Update app.py

app.py

@@ -3,174 +3,14 @@ import gradio as gr
 import requests
 import inspect
 import pandas as pd
-from google import genai
-from google.genai import types
 import time
-from smolagents import CodeAgent, DuckDuckGoSearchTool, WikipediaSearchTool, PythonInterpreterTool, FinalAnswerTool, LiteLLMModel, tool
+from smolagents import CodeAgent, DuckDuckGoSearchTool, WikipediaSearchTool, PythonInterpreterTool, FinalAnswerTool, LiteLLMModel
 
 
 # (Keep Constants as is)
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# --- Basic Agent Definition ---
-# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
-# class BasicAgent:
-#     def __init__(self):
-#         print("BasicAgent initialized.")
-#     def __call__(self, question: str) -> str:
-#         print(f"Agent received question (first 50 chars): {question[:50]}...")
-#         fixed_answer = "This is a default answer."
-#         print(f"Agent returning fixed answer: {fixed_answer}")
-#         return fixed_answer
-
-# class GeminiModel:
-#     def __init__(self, model_name="gemini-2.0-flash-exp"):
-#         api_key = os.getenv("GEMINI_API_KEY")
-#         if not api_key:
-#             raise ValueError("GEMINI_API_KEY is missing.")
-
-#         os.environ["GOOGLE_API_KEY"] = api_key
-#         self.client = genai.Client()
-#         self.model_id = model_name
-
-#         self.generation_config = types.GenerateContentConfig(
-#             temperature=0.4,
-#             top_p=0.9,
-#             top_k=40,
-#             candidate_count=1,
-#             seed=42,
-#             presence_penalty=0.0,
-#             frequency_penalty=0.0,
-#         )
-
-#     def __call__(self, prompt: str, **kwargs) -> str:
-#         """Send prompt to Gemini."""
-#         try:
-#             response = self.client.generate_content(
-#                 model=self.model_id,
-#                 contents=[{"role": "user", "parts": [{"text": prompt}]}],
-#                 generation_config=self.generation_config
-#             )
-#             # Return a dictionary that CodeAgent expects
-#             return {"content": response.candidates[0].content.parts[0].text.strip()}
-#         except Exception as e:
-#             return {"content": f"Error during Gemini call: {str(e)}"}
-
-# # Define BasicAgent properly
-# class BasicAgent:
-#     def __init__(self):
-#         print("Initializing CodeAgent with Gemini + tools.")
-
-#         # Load tools
-#         self.search_tool = DuckDuckGoSearchTool()
-
-#         # Build the agent
-#         self.agent = CodeAgent(
-#             tools=[self.search_tool],
-#             model=GeminiModel(),  # Our simple Gemini wrapper
-#             planning_interval=3  # Activate planning
-#         )
-
-#     def __call__(self, question: str) -> str:
-#         """Call the CodeAgent."""
-#         print(f"Running agent for task: {question[:50]}...")
-#         try:
-#             result = self.agent.run(question)
-
-#             # Sleep to respect rate limits
-#             time.sleep(7)
-#             return result
-#         except Exception as e:
-#             return f"Error running agent: {str(e)}"
-            
-# class BasicAgent(ReActAgent):
-#     def __init__(self):
-#         print("BasicAgent using local LLM initialized.")
-
-#         # Load a small model from Hugging Face
-#         model_name = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"  
-#         self.tokenizer = AutoTokenizer.from_pretrained(model_name)
-#         self.model = AutoModelForCausalLM.from_pretrained(
-#             model_name,
-#             torch_dtype=torch.float16,
-#             device_map="auto"  # Automatically choose GPU/CPU
-#         )
-
-#         super().__init__(tools=[])  # No tools for now
-
-#     def call(self, task: str) -> str:
-#         """Core method for answering a task."""
-#         prompt = f"Answer the following question concisely:\n\n{task}\n\nAnswer:"
-#         inputs = self.tokenizer(prompt, return_tensors="pt").to(self.model.device)
-
-#         with torch.no_grad():
-#             outputs = self.model.generate(
-#                 **inputs,
-#                 max_new_tokens=200,
-#                 do_sample=True,
-#                 temperature=0.7,
-#                 top_p=0.95,
-#                 top_k=50,
-#             )
-#         answer = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
-
-#         # Extract only the answer part
-#         return answer.split("Answer:")[-1].strip()
-
-# class BasicAgent:
-#     def __init__(self):
-#         print("BasicAgent using local LLM initialized.")
-
-#         # Load a small Hugging Face model
-#         model_name = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"  # Change if you want
-#         self.tokenizer = AutoTokenizer.from_pretrained(model_name)
-#         self.model = AutoModelForCausalLM.from_pretrained(
-#             model_name,
-#             torch_dtype=torch.float16,
-#             device_map="auto"  # Use GPU if available
-#         )
-
-#     def __call__(self, task: str) -> str:
-#         """Answer a question."""
-#         prompt = f"Answer the following question clearly and concisely:\n\n{task}\n\nAnswer:"
-#         inputs = self.tokenizer(prompt, return_tensors="pt").to(self.model.device)
-
-#         with torch.no_grad():
-#             outputs = self.model.generate(
-#                 **inputs,
-#                 max_new_tokens=256,
-#                 do_sample=True,
-#                 temperature=0.7,
-#                 top_p=0.9,
-#                 top_k=50,
-#             )
-#         decoded = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
-
-#         # Extract the answer part
-#         if "Answer:" in decoded:
-#             return decoded.split("Answer:")[-1].strip()
-#         return decoded.strip()
-
-
-
-# # Setup Gemini Client
-# api_key = os.getenv("GEMINI_API_KEY")
-# if not api_key:
-#     raise ValueError("GEMINI_API_KEY is missing.")
-# os.environ["GOOGLE_API_KEY"] = api_key
-# client = genai.Client()
-# model_id = "gemini-2.0-flash-exp"
-
-# generation_config = types.GenerateContentConfig(
-#     temperature=0.4,
-#     top_p=0.9,
-#     top_k=40,
-#     candidate_count=1,
-#     seed=42,
-#     presence_penalty=0.0,
-#     frequency_penalty=0.0,
-# )
 
     
 # Define the real agent
@@ -178,7 +18,7 @@ class BasicAgent:
     def __init__(self):
         print("Improved BasicAgent initialized with Gemini and enhanced tools.")
 
-        # Load Gemini through LiteLLM (ensure GEMINI_API_TOKEN is set)
+        # Load Gemini through LiteLLM
         self.model = LiteLLMModel(
             model_id="gemini/gemini-2.0-flash-lite",
             api_key=os.getenv("GEMINI_API_TOKEN"),
@@ -219,8 +59,8 @@ class BasicAgent:
             result = f"Error: {str(e)}"
 
         # Rate limiting to avoid 429 errors or API limits
-        print("Waiting 3 seconds to respect rate limits...")
-        time.sleep(3)
+        print("Waiting 5 seconds to respect rate limits...")
+        time.sleep(5)
 
         return result