Create CustomAgent.py

CustomAgent.py

@@ -0,0 +1,137 @@
+import datasets
+from langchain.docstore.document import Document
+# Load the dataset
+# guest_dataset = datasets.load_dataset("agents-course/unit3-invitees", split="train")
+
+# Convert dataset entries into Document objects
+# docs = [
+#     Document(
+#         page_content="\n".join([
+#             f"Name: {guest['name']}",
+#             f"Relation: {guest['relation']}",
+#             f"Description: {guest['description']}",
+#             f"Email: {guest['email']}"
+#         ]),
+#         metadata={"name": guest["name"]}
+#     )
+#     for guest in guest_dataset
+# ]
+
+# from langchain_community.retrievers import BM25Retriever
+# from langchain.tools import Tool
+
+# bm25_retriever = BM25Retriever.from_documents(docs)
+
+# def extract_text(query: str) -> str:
+#     """Retrieves detailed information about gala guests based on their name or relation."""
+#     results = bm25_retriever.invoke(query)
+#     if results:
+#         return "\n\n".join([doc.page_content for doc in results[:3]])
+#     else:
+#         return "No matching guest information found."
+
+# guest_info_tool = Tool(
+#     name="guest_info_retriever",
+#     func=extract_text,
+#     description="Retrieves detailed information about gala guests based on their name or relation."
+# )
+#######################################################################################################################################################
+from typing import TypedDict, Annotated
+from langgraph.graph.message import add_messages
+from langchain_core.messages import AnyMessage, HumanMessage, AIMessage,SystemMessage
+from langgraph.prebuilt import ToolNode
+from langgraph.graph import START, StateGraph
+from langgraph.prebuilt import tools_condition
+from langchain_openai import ChatOpenAI
+from Webserch_tool import weather_info_tool
+from other_tools import (
+    wiki_search, arvix_search, web_search, vector_search, 
+    multiply, add, subtract, divide, modulus, power, square_root
+)
+import os
+from dotenv import load_dotenv
+load_dotenv()
+# Generate the chat interface, including the tools
+llm = ChatOpenAI(temperature=0
+                   , model="gpt-4o-mini", openai_api_key=os.getenv("OPENAI_KEY"))
+
+tools = [
+    weather_info_tool, wiki_search, arvix_search, web_search, 
+    multiply, add, subtract, divide, modulus, power, square_root
+]
+chat_with_tools = llm.bind_tools(tools)
+
+#setting up prompt
+ai_message = SystemMessage(content="""You are a helpful assistant tasked with answering questions using a set of tools and reference materials.
+
+You may be provided with a reference set of questions and answers from a retriever. 
+If the current question is identical to or semantically equivalent to a reference question, or if a reference answer clearly applies, use that reference answer directly.
+
+Otherwise, reason through the question as needed to determine the correct answer.
+
+Your output must follow these formatting rules:
+- If the answer is a number, do not use commas or units (unless specifically requested).
+- If the answer is a string, do not use articles, abbreviations, or short forms. Write digits in full unless specified otherwise.
+- If the answer is a comma-separated list, apply the above rules to each item and include exactly one space after each comma.
+- If the question matches a reference question, return the reference answer exactly as it appears.
+
+Do not include any explanation, prefix, or extra text—output only the final answer.
+""")
+
+
+
+
+
+
+# Generate the AgentState and Agent graph
+from langgraph.graph import MessagesState #the same as AgentState
+# class AgentState(TypedDict):
+#     messages: Annotated[list[AnyMessage], add_messages]
+
+def assistant(state: MessagesState):
+    return {
+        "messages": [chat_with_tools.invoke(state["messages"])],
+    }
+
+def retriever(state: MessagesState):
+    """Retriever node"""
+    similar_question = vector_search(state["messages"][0].content)
+
+    if similar_question:
+        example_msg = HumanMessage(
+            content=f"Here I provide a similar question and answer for reference: \n\n{similar_question}",
+        )
+        print(f"Similar question found: {similar_question}")
+        return {"messages": [ai_message] + state["messages"] + [example_msg]}
+    else:
+        # Handle the case when no similar questions are found
+        print( "No similar question found.")
+        return {"messages": [ai_message] + state["messages"]}
+
+## The graph
+builder = StateGraph(MessagesState)
+
+
+# Define nodes: these do the work
+builder.add_node("assistant", assistant)
+builder.add_node("retriever", retriever)
+builder.add_node("tools", ToolNode(tools))
+
+# Define edges: these determine how the control flow moves
+builder.add_edge(START, "retriever")
+builder.add_edge("retriever", "assistant")
+builder.add_conditional_edges(
+    "assistant",
+    # If the latest message requires a tool, route to tools
+    # Otherwise, provide a direct response
+    tools_condition,
+)
+builder.add_edge("tools", "assistant")
+alfred = builder.compile()
+
+# messages = [HumanMessage(content="When was a picture of St. Thomas Aquinas first added to the Wikipedia page on the Principle of double effect?")]
+# #messages = [HumanMessage(content="What the remainder of 30 divided by 7?")]
+# response = alfred.invoke({"messages": messages})
+
+
+# print(response['messages'][-1].content)