 
 
+@deprecated("The RetrieveAssistantAgent is deprecated. Please use the AssistantAgent instead.")
 class RetrieveAssistantAgent(AssistantAgent):
     """(Experimental) Retrieve Assistant agent, designed to solve a task with LLM.
 

autogen/agentchat/contrib/swarm_agent.py

@@ -1163,8 +1163,9 @@ async def a_run_swarm(
             except Exception as e:
                 response.iostream.send(ErrorEvent(error=e))  # type: ignore[call-arg]
 
-    asyncio.create_task(stream_run())
-
+    task = asyncio.create_task(stream_run())
+    # prevent the task from being garbage collected
+    response._task_ref = task  # type: ignore[attr-defined]
     return response
 
 

autogen/agentchat/contrib/vectordb/couchbase.py

@@ -279,7 +279,7 @@ class CouchbaseVectorDB(VectorDB):
 
         for i in range(0, len(docs), batch_size):
             batch = docs[i : i + batch_size]
-            docs_to_upsert = dict()
+            docs_to_upsert = {}
             for doc in batch:
                 doc_id = doc["id"]
                 embedding = self.embedding_function([

autogen/agentchat/contrib/vectordb/mongodb.py

@@ -437,7 +437,7 @@ class MongoDBAtlasVectorDB(VectorDB):
         if include is None:
             include_fields = {"_id": 1, "content": 1, "metadata": 1}
         else:
-            include_fields = {k: 1 for k in set(include).union({"_id"})}
+            include_fields = dict.fromkeys(set(include).union({"_id"}), 1)
         collection = self.get_collection(collection_name)
         if ids is not None:
             docs = collection.find({"_id": {"$in": ids}}, include_fields)

autogen/agentchat/contrib/web_surfer.py

@@ -277,7 +277,7 @@ class WebSurferAgent(ConversableAgent):
         self._assistant.reset()  # type: ignore[no-untyped-call]
 
         # Clone the messages to give context
-        self._assistant.chat_messages[self._user_proxy] = list()
+        self._assistant.chat_messages[self._user_proxy] = []
         history = messages[0 : len(messages) - 1]
         for message in history:
             self._assistant.chat_messages[self._user_proxy].append(message)