app.py

import requests
import geopy
import joblib
import gradio as gr

# Load the trained model
model = joblib.load('hackathonrf.joblib')

# Function to get latitude and longitude from location name
def get_coordinates(location):
    geolocator = geopy.geocoders.Nominatim(user_agent="air_quality_app")
    location = geolocator.geocode(location)
    return location.latitude, location.longitude, location.address

# Function to get AQI value from OpenWeatherMap API
def get_aqi(latitude, longitude):
    api_key = "78b94879cbb50e02397e93687aa24adc"  # Hidden API Key
    url = f"http://api.openweathermap.org/data/2.5/air_pollution?lat={latitude}&lon={longitude}&appid={api_key}"
    response = requests.get(url)
    data = response.json()
    if 'list' in data and data['list']:
        aqi_value = data['list'][0]['main']['aqi']
        return aqi_value
    else:
        return None

# Function to convert AQI value to corresponding label
def convert_to_label(aqi_value):
    if aqi_value is None:
        return "unknown"
    elif aqi_value <= 50:
        return "good"
    elif aqi_value <= 100:
        return "moderate"
    elif aqi_value <= 150:
        return "unhealthy for sensitive groups"
    elif aqi_value <= 200:
        return "unhealthy"
    elif aqi_value <= 300:
        return "very unhealthy"
    else:
        return "hazardous"

# Function to make prediction
def predict_air_quality(location):
    latitude, longitude, city = get_coordinates(location)
    aqi_value = get_aqi(latitude, longitude)
    air_quality_label = convert_to_label(aqi_value)
    return f"{city} air quality is currently '{air_quality_label}', with AQI {aqi_value}"

# Create Gradio interface
iface = gr.Interface(fn=predict_air_quality, 
                      inputs=["text"], 
                      outputs="text", 
                      title="Air Quality Prediction",
                      description="Enter location:")
iface.launch()