wt002's picture
Update app.py
396779e verified
raw
history blame
31.6 kB
import os
import gradio as gr
import requests
import inspect
import pandas as pd
from smolagents import tool, Tool, CodeAgent, DuckDuckGoSearchTool, HfApiModel, VisitWebpageTool, SpeechToTextTool, FinalAnswerTool
from dotenv import load_dotenv
import heapq
from collections import Counter
import re
from io import BytesIO
from youtube_transcript_api import YouTubeTranscriptApi
from langchain_community.tools.tavily_search import TavilySearchResults
from langchain_community.document_loaders import WikipediaLoader
from langchain_community.utilities import WikipediaAPIWrapper
from langchain_community.document_loaders import ArxivLoader
# (Keep Constants as is)
# --- Constants ---
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
#Load environment variables
load_dotenv()
from langgraph.graph import END, StateGraph
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.messages import HumanMessage, AIMessage, ToolMessage
from langchain_core.tools import tool
from typing import Dict, List, TypedDict, Annotated
import operator
from langchain_community.llms import HuggingFaceHub
from langchain_community.chat_models import ChatHuggingFace
from langchain.schema import HumanMessage # Or your framework's equivalent
def init_state(question: str):
return {
"question": question,
"history": [HumanMessage(content=question)],
"context": {} # <- Add this line
}
# ====== Tool Definitions ======
@tool
def duckduckgo_search(query: str) -> str:
"""Search web using DuckDuckGo. Returns top 3 results."""
from duckduckgo_search import DDGS
with DDGS() as ddgs:
return "\n\n".join(
f"Title: {res['title']}\nURL: {res['href']}\nSnippet: {res['body']}"
for res in ddgs.text(query, max_results=3)
)
@tool
def wikipedia_search(query: str) -> str:
"""Get Wikipedia summaries. Returns first 3 sentences."""
import wikipedia
try:
return wikipedia.summary(query, sentences=3)
except wikipedia.DisambiguationError as e:
return f"Disambiguation options: {', '.join(e.options[:3])}"
except wikipedia.PageError:
return "Page not found"
@tool
def arxiv_search(query: str) -> str:
"""Search academic papers on arXiv. Returns top 3 results."""
import arxiv
results = arxiv.Search(
query=query,
max_results=3,
sort_by=arxiv.SortCriterion.Relevance
).results()
return "\n\n".join(
f"Title: {r.title}\nAuthors: {', '.join(a.name for a in r.authors)}\n"
f"Published: {r.published.strftime('%Y-%m-%d')}\nSummary: {r.summary[:250]}..."
for r in results
)
@tool
def document_qa(input_str: str) -> str:
"""Answer questions from documents. Input format: 'document_text||question'"""
from transformers import pipeline
if '||' not in input_str:
return "Invalid format. Use: 'document_text||question'"
context, question = input_str.split('||', 1)
qa_model = pipeline('question-answering', model='deepset/roberta-base-squad2')
return qa_model(question=question, context=context)['answer']
@tool
def python_execution(code: str) -> str:
"""Execute Python code and return output."""
try:
# Create isolated environment
env = {}
exec(f"def __exec_fn__():\n {indent_code(code)}\nresult = __exec_fn__()", env)
return str(env.get('result', 'No output'))
except Exception as e:
return f"Error: {str(e)}"
from typing import Optional
from langchain_core.tools import BaseTool
from youtube_transcript_api import YouTubeTranscriptApi
class VideoTranscriptionTool(BaseTool):
name: str = "transcript_video"
description: str = "Fetch text transcript from YouTube videos using URL or ID. Optionally include timestamps."
def _run(self, url: str, include_timestamps: Optional[bool] = False) -> str:
# Extract video ID
video_id = None
if "youtube.com/watch?v=" in url:
video_id = url.split("v=")[1].split("&")[0]
elif "youtu.be/" in url:
video_id = url.split("youtu.be/")[1].split("?")[0]
elif len(url.strip()) == 11 and not ("http://" in url or "https://" in url):
video_id = url.strip()
if not video_id:
return f"Invalid or unsupported YouTube URL/ID: {url}"
try:
transcription = YouTubeTranscriptApi.get_transcript(video_id)
if include_timestamps:
formatted = []
for part in transcription:
timestamp = f"{int(part['start']//60)}:{int(part['start']%60):02d}"
formatted.append(f"[{timestamp}] {part['text']}")
return "\n".join(formatted)
else:
return " ".join([part['text'] for part in transcription])
except Exception as e:
return f"Error fetching transcript: {str(e)}"
def _arun(self, *args, **kwargs):
raise NotImplementedError("Async not supported for this tool.")
import os
import time
import json
from typing import TypedDict, List, Union, Any, Dict
from langchain_huggingface import ChatHuggingFace
from langchain_huggingface.llms import HuggingFaceEndpoint
from langchain.schema import HumanMessage, AIMessage, SystemMessage
from langchain.prompts import ChatPromptTemplate
from langgraph.graph import StateGraph, END
from langchain.tools import Tool
# Assume these tools are defined elsewhere and imported
# Placeholder for your actual tool implementations
def duckduckgo_search(query: str) -> str:
"""Performs a DuckDuckGo search for current events or general facts."""
print(f"DEBUG: duckduckgo_search called with: {query}")
return f"Search result for '{query}': Example relevant information from web."
def wikipedia_search(query: str) -> str:
"""Searches Wikipedia for encyclopedic information."""
print(f"DEBUG: wikipedia_search called with: {query}")
return f"Wikipedia result for '{query}': Found detailed article."
def arxiv_search(query: str) -> str:
"""Searches ArXiv for scientific preprints and papers."""
print(f"DEBUG: arxiv_search called with: {query}")
return f"ArXiv result for '{query}': Found relevant research paper."
def document_qa(document_path: str, question: str) -> str:
"""Answers questions based on the content of a given document file (PDF, DOCX, TXT)."""
print(f"DEBUG: document_qa called with: {document_path}, question: {question}")
return f"Document QA result for '{question}': Answer extracted from document."
def python_execution(code: str) -> str:
"""Executes Python code in a sandboxed environment for calculations or data manipulation."""
try:
exec_globals = {}
exec_locals = {}
# WARNING: This is a highly insecure way to execute arbitrary Python code.
# For production, use a secure, sandboxed environment (e.g., Docker container, dedicated service).
exec(code, exec_globals, exec_locals)
return str(exec_locals.get('result', 'Code executed, no explicit result assigned to "result" variable.'))
except Exception as e:
return f"Python execution error: {str(e)}"
class VideoTranscriptionTool:
"""Transcribes and analyzes video content from a URL or ID."""
def __call__(self, video_id_or_url: str) -> str:
print(f"DEBUG: VideoTranscriptionTool called with: {video_id_or_url}")
return f"Video transcription/analysis result for '{video_id_or_url}': Summary of video content."
# --- Agent State Definition ---
class AgentState(TypedDict):
question: str
history: List[Union[HumanMessage, AIMessage]]
context: Dict[str, Any]
reasoning: str
iterations: int
final_answer: Union[str, float, int, None]
current_task: str
current_thoughts: str
tools: List[Tool]
# --- Utility Functions ---
def parse_agent_response(response_content: str) -> tuple[str, str, str]:
"""
Parses the LLM's JSON output for reasoning, action, and action input.
Returns (reasoning, action, action_input).
If JSON parsing fails, it attempts heuristic parsing.
"""
try:
response_json = json.loads(response_content)
reasoning = response_json.get("Reasoning", "").strip()
action = response_json.get("Action", "").strip()
action_input = response_json.get("Action Input", "").strip()
return reasoning, action, action_input
except json.JSONDecodeError:
print(f"WARNING: JSONDecodeError: LLM response was not valid JSON. Attempting heuristic parse: {response_content[:200]}...")
# Heuristic parsing for non-JSON or partial JSON responses
reasoning = ""
action = ""
action_input = ""
# Attempt to find Reasoning
reasoning_idx = response_content.find("Reasoning:")
action_idx = response_content.find("Action:")
if reasoning_idx != -1 and action_idx != -1 and reasoning_idx < action_idx:
reasoning = response_content[reasoning_idx + len("Reasoning:"):action_idx].strip()
# Clean up leading/trailing quotes if present
if reasoning.startswith('"') and reasoning.endswith('"'):
reasoning = reasoning[1:-1]
elif reasoning_idx != -1: # If only reasoning is found
reasoning = response_content[reasoning_idx + len("Reasoning:"):].strip()
if reasoning.startswith('"') and reasoning.endswith('"'):
reasoning = reasoning[1:-1]
# Attempt to find Action and Action Input
if action_idx != -1:
action_input_idx = response_content.find("Action Input:", action_idx)
if action_input_idx != -1:
action_part = response_content[action_idx + len("Action:"):action_input_idx].strip()
action = action_part
action_input = response_content[action_input_idx + len("Action Input:"):].strip()
else:
action = response_content[action_idx + len("Action:"):].strip()
# Clean up action and action_input
if action.startswith('"') and action.endswith('"'):
action = action[1:-1]
if action_input.startswith('"') and action_input.endswith('"'):
action_input = action_input[1:-1]
# Final cleanup for any trailing JSON artifacts if heuristic grabs too much
action = action.split('"', 1)[0].strip() # Stop at first quote for safety if it's "Action": "tool_name",
action_input = action_input.split('"', 1)[0].strip() # Similar for input
return reasoning, action, action_input
# --- Graph Nodes ---
def should_continue(state: AgentState) -> str:
"""
Determines if the agent should continue reasoning, use a tool, or end.
"""
print(f"DEBUG: Entering should_continue. Current context: {state.get('context', {})}")
if state.get("final_answer") is not None:
print("DEBUG: should_continue -> END (Final Answer set in state)")
return "end"
if state.get("context", {}).get("pending_action"):
print("DEBUG: should_continue -> ACTION (Pending action in context)")
return "action"
print("DEBUG: should_continue -> REASON (Default to reasoning)")
return "reason"
def reasoning_node(state: AgentState) -> AgentState:
"""
Node for the agent to analyze the question, determine next steps,
and select tools.
"""
print(f"DEBUG: Entering reasoning_node. Iteration: {state['iterations']}")
print(f"DEBUG: Current history length: {len(state.get('history', []))}")
HF_TOKEN = os.getenv("HF_TOKEN")
if not HF_TOKEN:
raise ValueError("HF_TOKEN not set in environment variables.")
state.setdefault("context", {})
state.setdefault("reasoning", "")
state.setdefault("iterations", 0)
state.setdefault("current_task", "Understand the question and plan the next step.")
state.setdefault("current_thoughts", "")
state["context"].pop("pending_action", None)
model_id = "meta-llama/Llama-2-7b-chat-hf"
llm = ChatHuggingFace(
llm=HuggingFaceEndpoint(
repo_id=model_id,
max_new_tokens=1024, # Increased max_new_tokens
temperature=0.1,
huggingfacehub_api_token=HF_TOKEN,
)
)
tool_descriptions = "\n".join([
f"- **{t.name}**: {t.description}" for t in state.get("tools", [])
])
system_prompt = (
"You are an expert problem solver, designed to provide concise and accurate answers. "
"Your process involves analyzing the question, intelligently selecting and using tools, "
"and synthesizing information.\n\n"
"**Available Tools:**\n"
f"{tool_descriptions}\n\n"
"**Tool Usage Guidelines:**\n"
"- Use **duckduckgo_search** for current events, general facts, or quick lookups.\n"
"- Use **wikipedia_search** for encyclopedic information, historical context, or detailed topics.\n"
"- Use **arxiv_search** for scientific papers, research, or cutting-edge technical information.\n"
"- Use **document_qa** when the question explicitly refers to a specific document file (e.g., 'Analyze this PDF').\n"
"- Use **python_execution** for complex calculations, data manipulation, or logical operations that cannot be done with simple reasoning. Always provide the full Python code, ensuring it's valid and executable, and assign the final result to a variable named 'result' if applicable.\n"
"- Use **VideoTranscriptionTool** for any question involving video or audio content. Provide the full YouTube URL or video ID.\n\n"
"**Current Context:**\n{context}\n\n"
"**Previous Reasoning Steps:**\n{reasoning}\n\n"
"**Current Task:** {current_task}\n"
"**Current Thoughts:** {current_thoughts}\n\n"
"**Your Response MUST be a valid JSON object with the following keys:**\n"
"```json\n"
"{\n"
" \"Reasoning\": \"Your detailed analysis of the question and why you chose a specific action.\",\n"
" \"Action\": \"[Tool name OR 'Final Answer']\",\n"
" \"Action Input\": \"[Input for the selected tool OR the final response]\"\n"
"}\n"
"```\n"
"**CRITICAL RULE: 'Action' and 'Action Input' MUST NOT be empty, unless 'Action' is 'Final Answer' and 'Action Input' is the conclusive response.**\n"
"If you cannot determine a suitable tool or a final answer, return Action: 'Final Answer' with a message like 'I cannot answer this question with the available tools.' or 'More information is needed.'\n"
"Ensure 'Action Input' is appropriate for the chosen 'Action'. If 'Action' is 'Final Answer', provide the complete, concise answer."
)
prompt = ChatPromptTemplate.from_messages([
SystemMessage(content=system_prompt),
*state["history"]
])
chain = prompt | llm
def call_with_retry(inputs, retries=3, delay=30):
for attempt in range(retries):
try:
response = chain.invoke(inputs)
# --- NEW DEBUGGING PRINT ---
print(f"DEBUG: RAW LLM Response (Attempt {attempt+1}):\n---\n{response.content}\n---")
# --- END NEW DEBUGGING PRINT ---
json.loads(response.content) # Attempt to parse to validate structure
return response
except json.JSONDecodeError as e:
print(f"[Retry {attempt+1}/{retries}] LLM returned invalid JSON. Error: {e}. Retrying...")
print(f"Invalid JSON content (partial): {response.content[:200]}...")
# Add specific error message to history to guide LLM
state["history"].append(AIMessage(content=f"[Parsing Error] The previous LLM output was not valid JSON. Expected format: ```json{{\"Reasoning\": \"...\", \"Action\": \"...\", \"Action Input\": \"...\"}}```. Please ensure your response is ONLY valid JSON and strictly follows the format. Error: {e}"))
time.sleep(5)
except Exception as e:
print(f"[Retry {attempt+1}/{retries}] An unexpected error occurred during LLM call: {e}. Waiting {delay}s...")
state["history"].append(AIMessage(content=f"[API Error] Failed to get a response from the LLM due to an API error: {e}. Trying again."))
time.sleep(delay)
raise RuntimeError("Failed after multiple retries due to Hugging Face API issues or invalid JSON.")
response = call_with_retry({
"context": state["context"],
"reasoning": state["reasoning"],
"question": state["question"],
"current_task": state["current_task"],
"current_thoughts": state["current_thoughts"]
})
content = response.content
reasoning, action, action_input = parse_agent_response(content)
print(f"DEBUG: Parsed Action: '{action}', Action Input: '{action_input[:100]}...'")
# Only append the LLM's raw output if it was valid JSON and processed successfully
# Otherwise, the specific error message from the retry loop will already be in history.
if isinstance(response, AIMessage) and content == response.content: # Check if it's the original response, not an error message
state["history"].append(AIMessage(content=content))
state["reasoning"] += f"\nStep {state['iterations'] + 1}: {reasoning}"
state["iterations"] += 1
state["current_thoughts"] = reasoning
if "final answer" in action.lower():
state["final_answer"] = action_input
else:
state["context"]["pending_action"] = {
"tool": action,
"input": action_input
}
# Add a message to history to indicate the agent's intent for the LLM
# This will be shown to the LLM in the next turn.
state["history"].append(AIMessage(content=f"Agent decided to use tool: {action} with input: {action_input}"))
print(f"DEBUG: Exiting reasoning_node. New history length: {len(state['history'])}")
return state
def tool_node(state: AgentState) -> AgentState:
"""
Node for executing the chosen tool and returning its output.
"""
print(f"DEBUG: Entering tool_node. Iteration: {state['iterations']}")
tool_call_dict = state["context"].pop("pending_action", None)
if not tool_call_dict:
error_message = "[Tool Error] No pending_action found in context. This indicates an issue with graph flow."
print(f"ERROR: {error_message}")
state["history"].append(AIMessage(content=error_message))
return state
tool_name = tool_call_dict.get("tool")
tool_input = tool_call_dict.get("input")
if not tool_name or tool_input is None:
error_message = f"[Tool Error] Invalid action request from LLM: Tool name '{tool_name}' or input '{tool_input}' was empty. LLM needs to provide valid 'Action' and 'Action Input'."
print(f"ERROR: {error_output}") # Print the error message for debugging
state["history"].append(AIMessage(content=error_message))
state["context"].pop("pending_action", None)
return state
available_tools = state.get("tools", [])
tool_fn = next((t for t in available_tools if t.name == tool_name), None)
if tool_fn is None:
tool_output = f"[Tool Error] Tool '{tool_name}' not found or not available. Please choose from: {', '.join([t.name for t in available_tools])}"
print(f"ERROR: {tool_output}")
else:
try:
print(f"DEBUG: Invoking tool '{tool_name}' with input: '{tool_input[:100]}...'")
tool_output = tool_fn.run(tool_input)
if not tool_output and tool_output is not False:
tool_output = f"[{tool_name} output] No specific result found for '{tool_input}'. The tool might have returned an empty response."
except Exception as e:
tool_output = f"[Tool Error] An error occurred while running '{tool_name}': {str(e)}"
print(f"ERROR: {tool_output}")
state["history"].append(AIMessage(content=f"[{tool_name} output]\n{tool_output}"))
print(f"DEBUG: Exiting tool_node. Tool output added to history. New history length: {len(state['history'])}")
return state
# ====== Agent Graph ======
def create_agent_workflow(tools: List[Tool]):
workflow = StateGraph(AgentState)
workflow.add_node("reason", reasoning_node)
workflow.add_node("action", tool_node)
workflow.set_entry_point("reason")
workflow.add_conditional_edges(
"reason",
should_continue,
{
"action": "action",
"reason": "reason",
"end": END
}
)
workflow.add_edge("action", "reason")
app = workflow.compile()
return app
# ====== Agent Interface ======
class BasicAgent:
def __init__(self):
self.tools = [
Tool(name="duckduckgo_search", func=duckduckgo_search, description="Performs a DuckDuckGo search for current events or general facts."),
Tool(name="wikipedia_search", func=wikipedia_search, description="Searches Wikipedia for encyclopedic information."),
Tool(name="arxiv_search", func=arxiv_search, description="Searches ArXiv for scientific preprints and papers."),
Tool(name="document_qa", func=document_qa, description="Answers questions based on the content of a given document file (PDF, DOCX, TXT). Requires 'document_path' and 'question' as input."),
Tool(name="python_execution", func=python_execution, description="Executes Python code in a sandboxed environment for complex calculations or data manipulation."),
Tool(name="VideoTranscriptionTool", func=VideoTranscriptionTool(), description="Transcribes and analyzes video content from a URL or ID. Use for any question involving video or audio.")
]
self.workflow = create_agent_workflow(self.tools)
def __call__(self, question: str) -> str:
print(f"\n--- Agent received question: {question[:80]}{'...' if len(question) > 80 else ''} ---")
state = {
"question": question,
"context": {},
"reasoning": "",
"iterations": 0,
"history": [HumanMessage(content=question)],
"final_answer": None,
"current_task": "Understand the question and plan the next step.",
"current_thoughts": "",
"tools": self.tools
}
final_state = self.workflow.invoke(state)
if final_state.get("final_answer") is not None:
answer = final_state["final_answer"]
print(f"--- Agent returning FINAL ANSWER: {answer} ---")
return answer
else:
print(f"--- ERROR: Agent finished without setting 'final_answer' for question: {question} ---")
# You might want to return the last message from history if no final answer was set
if final_state["history"]:
last_message = final_state["history"][-1].content
print(f"Last message in history: {last_message}")
return f"Agent could not fully answer. Last message: {last_message}"
else:
raise ValueError("Agent finished without providing a final answer and no history messages.")
def run_and_submit_all( profile: gr.OAuthProfile | None):
"""
Fetches all questions, runs the BasicAgent 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 Agent ( modify this part to create your agent)
try:
agent = BasicAgent()
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)
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
except requests.exceptions.JSONDecodeError as e:
print(f"Error decoding JSON response from questions endpoint: {e}")
print(f"Response text: {response.text[:500]}")
return f"Error decoding server response for questions: {e}", None
except Exception as e:
print(f"An unexpected error occurred fetching questions: {e}")
return f"An unexpected error occurred fetching questions: {e}", None
# 3. Run your Agent
results_log = []
answers_payload = []
print(f"Running agent on {len(questions_data)} questions...")
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:
submitted_answer = agent(question_text)
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 running agent on task {task_id}: {e}")
results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
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.HTTPError as e:
error_detail = f"Server responded with status {e.response.status_code}."
try:
error_json = e.response.json()
error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
except requests.exceptions.JSONDecodeError:
error_detail += f" Response: {e.response.text[:500]}"
status_message = f"Submission Failed: {error_detail}"
print(status_message)
results_df = pd.DataFrame(results_log)
return status_message, results_df
except requests.exceptions.Timeout:
status_message = "Submission Failed: The request timed out."
print(status_message)
results_df = pd.DataFrame(results_log)
return status_message, results_df
except requests.exceptions.RequestException as e:
status_message = f"Submission Failed: Network error - {e}"
print(status_message)
results_df = pd.DataFrame(results_log)
return status_message, results_df
except Exception as e:
status_message = f"An unexpected error occurred during submission: {e}"
print(status_message)
results_df = pd.DataFrame(results_log)
return status_message, results_df
# --- Build Gradio Interface using Blocks ---
with gr.Blocks() as demo:
gr.Markdown("# Basic Agent Evaluation Runner")
gr.Markdown(
"""
**Instructions:**
1. Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
2. Log in to your Hugging Face account using the button below. This uses your HF username for submission.
3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
---
**Disclaimers:**
Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async.
"""
)
gr.LoginButton()
run_button = gr.Button("Run Evaluation & Submit All Answers")
status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
# Removed max_rows=10 from DataFrame constructor
results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
run_button.click(
fn=run_and_submit_all,
outputs=[status_output, results_table]
)
if __name__ == "__main__":
print("\n" + "-"*30 + " App Starting " + "-"*30)
# Check for SPACE_HOST and SPACE_ID at startup for information
space_host_startup = os.getenv("SPACE_HOST")
space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup
if space_host_startup:
print(f"✅ SPACE_HOST found: {space_host_startup}")
print(f" Runtime URL should be: https://{space_host_startup}.hf.space")
else:
print("ℹ️ SPACE_HOST environment variable not found (running locally?).")
if space_id_startup: # Print repo URLs if SPACE_ID is found
print(f"✅ SPACE_ID found: {space_id_startup}")
print(f" Repo URL: https://huggingface.co/spaces/{space_id_startup}")
print(f" Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main")
else:
print("ℹ️ SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined.")
print("-"*(60 + len(" App Starting ")) + "\n")
print("Launching Gradio Interface for Basic Agent Evaluation...")
demo.launch(debug=True, share=False)