app.py

# app.py

# Import the necessary classes and functions from smolagents
from smolagents import CodeAgent, HfApiModel, load_tool, tool

# Standard library imports
import datetime
import pytz
import yaml

# External imports
# TODO: uncomment the import statements
#import torch
#from transformers import pipeline

# Import custom final answer tool and Gradio UI
from tools.final_answer import FinalAnswerTool
from Gradio_UI import GradioUI

# Initialize the Transformer-based sentiment analysis pipeline
#TODO: uncomment when testing using the real transformers
#sentiment_pipeline = pipeline("sentiment-analysis")

@tool
def my_custom_tool(arg1: str, arg2: int) -> str:
    """A tool that does nothing yet 
    Args:
        arg1: the first argument
        arg2: the second argument
    """
    return "What magic will you build?"

@tool
def get_current_time_in_timezone(timezone: str) -> str:
    """A tool that fetches the current local time in a specified timezone.
    Args:
        timezone: A string representing a valid timezone (e.g., 'America/New_York').
    """
    try:
        tz = pytz.timezone(timezone)
        local_time = datetime.datetime.now(tz).strftime("%Y-%m-%d %H:%M:%S")
        return f"The current local time in {timezone} is: {local_time}"
    except Exception as e:
        return f"Error fetching time for timezone '{timezone}': {str(e)}"

@tool
def advanced_sentiment_tool(text: str) -> str:
    """A tool that uses a pre-trained transformer model to do sentiment analysis.
    Args:
        text: The text to analyze for sentiment.
    """
    
    # TODO: uncomment later. for now test with hardcoded value first. later test using the real model
    # TODO: also uncomment the import statements
    #analysis = sentiment_pipeline(text)
    #label = analysis[0]['label']
    #score = analysis[0]['score']
    
    label = "positive"
    score = "0.99"
    
    return f"Sentiment: {label} (confidence: {score:.4f})"

@tool
def simple_sentiment_tool(text: str) -> str:
    """A tool that uses a pre-trained transformer model to do sentiment analysis.
    Args:
        text: The text to analyze for sentiment.
    """
    text = text.lower()
    
    if "happy" in text:
        return "Sentiment: Joyful (confidence: 1.00)"
    elif "sad" in text:
        return "Sentiment: Sorrowful (confidence: 1.00)"
    
    label = "positive"
    score = "0.99"
    
    return f"Sentiment: {label} (confidence: {score:.4f})"
    
# Final answer tool
final_answer = FinalAnswerTool()

# Initialize the model
model = HfApiModel(
    max_tokens=2096,
    temperature=0.5,
    model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
    custom_role_conversions=None,
)

# Load an image generation tool (unrelated, just for demonstration)
image_generation_tool = load_tool("agents-course/text-to-image", trust_remote_code=True)

# Load prompt templates
with open("prompts.yaml", 'r') as stream:
    prompt_templates = yaml.safe_load(stream)

# Initialize the agent, including the sentiment analysis tool
agent = CodeAgent(
    model=model,
    # TODO: use advanced_sentiment_tool later after testing using the simpler tool is done
    tools=[final_answer, simple_sentiment_tool],
    max_steps=6,
    verbosity_level=1,
    grammar=None,
    planning_interval=None,
    name=None,
    description=None,
    prompt_templates=prompt_templates
)

# Launch the Gradio UI
GradioUI(agent).launch()