import streamlit as st
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import joblib

from sklearn.ensemble import RandomForestClassifier, VotingClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC
from sklearn.naive_bayes import GaussianNB
from sklearn.neural_network import MLPClassifier
from sklearn.ensemble import GradientBoostingClassifier
from xgboost import XGBClassifier
from lightgbm import LGBMClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score


st.title('Disease Prediction Application')
st.write('''
         Please fill in the attributes below, then hit the Predict button
         to get your results. 
         ''')

st.header('General Information')
gen = st.radio("Your Gender", ('Male', 'Female'))
st.write(''' ''')
age = st.slider('Your Age (Years)', min_value=0.0, max_value=100.0, value=50.0, step=1.0)
st.write(''' ''')
bmi = st.slider('BMI', min_value=0.0, max_value=50.0, value=25.0, step=0.1)
st.write(''' ''')


st.header('Blood Test Information')
bs = st.radio("Is Your Fasting Blood Sugar > 120 mg/dl?", ('Yes', 'No'))
st.write(''' ''')
urea = st.slider('Urea', min_value=0.0, max_value=100.0, value=50.0, step=0.1)
st.write(''' ''')
cr = st.slider('Creatinine Ratio(Cr)', min_value=0.0, max_value=1000.0, value=500.0, step=1.0)
st.write(''' ''')
hb = st.slider('HbA1c', min_value=0.0, max_value=20.0, value=10.0, step=0.1)
st.write(''' ''')

st.header('Cholesterol Test Information')
chol = st.slider('Cholesterol (Chol)', min_value=0.0, max_value=20.0, value=10.0, step=0.1)
st.write(''' ''')
tg = st.slider('Triglycerides(TG) Cholesterol', min_value=0.0, max_value=20.0, value=10.0, step=0.1)
st.write(''' ''')
hdl = st.slider('HDL Cholesterol', min_value=0.0, max_value=20.0, value=10.0, step=0.1)
st.write(''' ''')
ldl = st.slider('LDL Cholesterol', min_value=0.0, max_value=20.0, value=10.0, step=0.1)
st.write(''' ''')
vldl = st.slider('VLDL Cholesterol', min_value=0.0, max_value=50.0, value=25.0, step=0.1)
st.write(''' ''')

st.header('Cardio Test Information')
cp = st.radio("Chest Pain", ('Typical Angina', 'Atypical Angina', 'Non-Anginal Pain', 'Asymptomatic'))
st.write(''' ''')
max_heart = st.slider('Maximum Heart Rate', min_value=0.0, max_value=300.0, value=150.0, step=1.0)
st.write(''' ''')
resting_bp = st.slider('Resting Blood Pressure (In mm Hg)', min_value=0.0, max_value=200.0, value=100.0, step=1.0)
st.write(''' ''')
re = st.radio("Resting Electrocardiogram Results", ('Normal', 'ST-T Wave Abnormality (T Wave Inversions and/or ST Elevation or Depression of > 0.05 mV)', 'Showing Probable or Definite Left Ventricular Hypertrophy by Estes Criteria'))
st.write(''' ''')
ex = st.radio("Exercise Induced Angina", ('Yes', 'No'))
st.write(''' ''')
oldpeak = st.slider('ST Depression Induced by Exercise Relative to Rest', min_value=-5.0, max_value=5.0, value=0.0, step=0.01)
st.write(''' ''')
sp = st.radio("The Slope of the Peak Exercise ST Segment", ('Upsloping', 'Flat', 'Downsloping'))
st.write(''' ''')

# gender conversion
if gen == "Male":
    gender = 1
else:
    gender = 0

# Chest Pain
if cp == "Typical Angina":
    chest = 1
elif cp == "Atypical Angina":
    chest = 2
elif cp == "Non-Anginal Pain":
    chest = 3
else:
    chest = 4

# blood_sugar conversion
if bs == "Yes":
    blood_sugar = 1
else:
    blood_sugar = 0

# electro conversion
if re == "Normal":
    electro = 0
elif re == "ST-T Wave Abnormality (T Wave Inversions and/or ST Elevation or Depression of > 0.05 mV)":
    electro = 1
else:
    electro = 2

# exercise conversion
if ex == "Yes":
    exercise = 1
else:
    exercise = 0

# slope conversion
if sp == "Upsloping":
    slope = 1
elif sp == "Flat":
    slope = 2
else:
    slope = 3

user_input = np.array([age, gender, chest, blood_sugar, resting_bp, electro, max_heart,
                       exercise, oldpeak, slope, urea, cr, hb, chol, tg, hdl, ldl, vldl, bmi]).reshape(1, -1)


# import dataset
def get_dataset():
    data = pd.read_csv('Fianl Dataset.csv')

    # Calculate the correlation matrix
    # corr_matrix = data.corr()

    # Create a heatmap of the correlation matrix
    # plt.figure(figsize=(10, 8))
    # sns.heatmap(corr_matrix, annot=True, cmap='coolwarm')
    # plt.title('Correlation Matrix')
    # plt.xticks(rotation=45)
    # plt.yticks(rotation=0)
    # plt.tight_layout()

    # Display the heatmap in Streamlit
    # st.pyplot()

    return data
    

if st.button('Submit'):
    # Load your dataset for prediction
    df = get_dataset()

    # Split the dataset into features and targets for Heart and Diabetes prediction
    X = df.drop(['Diabetes', 'Heart'], axis=1)
    y_heart = df['Heart']
    y_diabetes = df['Diabetes']

    # Create an ensemble model for Heart prediction
    random_forest_model_heart = RandomForestClassifier(random_state=42)
    naive_bayes_model_heart = GaussianNB()
    gradient_boosting_model_heart = GradientBoostingClassifier(random_state=42)

    # Create a voting ensemble with soft voting for Heart prediction
    ensemble_model_heart = VotingClassifier(estimators=[
        ('Random Forest', random_forest_model_heart),
        ('Naive Bayes', naive_bayes_model_heart),
        ('Gradient Boosting', gradient_boosting_model_heart)
    ], voting='soft')

    # Create an ensemble model for Diabetes prediction
    random_forest_model_diabetes = RandomForestClassifier(random_state=42)
    naive_bayes_model_diabetes = GaussianNB()
    gradient_boosting_model_diabetes = GradientBoostingClassifier(random_state=42)

    # Create a voting ensemble with soft voting for Diabetes prediction
    ensemble_model_diabetes = VotingClassifier(estimators=[
        ('Random Forest', random_forest_model_diabetes),
        ('Naive Bayes', naive_bayes_model_diabetes),
        ('Gradient Boosting', gradient_boosting_model_diabetes)
    ], voting='soft')

    # Split the data into training and testing sets for both targets
    X_train, X_test, y_heart_train, y_heart_test, y_diabetes_train, y_diabetes_test = train_test_split(
        X, y_heart, y_diabetes, test_size=0.2, random_state=42
    )

    # Ensure the user input has the correct number of features for Heart prediction
    if user_input.shape[1] == X_train.shape[1]:
        # Fit the ensemble model for Heart prediction on the training data
        ensemble_model_heart.fit(X_train, y_heart_train)

        # Make predictions on user input for Heart
        prediction_heart = ensemble_model_heart.predict(user_input)
        prediction_proba_heart = ensemble_model_heart.predict_proba(user_input)

        # Check if the dimensions of prediction_heart match y_heart_test
        if prediction_heart.shape[0] == y_heart_test.shape[0]:
            # You can add a threshold and provide a prediction based on class 1 for Heart
            threshold_heart = 0.5
            if prediction_proba_heart[0][1] >= threshold_heart:
                st.header("Predicted Heart Disease: You might have Heart Disease")
                st.write("Predicted Probability of Having Heart Disease:", prediction_proba_heart[0][1] * 100)
            else:
                st.header("Predicted Heart Disease: You do not have Heart Disease")
                st.write("Predicted Probability of Having Heart Disease:", prediction_proba_heart[0][1] * 100)

            # Calculate and print Heart prediction metrics
            accuracy_heart = accuracy_score(y_heart_test, prediction_heart)
            precision_heart = precision_score(y_heart_test, prediction_heart)
            recall_heart = recall_score(y_heart_test, prediction_heart)
            f1_heart = f1_score(y_heart_test, prediction_heart)

            st.write("Heart Prediction Metrics:")
            st.write("Accuracy:", accuracy_heart)
            st.write("Precision:", precision_heart)
            st.write("Recall:", recall_heart)
            st.write("F1-score:", f1_heart)
            st.write("____________________________________________________________________________________________")
        else:
            st.write("Error: Inconsistent dimensions in Heart prediction. Please check your data.")
    else:
        st.write("Error: Input features do not match the dataset. Please provide valid input.")

    # Ensure the user input has the correct number of features for Diabetes prediction
    if user_input.shape[1] == X_train.shape[1]:
        # Fit the ensemble model for Diabetes prediction on the training data
        ensemble_model_diabetes.fit(X_train, y_diabetes_train)

        # Make predictions on user input for Diabetes
        pred_diabetes = ensemble_model_diabetes.predict(user_input)
        pred_diabetes_proba = ensemble_model_diabetes.predict_proba(user_input)

        # Check if the dimensions of pred_diabetes match y_diabetes_test
        if pred_diabetes.shape[0] == y_diabetes_test.shape[0]:
            # You can add a threshold and provide a prediction based on class 1 for Diabetes
            threshold_diabetes = 0.5
            if pred_diabetes_proba[0][1] >= threshold_diabetes:
                st.header("Predicted Diabetes: You might have Diabetes")
                st.write("Predicted Probability of Having Diabetes:", pred_diabetes_proba[0][1] * 100)
            else:
                st.header("Predicted Diabetes: You do not have Diabetes")
                st.write("Predicted Probability of Having Diabetes:", pred_diabetes_proba[0][1] * 100)

            # Calculate and print Diabetes prediction metrics
            accuracy_diabetes = accuracy_score(y_diabetes_test, pred_diabetes)
            precision_diabetes = precision_score(y_diabetes_test, pred_diabetes)
            recall_diabetes = recall_score(y_diabetes_test, pred_diabetes)
            f1_diabetes = f1_score(y_diabetes_test, pred_diabetes)

            st.write("Diabetes Prediction Metrics:")
            st.write("Accuracy:", accuracy_diabetes)
            st.write("Precision:", precision_diabetes)
            st.write("Recall:", recall_diabetes)
            st.write("F1-score:", f1_diabetes)
        else:
            st.write("Error: Inconsistent dimensions in Diabetes prediction. Please check your data.")
    else:
        st.write("Error: Input features do not match the dataset. Please provide valid input.")