merttaymaz's picture
Update app.py
8c47dfc verified
from smolagents import CodeAgent, DuckDuckGoSearchTool, HfApiModel, load_tool, tool
import datetime
import requests
import pytz
import yaml
import os
import numpy as np
import cv2
import gradio as gr
from tools.final_answer import FinalAnswerTool
from PIL import Image
from Gradio_UI import GradioUI
# OpenCV Haarcascade face model
HAAR_CASCADE_PATH = cv2.data.haarcascades + 'haarcascade_frontalface_default.xml'
# Supported mime types
ALLOWED_MIME_TYPES = ("image/jpeg", "image/png")
# Face detection parameters
FACE_DETECTION_SCALE_FACTOR = 1.1
FACE_DETECTION_MIN_NEIGHBORS = 5
MIN_FACE_SIZE = (80, 80) # Min Face Size (width, height)
@tool
def check_passport_photo(image: Image.Image) -> bool:
"""
Check if the given image is a valid passport photo.
Args:
image: The image file uploaded through Gradio UI.
Returns:
bool: True if the image is a valid passport photo, False otherwise.
"""
if image is None:
raise ValueError("No image uploaded.")
# Convert PIL image to NumPy array
image = np.array(image) # PIL -> NumPy
image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR) # Convert RGB to BGR (OpenCV uses BGR)
if image is None:
raise ValueError("Invalid or corrupted image file.")
# Convert to grayscale
gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# Load face detector
face_detector = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
# Detect faces
detected_faces = face_detector.detectMultiScale(gray_image, scaleFactor=1.1, minNeighbors=5, minSize=(80, 80))
return len(detected_faces) == 1
@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)}"
final_answer = FinalAnswerTool()
model = HfApiModel(
max_tokens=2096,
temperature=0.5,
model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
custom_role_conversions=None,
)
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, check_passport_photo], # check_passport_photo fonksiyonunu ekledik
max_steps=6,
verbosity_level=1,
grammar=None,
planning_interval=None,
name=None,
description=None,
prompt_templates=prompt_templates
)
def gradio_check_passport_photo(image):
return check_passport_photo(image)
iface = gr.Interface(fn=gradio_check_passport_photo, inputs="image", outputs="text")
iface.launch()