import streamlit as st
import tensorflow as tf
from PIL import Image
import numpy as np
import os
import pandas as pd
import io




def import_and_predict(image_data, model, class_labels):
    size = (256, 256)

    if image_data is not None:
        image = Image.open(io.BytesIO(image_data.read()))
        image = image.resize(size)
        image = np.array(image)
        img_reshape = image / 255.0
        img_reshape = np.expand_dims(img_reshape, axis=0)

        prediction = model.predict(img_reshape)
        st.image(image, width=300)
        predictions_label = class_labels[np.argmax(prediction[0])]
        return predictions_label
    else:
        st.warning("Please upload an image.")
        return None




st.title("Plant Disease Detection")

selected_item = st.radio("Select an item:", ["Apple", "Mango", "Grape", "Tomato"])

if selected_item:
    st.write(f"You selected:  {selected_item}")

    uploaded_image = st.file_uploader(f"Upload an image of {selected_item.lower()}", type=["jpg", "jpeg", "png"])

    df = pd.read_excel('./final format.xlsx')

    models_path = [
        './models/best_model_50_apple_plant (1).h5','./models/best_model_100_subset (1).h5','./models/best_model_50_grape_plant2.h5','./models/model_inception_epoch_50_mango.h5']
    
    @st.cache_resource
    def load_model(models_path):
        model = tf.keras.models.load_model(models_path)
        return model

    if selected_item == 'Apple':
        CLASS_LABELS =  ['apple scab', 'Apple black rot', 'cedar apple rust', 'Apple healthy']

        model =load_model(models_path[0])
        prediction = import_and_predict(uploaded_image, model, CLASS_LABELS)
        st.write("disease name: ", prediction)


        if prediction != None:
            new_title = '<p style="font-size: 38px">Measures you can take to control: </p>'
            st.markdown(new_title, unsafe_allow_html=True)
            if prediction == CLASS_LABELS[0]:
                response = df['Measures'].dropna().head(28)
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[1]:
                response = df['Measures'].dropna()[28:46]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[2]:
                response = df['Measures'].dropna()[46:63]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[3]:
                st.write("Plant is healthy take good care of it")





    elif selected_item == 'Tomato':
                
        CLASS_LABELS = ['Tomato Early blight', 'Tomato Leaf Mold', 'Tomato YellowLeaf Curl Virus',
                                 'Tomato mosaic virus', 'Tomato healthy']

        model =load_model(models_path[1])
   
        prediction = import_and_predict(uploaded_image, model, CLASS_LABELS)
        st.write("disease name: ", prediction)


        if prediction != None:
            new_title = '<p style="font-size: 38px">Measures you can take to control: </p>'
            st.markdown(new_title, unsafe_allow_html=True)
            if prediction == CLASS_LABELS[0]:
                response = df['Measures'].dropna()[63:79]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[1]:                                                             # remaining
                response = df['Measures'].dropna()[187:196]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[2]:
                response = df['Measures'].dropna()[99:113]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[3]:
                response = df['Measures'].dropna()[79:99]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[4]:
                st.write("Plant is healthy take good care of it")           

    elif selected_item == 'Grape':

        CLASS_LABELS = ['Grape Black rot', 'Grape Black Measles','Grape Leaf blight', 'Grape healthy']

        model =load_model(models_path[2])
        prediction = import_and_predict(uploaded_image, model, CLASS_LABELS)
        st.write("disease name: ", prediction)

        if prediction != None:
            new_title = '<p style="font-size: 38px">Measures you can take to control: </p>'
            st.markdown(new_title, unsafe_allow_html=True)
            if prediction == CLASS_LABELS[0]:
                response = df['Measures'].dropna()[123:134]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[1]:                                                             
                response = df['Measures'].dropna()[196:204]                                                    #Remaining
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[2]:
                response = df['Measures'].dropna()[113:123]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)


            elif prediction == CLASS_LABELS[3]:
                st.write("Plant is healthy take good care of it") 


    elif selected_item == 'Mango':

        CLASS_LABELS = ['Anthracnose', 'Bacterial Canker', 'Cutting Weevil', 'Die Back', 'Gall Midge', 'Healthy','Sooty Mould' ,'Powdery Mildew']

        model =load_model(models_path[3])
        prediction = import_and_predict(uploaded_image, model, CLASS_LABELS)
        st.write("disease name: ", prediction)
        
        if prediction != None:
            new_title = '<p style="font-size: 38px">Measures you can take to control: </p>'
            st.markdown(new_title, unsafe_allow_html=True)

            if prediction == CLASS_LABELS[0]:
                response = df['Measures'].dropna()[134:151]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[1]:                                                             
                response = df['Measures'].dropna()[151:165]                                                    
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[2]:
                response = df['Measures'].dropna()[164:173]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])   #remaining 2,3,4
                st.write(response_text, unsafe_allow_html=True) 

            elif prediction == CLASS_LABELS[3]:
                response = df['Measures'].dropna()[172:180]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[4]:                                                             
                response = df['Measures'].dropna()[181:189]                                                    
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[5]:
                st.write("Plant is healthy take good care of it")

            elif prediction == CLASS_LABELS[6]:
                response = df['Measures'].dropna()[164:174]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)

            elif prediction == CLASS_LABELS[7]:
                response = df['Measures'].dropna()[174:189]
                response_text = " ".join([f"<p style='font-size: 18px;'>{text}</p>" for text in response])
                st.write(response_text, unsafe_allow_html=True)