MilindChawre's picture
Added image generation and translator tool
d0cc95b
from smolagents import CodeAgent,DuckDuckGoSearchTool, HfApiModel,load_tool,tool
import datetime
import requests
import pytz
import yaml
from tools.final_answer import FinalAnswerTool
from Gradio_UI import GradioUI
# Below is an example of a tool that does nothing. Amaze us with your creativity !
@tool
def my_custom_tool(arg1:str, arg2:int)-> str: #it's import to specify the return type
#Keep this format for the description / args / args description but feel free to modify the tool
"""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:
# Create timezone object
tz = pytz.timezone(timezone)
# Get current time in that 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 image_generator(prompt: str) -> str:
"""A tool that generates an image based on a text prompt.
Args:
prompt: A text description of the image to generate.
"""
try:
# Use the imported image generation tool
image_path = image_generation_tool(prompt=prompt)
return image_path
except Exception as e:
return f"Error generating image: {str(e)}"
@tool
def translator(question: str, src_lang: str, tgt_lang: str) -> str:
"""A tool that translates text from one language to another.
Args:
question: The text to translate.
src_lang: The source language of the text.
tgt_lang: The target language to translate to.
"""
try:
# Import the deep_translator library
from deep_translator import GoogleTranslator
# Map common language names to language codes if needed
language_map = {
"english": "en", "french": "fr", "spanish": "es", "german": "de",
"italian": "it", "portuguese": "pt", "russian": "ru", "japanese": "ja",
"chinese": "zh-CN", "korean": "ko", "arabic": "ar", "hindi": "hi"
}
# Convert language names to codes if necessary
src = language_map.get(src_lang.lower(), src_lang.lower())
tgt = language_map.get(tgt_lang.lower(), tgt_lang.lower())
# Perform the translation
translated = GoogleTranslator(source=src, target=tgt).translate(question)
return translated
except Exception as e:
return f"Error translating text: {str(e)}"
@tool
def search(query: str) -> str:
"""A tool that searches the web for information and returns a list of relevant pages.
Args:
query: The search query.
"""
try:
# Use DuckDuckGo search tool that's already imported
search_tool = DuckDuckGoSearchTool()
results = search_tool(query)
# Check if results is empty
if not results:
return f"No result found for query \"{query}\"."
# Handle different return formats
formatted_results = []
# Check if results is a list of strings or objects
if isinstance(results, list):
if results and isinstance(results[0], str):
# Handle list of strings
formatted_results = results[:5] # Limit to top 5 results
else:
# Try to handle objects with title/url attributes
for result in results[:5]:
try:
if hasattr(result, 'title') and hasattr(result, 'url'):
formatted_results.append(f"[{result.title}]({result.url})")
elif isinstance(result, dict) and 'title' in result and 'url' in result:
formatted_results.append(f"[{result['title']}]({result['url']})")
else:
formatted_results.append(str(result))
except:
formatted_results.append(str(result))
elif isinstance(results, str):
# Handle string result
return results
return f"Found {len(formatted_results)} pages:\n" + "\n".join(formatted_results)
except Exception as e:
return f"Error searching the web: {str(e)}"
@tool
def web_search(query: str) -> str:
"""A tool that searches the web for information and returns detailed results.
Args:
query: The search query.
"""
try:
# Use DuckDuckGo search tool that's already imported
search_tool = DuckDuckGoSearchTool()
results = search_tool(query)
# Check if results is empty
if not results:
return f"No results found for query \"{query}\"."
# Handle different return formats
if isinstance(results, str):
return results
combined_info = ""
# Process up to 3 results
count = 0
for result in results:
if count >= 3:
break
try:
if hasattr(result, 'snippet'):
combined_info += f"Result {count+1}: {result.snippet}\n\n"
elif isinstance(result, dict) and 'snippet' in result:
combined_info += f"Result {count+1}: {result['snippet']}\n\n"
elif hasattr(result, 'title'):
combined_info += f"Result {count+1}: {result.title}\n\n"
elif isinstance(result, dict) and 'title' in result:
combined_info += f"Result {count+1}: {result['title']}\n\n"
else:
combined_info += f"Result {count+1}: {str(result)}\n\n"
count += 1
except:
combined_info += f"Result {count+1}: {str(result)}\n\n"
count += 1
return combined_info.strip() if combined_info else "No detailed information found."
except Exception as e:
return f"Error searching the web: {str(e)}"
@tool
def visit_webpage(url: str) -> str:
"""A tool that visits a webpage and returns its content.
Args:
url: The URL of the webpage to visit.
"""
try:
# Add headers to mimic a browser request
headers = {
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.114 Safari/537.36'
}
# Make the request
response = requests.get(url, headers=headers, timeout=10)
response.raise_for_status()
# Try to extract main content using a simple approach
from bs4 import BeautifulSoup
soup = BeautifulSoup(response.text, 'html.parser')
# Remove script and style elements
for script in soup(["script", "style"]):
script.extract()
# Get text
text = soup.get_text()
# Break into lines and remove leading and trailing space on each
lines = (line.strip() for line in text.splitlines())
# Break multi-headlines into a line each
chunks = (phrase.strip() for line in lines for phrase in line.split(" "))
# Remove blank lines
text = '\n'.join(chunk for chunk in chunks if chunk)
# Limit the length of the response
max_length = 5000
if len(text) > max_length:
text = text[:max_length] + "... (content truncated)"
return text
except Exception as e:
return f"Error visiting webpage: {str(e)}"
final_answer = FinalAnswerTool()
# If the agent does not answer, the model is overloaded, please use another model or the following Hugging Face Endpoint that also contains qwen2.5 coder:
# model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud'
model = HfApiModel(
max_tokens=2096,
temperature=0.5,
#model_id='Qwen/Qwen2.5-Coder-32B-Instruct',# it is possible that this model may be overloaded
model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud',
custom_role_conversions=None,
)
# Import tool from Hub
image_generation_tool = load_tool("agents-course/text-to-image", trust_remote_code=True)
with open("prompts.yaml", 'r') as stream:
prompt_templates = yaml.safe_load(stream)
agent = CodeAgent(
model=model,
tools=[final_answer, image_generator, translator, get_current_time_in_timezone, search, web_search, visit_webpage], ## add your tools here (don't remove final answer)
max_steps=6,
verbosity_level=1,
grammar=None,
planning_interval=None,
name=None,
description=None,
prompt_templates=prompt_templates
)
GradioUI(agent).launch()