Update app.py

app.py

@@ -1,155 +1,144 @@
+
+# app.py
 import os
+import gradio as gr
 import requests
-from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI
-from llama_index.embeddings.huggingface import HuggingFaceEmbedding
-from langchain_community.document_loaders import WikipediaLoader
-from llama_index.core.tools.types import ToolMetadata
-from llama_index.core.schema import Document
-from llama_index.core.tools import FunctionTool
-from langchain_community.tools.tavily_search import TavilySearchResults
-from llama_index.core.agent.workflow import AgentWorkflow
-
-hf_token = os.getenv("HF_TOKEN")
-
-# List of models to try in order
-model_list = [
-    "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
-    "microsoft/phi-3-mini-128k-instruct",
-    "google/gemma-2b-it",
-    "gpt2"
-]
-
-current_model_index = 0
-llm = HuggingFaceInferenceAPI(
-    model_name=model_list[current_model_index],
-    token=hf_token,
-)
-
-# Numerical operation functions
-def multiply(a: int, b: int) -> int:
-    """Multiply two numbers."""
-    return a * b
-
-def add(a: int, b: int) -> int:
-    """Add two numbers."""
-    return a + b
-
-def subtract(a: int, b: int) -> int:
-    """Subtract two numbers."""
-    return a - b
-
-def divide(a: int, b: int) -> float:
-    """Divide two numbers, raises error on zero divisor."""
-    if b == 0:
-        raise ValueError("Cannot divide by zero.")
-    return a / b
-
-def modulus(a: int, b: int) -> int:
-    """Get the modulus of two numbers."""
-    return a % b
-
-# Web search tool function
-def web_search(query: str) -> list:
-    """Search Tavily for a query and return up to 3 results."""
-    results = TavilySearchResults(max_results=3).invoke(query=query)
-    docs = []
-    for r in results:
-        meta = {"source": r.metadata.get("source", ""), "page": r.metadata.get("page", "")}
-        docs.append(Document(text=r.page_content, metadata=meta))
-    return docs
-
-# Wikipedia search tool function
-def wiki_search(query: str) -> list:
-    """Search Wikipedia for a query and return up to 2 results."""
-    results = WikipediaLoader(query=query, load_max_docs=2).load()
-    docs = []
-    for r in results:
-        meta = {"source": r.metadata.get("source", ""), "page": r.metadata.get("page", "")}
-        docs.append(Document(text=r.page_content, metadata=meta))
-    return docs
-
-# Wrap functions into FunctionTool instances
-web_search_tool = FunctionTool(
-    web_search,
-    metadata=ToolMetadata(name="web_search", description="Tavily 3-hit search")
-)
-wiki_search_tool = FunctionTool(
-    wiki_search,
-    metadata=ToolMetadata(name="wiki_search", description="Wikipedia 2-hit search")
-)
-
-multiply_tool = FunctionTool(multiply, metadata=ToolMetadata(name="multiply", description="Multiply two numbers."))
-add_tool      = FunctionTool(add,      metadata=ToolMetadata(name="add",      description="Add two numbers."))
-subtract_tool = FunctionTool(subtract, metadata=ToolMetadata(name="subtract", description="Subtract two numbers."))
-divide_tool   = FunctionTool(divide,   metadata=ToolMetadata(name="divide",   description="Divide two numbers."))
-modulus_tool  = FunctionTool(modulus,  metadata=ToolMetadata(name="modulus",  description="Modulus operation on two numbers."))
-
-# Aggregate all tools
-tools = [
-    web_search_tool,
-    wiki_search_tool,
-    multiply_tool,
-    add_tool,
-    subtract_tool,
-    divide_tool,
-    modulus_tool,
-]
-
-# Initialize agent
-agent = AgentWorkflow.from_tools_or_functions(tools, llm=llm)
-
-# Function to try the next model in the list
-def try_next_model():
-    """Switch to the next model in the list and reinitialize the agent.
-    Returns True if successful, False if we've tried all models."""
-    global current_model_index, llm, agent
-    
-    current_model_index += 1
-    if current_model_index >= len(model_list):
-        return False
-    
-    # Reinitialize LLM with new model
-    llm = HuggingFaceInferenceAPI(
-        model_name=model_list[current_model_index],
-        token=hf_token,
-    )
+import pandas as pd
+from agent import agent, run_with_fallback  # Import run_with_fallback directly
+import asyncio
+import nest_asyncio
+nest_asyncio.apply()
+
+# Constants
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+
+# Async helper to run the agent - modified to use run_with_fallback
+async def run_agent(agent, question_text):
+    """Run the agent in a way that's compatible with asyncio"""
+    # Create a new event loop for this function call to avoid nesting issues
+    loop = asyncio.get_event_loop()
+    # Run the synchronous function in the executor
+    return await loop.run_in_executor(None, run_with_fallback, question_text)
+
+# Gradio Agent Interface
+def run_and_submit_all(profile: gr.OAuthProfile | None):
+    """
+    Fetches all questions, runs the LlamaIndexAgent on them, submits all answers,
+    and displays the results.
+    """
+    # --- Determine HF Space Runtime URL and Repo URL ---
+    space_id = os.getenv("SPACE_ID")  # Get the SPACE_ID for sending link to the code
+
+    if profile:
+        username = f"{profile.username}"
+        print(f"User logged in: {username}")
+    else:
+        print("User not logged in.")
+        return "Please Login to Hugging Face with the button.", None
+
+    api_url = DEFAULT_API_URL
+    questions_url = f"{api_url}/questions"
+    submit_url = f"{api_url}/submit"
+
+    # 1. Instantiate LlamaIndexAgent
+    print("Using imported agent instance.")
+
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
+    print(agent_code)
+
+    # 2. Fetch Questions
+    print(f"Fetching questions from: {questions_url}")
+    try:
+        response = requests.get(questions_url, timeout=15)
+        response.raise_for_status()
+        questions_data = response.json()
+        if not questions_data:
+            print("Fetched questions list is empty.")
+            return "Fetched questions list is empty or invalid format.", None
+        print(f"Fetched {len(questions_data)} questions.")
+    except requests.exceptions.RequestException as e:
+        print(f"Error fetching questions: {e}")
+        return f"Error fetching questions: {e}", None
+
+    # 3. Run your LlamaIndex Agent
+    results_log = []
+    answers_payload = []
+    print(f"Running agent on {len(questions_data)} questions...")
     
-    # Reinitialize agent with new LLM
-    agent = AgentWorkflow.from_tools_or_functions(tools, llm=llm)
-    return True
-
-# Run with fallback logic
-def run_with_fallback(query: str):
-    global current_model_index, llm, agent
-    
-    # Reset to first model if we're not already on it
-    if current_model_index != 0:
-        current_model_index = 0
-        llm = HuggingFaceInferenceAPI(
-            model_name=model_list[current_model_index],
-            token=hf_token,
-        )
-        agent = AgentWorkflow.from_tools_or_functions(tools, llm=llm)
+    # Create a new event loop for this function
+    loop = asyncio.new_event_loop()
+    asyncio.set_event_loop(loop)
     
-    # Try each model in sequence
-    for i in range(len(model_list)):
+    for item in questions_data:
+        task_id = item.get("task_id")
+        question_text = item.get("question")
+        if not task_id or question_text is None:
+            print(f"Skipping item with missing task_id or question: {item}")
+            continue
         try:
-            result = agent.run(query)
-            print(f"Successfully ran query with model: {model_list[current_model_index]}")
-            return result
+            # Run the async function in the loop
+            submitted_answer = loop.run_until_complete(run_agent(agent, question_text))
+            
+            # Ensure serializable response
+            if not isinstance(submitted_answer, (str, dict, list, int, float, bool, type(None))):
+                submitted_answer = str(submitted_answer)
+                
+            answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
+            
         except Exception as e:
-            print(f"Error with model {model_list[current_model_index]}: {e}")
-            if i < len(model_list) - 1:  # If not the last model
-                try_next_model()
-            else:
-                break
+            print(f"Error running agent on task {task_id}: {e}")
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
     
-    return "Sorry, encountered issues with all models."
+    # Close the loop when done
+    loop.close()
+
+    if not answers_payload:
+        print("Agent did not produce any answers to submit.")
+        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
+
+    # 4. Prepare Submission
+    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
+    status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
+    print(status_update)
+
+    # 5. Submit
+    print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
+    try:
+        response = requests.post(submit_url, json=submission_data, timeout=60)
+        response.raise_for_status()
+        result_data = response.json()
+        final_status = (
+            f"Submission Successful!\n"
+            f"User: {result_data.get('username')}\n"
+            f"Overall Score: {result_data.get('score', 'N/A')}% "
+            f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
+            f"Message: {result_data.get('message', 'No message received.')}"
+        )
+        print("Submission successful.")
+        results_df = pd.DataFrame(results_log)
+        return final_status, results_df
+    except requests.exceptions.RequestException as e:
+        print(f"Submission failed: {e}")
+        return f"Submission failed: {e}", pd.DataFrame(results_log)
+
+
+# Gradio Interface
+with gr.Blocks() as demo:
+    gr.Markdown("# LlamaIndex Agent Evaluation Runner")
+    
+    gr.LoginButton()
 
-# Make agent.run() work with asyncio by adding async support
-async def run(query: str):
-    """Async wrapper for the agent.run method to be compatible with app.py"""
-    return run_with_fallback(query)
+    run_button = gr.Button("Run Evaluation & Submit All Answers")
+
+    status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
+    results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
+
+    run_button.click(
+        fn=run_and_submit_all,
+        outputs=[status_output, results_table]
+    )
 
-# Add the async run method to the agent object
-agent.run = run_with_fallback  # Replace with synchronous version for direct calls
+if __name__ == "__main__":
+    print("Launching Gradio Interface for LlamaIndex Agent Evaluation...")
+    demo.launch(debug=True, share=False)