Update space

.gitignore

@@ -0,0 +1,57 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# Distribution / packaging
+dist/
+build/
+*.egg-info/
+
+# Virtual environments
+venv/
+env/
+.env
+.venv/
+ENV/
+
+# IDE specific files
+.idea/
+.vscode/
+*.swp
+*.swo
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# Testing
+.coverage
+htmlcov/
+.pytest_cache/
+.tox/
+
+# Logs
+*.log
+
+# Local development settings
+.env.local
+.env.development.local
+.env.test.local
+.env.production.local
+
+# Dependencies
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Documentation
+docs/_build/
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# chroma
+chroma_db/
+
+

app.py

@@ -1,64 +1,84 @@
+import os
+from typing import TypedDict, Annotated
+from dotenv import load_dotenv
+from langgraph.graph.message import add_messages
+from langchain_core.messages import AnyMessage, HumanMessage, SystemMessage
+from langgraph.prebuilt import ToolNode
+from langgraph.graph import START, StateGraph
+from langgraph.prebuilt import tools_condition
+from langgraph.checkpoint.memory import MemorySaver
+from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
+from langfuse.callback import CallbackHandler
+from retriever import guest_info_tool
+from tools import search_tool, weather_info_tool, hub_stats_tool
+
 import gradio as gr
-from huggingface_hub import InferenceClient
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+load_dotenv()
 
+# Initialize Langfuse
+langfuse_handler = CallbackHandler(
+    secret_key=os.getenv("LANGFUSE_SECRET_KEY"),
+    public_key=os.getenv("LANGFUSE_PUBLIC_KEY"),
+    host=os.getenv("LANGFUSE_HOST"),
+)
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Generate the chat interface, including the tools
+llm = HuggingFaceEndpoint(
+    repo_id="Qwen/Qwen2.5-Coder-32B-Instruct",
+    huggingfacehub_api_token=os.getenv("HUGGINGFACEHUB_API_TOKEN")
+)
+chat = ChatHuggingFace(llm=llm, verbose=True)
+tools = [guest_info_tool, search_tool, weather_info_tool, hub_stats_tool]
+chat_with_tools = chat.bind_tools(tools)
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
 
-    messages.append({"role": "user", "content": message})
+# Define the AgentState structure
+class AgentState(TypedDict):
+    messages: Annotated[list[AnyMessage], add_messages]
 
-    response = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+def assistant(state: AgentState):
+    # Define the system prompt (this is not part of the conversation history)
+    system_message = SystemMessage(content="""You are Alfred, a helpful and sophisticated assistant.
 
-        response += token
-        yield response
+Your capabilities:
+- Answer questions using your knowledge
+- Search the web for recent or factual information using the DuckDuckGoSearchResults tool
+- Retrieve information about guests using the guest_info_tool
+- Use weather_info_tool to give information about the weather
 
+Guidelines:
+- Be concise, polite and helpful
+- When you don't know something, use the appropriate tool rather than guessing
+- For guest information requests, always use the guest_info_tool first
+- For factual or current information, use the search tool
+- Present information in a clear, organized manner
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+Always think carefully about which tool is most appropriate for the user's request.
+""")
+    # Call the agent with the system prompt and conversation history (state messages)
+    assistant_response = chat_with_tools.invoke([system_message] + state.get("messages"))
+    # Return the updated conversation state including the new assistant response
+    return {"messages": state.get("messages") + [assistant_response]}
+
+# Build the graph
+builder = StateGraph(AgentState)
+builder.add_node("assistant", assistant)
+builder.add_node("tools", ToolNode(tools))
+builder.add_edge(START, "assistant")
+builder.add_conditional_edges("assistant", tools_condition)
+builder.add_edge("tools", "assistant")
+checkpointer = MemorySaver()
+alfred = builder.compile(checkpointer=checkpointer)
+config = {"configurable": {"thread_id": "1"}, "callbacks": [langfuse_handler]}
 
+# Gradio chat function
+def chat_fn(message, history):
+    state = {"messages": [HumanMessage(content=message)]}
+    result = alfred.invoke(state, config)
+    return result["messages"][-1].content
 
-if __name__ == "__main__":
-    demo.launch()
+# Launch Gradio interface
+interface = gr.ChatInterface(fn=chat_fn,type="messages")
+interface.launch()

requirements.txt

@@ -1 +1,15 @@
-huggingface_hub==0.25.2
+huggingface_hub>=0.28.1
+chromadb>=1.0.0
+datasets>=3.5.0
+duckduckgo-search>=2025.4.4
+gradio>=5.23.3
+langchain>=0.3.23
+langchain-community>=0.3.21
+langchain-huggingface>=0.1.2
+langfuse>=2.60.2
+langgraph>=0.3.25
+langsmith>=0.3.24
+pydantic>=2.11.2
+python-dotenv>=1.1.0
+rank-bm25>=0.2.2
+sentence-transformers>=4.0.2

retriever.py

@@ -0,0 +1,47 @@
+import datasets
+import os
+
+from langchain.docstore.document import Document
+from langchain_core.tools import Tool
+from langchain_community.vectorstores import Chroma
+from langchain_huggingface import HuggingFaceEmbeddings
+
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+# 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
+]
+
+# Create embedding model
+embeddings = HuggingFaceEmbeddings()
+
+# Create Chroma vector store
+vector_store = Chroma.from_documents(documents=docs, embedding=embeddings, persist_directory="./chroma_db")
+retriever = vector_store.as_retriever(search_kwargs={"k": 3})
+
+def extract_text(query: str) -> str:
+    """Retrieves detailed information about gala guests based on their name or relation."""
+    results = retriever.invoke(query)
+    if results:
+        return "\n\n".join([doc.page_content for doc in results])
+    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."
+)

tools.py

@@ -0,0 +1,85 @@
+import requests
+import os
+
+from huggingface_hub import list_models
+from langchain_community.tools import DuckDuckGoSearchResults
+from langchain.tools import Tool
+from dotenv import load_dotenv
+
+# Make sure environment variables are loaded
+load_dotenv()
+
+
+def get_weather_info(location: str) -> str:
+    """Fetches real weather information for a given location using WeatherAPI.com."""
+    api_key = os.getenv("WEATHERAPI_KEY")
+    if not api_key:
+        return "Error: WeatherAPI key not found. Please set the WEATHERAPI_KEY environment variable."
+
+    base_url = "http://api.weatherapi.com/v1/current.json"
+    params = {
+        "key": api_key,
+        "q": location,
+        "aqi": "no"  # You can change this to "yes" if you want air quality info
+    }
+
+    try:
+        response = requests.get(base_url, params=params)
+        data = response.json()
+
+        if response.status_code == 200:
+            # Extract relevant information from the response
+            location_data = data.get("location", {})
+            current_data = data.get("current", {})
+
+            location_name = location_data.get("name", "Unknown")
+            region = location_data.get("region", "Unknown")
+            country = location_data.get("country", "Unknown")
+            temp_c = current_data.get("temp_c", "N/A")
+            feelslike_c = current_data.get("feelslike_c", "N/A")
+            condition_text = current_data.get("condition", {}).get("text", "N/A")
+            humidity = current_data.get("humidity", "N/A")
+            wind_kph = current_data.get("wind_kph", "N/A")
+
+            return (
+                f"Weather in {location_name}, {region}, {country}: {condition_text}, "
+                f"{temp_c}°C (feels like {feelslike_c}°C). Humidity: {humidity}%, Wind: {wind_kph} kph"
+            )
+        else:
+            error_message = data.get("error", {}).get("message", "Unknown error")
+            return f"Error fetching weather data: {error_message}"
+
+    except Exception as e:
+        return f"An error occurred while fetching weather data: {str(e)}"
+
+
+# Initialize the tool
+weather_info_tool = Tool(
+    name="get_weather_info",
+    func=get_weather_info,
+    description="Fetches real weather information for a given location. Provide the city name and optionally the country code (e.g., 'London' or 'London,UK')."
+)
+
+
+def get_hub_stats(author: str) -> str:
+    """Fetches the most downloaded model from a specific author on the Hugging Face Hub."""
+    try:
+        # List models from the specified author, sorted by downloads
+        models = list(list_models(author=author, sort="downloads", direction=-1, limit=1))
+
+        if models:
+            model = models[0]
+            return f"The most downloaded model by {author} is {model.id} with {model.downloads:,} downloads."
+        else:
+            return f"No models found for author {author}."
+    except Exception as e:
+        return f"Error fetching models for {author}: {str(e)}"
+
+# Initialize the tool
+hub_stats_tool = Tool(
+    name="get_hub_stats",
+    func=get_hub_stats,
+    description="Fetches the most downloaded model from a specific author on the Hugging Face Hub."
+)
+
+search_tool = DuckDuckGoSearchResults()