Upload 7 files

agentic/__init__.py

@@ -0,0 +1 @@
+from .langgraph_agent import LangGraphAgent4GAIA

agentic/langgraph_agent.py

@@ -0,0 +1,99 @@
+import os
+from dotenv import load_dotenv
+from langgraph.graph import START, StateGraph, MessagesState
+from langgraph.prebuilt import tools_condition, ToolNode
+from langchain_openai import ChatOpenAI
+from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
+from langchain_core.messages import SystemMessage, HumanMessage, AIMessage
+from omegaconf import OmegaConf
+from .tools import *
+
+
+def load_config(config_path: str):
+    config = OmegaConf.load(config_path)
+    return config
+
+# --- Constants ---
+CONFIG = load_config("config.yaml")
+SYSTEM_PROMPT = CONFIG["system_prompt"]["custom"]
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+class LangGraphAgent4GAIA:
+    def __init__(self, model_provider: str, model_name: str):
+        self.sys_prompt = SystemMessage(content=SYSTEM_PROMPT)
+        self.graph = self.get_agent(model_provider, model_name)
+
+    def assistant(self, state: MessagesState):
+        """Assistant node"""
+        return {"messages": [self.llm_with_tools.invoke([self.sys_prompt] + state["messages"])]}
+
+    def get_agent(self, provider: str, model_name: str):
+        tools = [
+            multiply,
+            add,
+            subtract,
+            divide,
+            modulo,
+            web_search,
+            arxiv_search,
+            wiki_search
+        ]
+
+        # 1. Build graph
+        if provider == "openai":
+            llm = ChatOpenAI(
+                model=model_name,
+                temperature=0,
+                max_retries=2,
+                api_key=os.getenv("OPENAI_API_KEY")
+            )
+        elif provider == "huggingface":
+            llm = ChatHuggingFace(
+                llm=HuggingFaceEndpoint(
+                    repo_id=model_name,
+                    task="text-generation",
+                    max_new_tokens=1024,
+                    do_sample=False,
+                    repetition_penalty=1.03,
+                    temperature=0
+                ),
+                verbose=True
+            )
+        else:
+            raise ValueError("Invalid provider. Choose 'openai' or 'huggingface'.")
+
+
+        # 2. Bind tools to LLM
+        self.llm_with_tools = llm.bind_tools(tools)
+
+        builder = StateGraph(MessagesState)
+        builder.add_node("assistant", self.assistant)
+        builder.add_node("tools", ToolNode(tools))
+        builder.add_edge(START, "assistant")
+        builder.add_conditional_edges(
+            "assistant",
+            tools_condition,
+        )
+        builder.add_edge("tools", "assistant")
+
+        # Compile graph
+        return builder.compile()
+
+if __name__ == "__main__":
+    from langchain_core.runnables.graph import MermaidDrawMethod
+
+    question = "What is the capital of Spain?"
+    # Build the graph
+    agent_manager = LangGraphAgent4GAIA(CONFIG["model"]["provider"], CONFIG["model"]["name"])
+    img_data = agent_manager.graph.get_graph().draw_mermaid_png(draw_method=MermaidDrawMethod.API)
+    with open('agentic/graph.png', "wb") as f:
+        f.write(img_data)
+
+    # Run the graph
+    messages = [HumanMessage(content=question)]
+    messages = agent_manager.graph.invoke({"messages": messages}, {"recursion_limit": 50})
+    for m in messages["messages"]:
+        m.pretty_print()

agentic/tools.py

@@ -0,0 +1,109 @@
+from langchain_core.tools import tool
+from langchain_community.tools import DuckDuckGoSearchResults
+from langchain_community.tools.tavily_search import TavilySearchResults
+from langchain_community.document_loaders import ArxivLoader, WikipediaLoader
+
+@tool
+def add(a: int, b: int) -> int:
+    """Add two numbers.
+
+    Args:
+        a: first int
+        b: second int
+    """
+    return a + b
+
+
+@tool
+def subtract(a: int, b: int) -> int:
+    """Subtract two numbers.
+
+    Args:
+        a: first int
+        b: second int
+    """
+    return a - b
+
+
+@tool
+def multiply(a: int, b: int) -> int:
+    """Multiply two numbers.
+
+    Args:
+        a: first int
+        b: second int
+    """
+    return a * b
+
+
+@tool
+def divide(a: int, b: int) -> int:
+    """Divide two numbers.
+
+    Args:
+        a: first int
+        b: second int
+    """
+    if b == 0:
+        raise ValueError("Cannot divide by zero.")
+    return a / b
+
+
+@tool
+def modulo(a: int, b: int) -> int:
+    """Get the modulus of two numbers.
+
+    Args:
+        a: first int
+        b: second int
+    """
+    return a % b
+
+
+@tool
+def web_search(query: str) -> str:
+    """Web search with Tavily for a query and return maximum 3 results.
+
+    Args:
+        query: The search query."""
+    # search_docs = DuckDuckGoSearchResults(max_results=3).invoke(query=query)
+    search_docs = TavilySearchResults(max_results=3).invoke(query=query)
+    formatted_search_docs = "\n\n---\n\n".join(
+        [
+            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
+            for doc in search_docs
+        ])
+    # return {"web_results": formatted_search_docs}
+    return formatted_search_docs
+
+
+@tool
+def arxiv_search(query: str) -> str:
+    """Search ArXiv for a query and return maximum 3 result.
+
+    Args:
+        query: The search query."""
+    search_docs = ArxivLoader(query=query, load_max_docs=3).load()
+    formatted_search_docs = "\n\n---\n\n".join(
+        [
+            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content[:1000]}\n</Document>'
+            for doc in search_docs
+        ])
+    # return {"arxiv_results": formatted_search_docs}
+    return formatted_search_docs
+
+
+@tool
+def wiki_search(query: str) -> str:
+    """Search Wikipedia for a query and return maximum 2 results.
+
+    Args:
+        query: The search query."""
+    search_docs = WikipediaLoader(query=query, load_max_docs=2).load()
+    formatted_search_docs = "\n\n---\n\n".join(
+        [
+            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
+            for doc in search_docs
+        ])
+    # return {"wiki_results": formatted_search_docs}
+    return formatted_search_docs

app.py

@@ -1,15 +1,38 @@
+"""
+This is a modified version of the original hf space code for submitting
+answers from the course.
+
+Instruction on submitting answers from
+https://huggingface.co/learn/agents-course/unit4/hands-on
+
+GET /questions: Retrieve the full list of filtered evaluation questions.
+GET /random-question: Fetch a single random question from the list.
+GET /files/{task_id}: Download a specific file associated with a given task ID.
+POST /submit: Submit agent answers, calculate the score, and update the leaderboard.
+
+The submit function will compare the answer to the ground truth in an EXACT MATCH manner,
+hence prompt it well ! The GAIA team shared a prompting example for your agent here
+(for the sake of this course, make sure you don't include the text "FINAL ANSWER" in your
+submission, just make your agent reply with the answer and nothing else).
+"""
 import os
 import gradio as gr
 import requests
-import inspect
 import pandas as pd
+from omegaconf import OmegaConf
+from langchain_core.messages import HumanMessage
+from agentic import LangGraphAgent4GAIA
+
+
+def load_config(config_path: str):
+    config = OmegaConf.load(config_path)
+    return config
 
-# (Keep Constants as is)
 # --- Constants ---
-DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+CONFIG = load_config("config.yaml")
 
 # --- Basic Agent Definition ---
-# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
+# ----- THIS IS WHERE YOU CAN BUILD WHAT YOU WANT ------
 class BasicAgent:
     def __init__(self):
         print("BasicAgent initialized.")
@@ -19,6 +42,23 @@ class BasicAgent:
         print(f"Agent returning fixed answer: {fixed_answer}")
         return fixed_answer
 
+class Agent:
+    def __init__(self):
+        self.agent_manager = LangGraphAgent4GAIA(
+            model_provider=CONFIG["model"]["provider"],
+            model_name=CONFIG["model"]["name"]
+        )
+        print("LangGraphAgent4GAIA initialized.")
+
+    def __call__(self, question: str) -> str:
+        print(f"Agent received question (first 50 chars): {question[:50]}...")
+        messages = [HumanMessage(content=question)]
+        result = self.agent_manager.graph.invoke({"messages": messages}, {"recursion_limit": 50})
+        answer = result['messages'][-1].content
+        final_answer = answer.split('FINAL ANSWER: ')[-1].strip()
+        return final_answer
+
+
 def run_and_submit_all( profile: gr.OAuthProfile | None):
     """
     Fetches all questions, runs the BasicAgent on them, submits all answers,
@@ -34,17 +74,17 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         print("User not logged in.")
         return "Please Login to Hugging Face with the button.", None
 
-    api_url = DEFAULT_API_URL
+    api_url = CONFIG["api_url"]
     questions_url = f"{api_url}/questions"
     submit_url = f"{api_url}/submit"
 
     # 1. Instantiate Agent ( modify this part to create your agent)
     try:
-        agent = BasicAgent()
+        agent = Agent()
     except Exception as e:
         print(f"Error instantiating agent: {e}")
         return f"Error initializing agent: {e}", None
-    # In the case of an app running as a hugging Face space, this link points toward your codebase ( usefull for others so please keep it public)
+    # In the case of an app running as a hugging Face space, this link points toward your codebase (useful for others so please keep it public)
     agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
     print(agent_code)
 

config.yaml

@@ -0,0 +1,22 @@
+api_url: "https://agents-course-unit4-scoring.hf.space"
+
+system_prompt:
+  default: "You are a general AI assistant. I will ask you a question. Report your thoughts, and finish your answer with the following template: FINAL ANSWER: [YOUR FINAL ANSWER]. YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string.\n"
+  custom: "You are a helpful assistant tasked with answering questions using a set of tools.\nNow, I will ask you a question. Report your thoughts, and finish your answer with the following template: \nFINAL ANSWER: [YOUR FINAL ANSWER]. \nYOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string.\nYour answer should only start with 'FINAL ANSWER: ', then follows with the answer. "
+
+model:
+  provider: "openai"
+  name: "gpt-4o-mini"
+#  provider: "huggingface"
+#  name: "Qwen/Qwen2.5-Coder-32B-Instruct"
+#  name: "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B"
+
+
+code:
+  username: "lautel"
+  hf_space: "https://huggingface.co/spaces/lautel/agents-course-final-project"
+
+paths:
+  output: "results"
+  output_filename: "results_{timestamp}.json"
+  logs_filename: "log_{timestamp}.txt"

requirements.txt

@@ -1,2 +1,20 @@
 gradio
-requests
+requests
+omegaconf
+pandas
+tabulate
+langchain
+langchain-community
+langchain-core
+langchain-huggingface
+langchain-openai
+langchain-tavily
+langgraph
+huggingface_hub
+qdrant-client
+arxiv
+pymupdf
+wikipedia
+pgvector
+python-dotenv
+beautifulsoup4