import pandas as pd
import numpy as np
import folium
from sklearn.cluster import KMeans
import gradio as gr
from sklearn.preprocessing import StandardScaler

# Function to load data from CSV file
def load_data(file):
    data = pd.read_csv(file)
    return data

# Function to preprocess and apply clustering model to classify soil moisture
def process_data(data):
    # Preprocessing data (standardizing)
    X = data[['temperature', 'humidity', 'soil_type']]  # Example feature columns
    scaler = StandardScaler()
    X_scaled = scaler.fit_transform(X)  # Scaling the data
    
    # Clustering model: KMeans (you can replace with regression models as well)
    kmeans = KMeans(n_clusters=3, random_state=42)
    data['moisture_category'] = kmeans.fit_predict(X_scaled)  # Classify into moisture categories (Low, Medium, High)
    
    return data

# Function to create an interactive folium map
def create_map(data):
    # Initializing the map at the center of the provided data
    m = folium.Map(location=[data['latitude'].mean(), data['longitude'].mean()], zoom_start=12)
    
    # Plotting data points on the map
    for _, row in data.iterrows():
        color = 'blue' if row['moisture_category'] == 0 else 'green' if row['moisture_category'] == 1 else 'red'
        folium.CircleMarker(location=[row['latitude'], row['longitude']],
                            radius=10,
                            color=color,
                            fill=True).add_to(m)
    return m

# Main function to process the uploaded file, analyze data, and create the map
def soil_moisture_mapping(file):
    # Load data
    data = load_data(file)
    
    # Process data and apply the model
    processed_data = process_data(data)
    
    # Generate map with the results
    map_result = create_map(processed_data)
    
    return map_result

# Setting up the Gradio interface for file upload
iface = gr.Interface(fn=soil_moisture_mapping, inputs="file", outputs="html", live=True)

# Launch the Gradio application
iface.launch()