import gradio as gr
import pandas as pd
import lightgbm as lgb
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
import os
import torch
from torchvision import models, transforms
from PIL import Image


# ---------------------------
# Crop Recommendation Setup
# ---------------------------
url = "https://raw.githubusercontent.com/Pushpinder-Singh06/CSV-Files/refs/heads/main/crop_cleaned%20data.csv "
data = pd.read_csv(url)

X = data.drop('label', axis=1)
y = data['label']
le = LabelEncoder()
y_encoded = le.fit_transform(y)

X_train, X_test, y_train, y_test = train_test_split(X, y_encoded, test_size=0.3, random_state=0)
model = lgb.LGBMClassifier()
model.fit(X_train, y_train)

def predict_crop(nitrogen, phosphorus, potassium, temperature, humidity, soil_pH, rainfall):
    input_data = np.array([[nitrogen, phosphorus, potassium, temperature, humidity, soil_pH, rainfall]])
    pred = model.predict(input_data)[0]
    crop_name = le.inverse_transform([pred])[0]
    image_path = f"crop_images/{crop_name}.jpeg"
    if not os.path.exists(image_path):
        image_path = None  
    return image_path, f"🌾 Recommended crop for your field:  *{crop_name}*"

with gr.Blocks() as demo:
    gr.Markdown("# 🌾 **Which Crop Should I Grow?**")

    with gr.Tabs():
        with gr.Row():
            nitrogen = gr.Slider(0, 140, step=1, label="Nitrogen (kg/ha)")
            phosphorus = gr.Slider(5, 95, step=1, label="Phosphorus (kg/ha)")
            potassium = gr.Slider(5, 82, step=1, label="Potassium (kg/ha)")
        with gr.Row():
            temperature = gr.Slider(15.63, 36.32, step=0.1, label="Temperature (°C)")
            humidity = gr.Slider(14.2, 99.98, step=1, label="Humidity (%)")
        with gr.Row():
            soil_pH = gr.Slider(0, 14, step=0.1, label="Soil pH")
            rainfall = gr.Slider(20.21, 253.72, step=1, label="Rainfall (mm)")

        predict_btn = gr.Button("Predict Crop")
        crop_image_output = gr.Image(label="🌿 Crop Image")
        crop_text_output = gr.Markdown()

        predict_btn.click(fn=predict_crop,
                          inputs=[nitrogen, phosphorus, potassium, temperature, humidity, soil_pH, rainfall],
                          outputs=[crop_image_output, crop_text_output])

demo.launch()