import streamlit as st
import pandas as pd
import pickle


# Title
st.write(""" 
# DEATH PREDICTION APP

This app predicts the likelihood of a patient's **Death** .

""")

@st.cache_data
def fetch_data():
    df = pd.read_csv('h8dsft_P1G3_AlselloDM.csv')
    return df

df = fetch_data()

st.sidebar.title("Patient's Medical Records")
st.sidebar.write('Please enter the condition here')
st.sidebar.title("Prediction Form📋")

def user_input_features():
    # Age of the patient
    age = st.sidebar.number_input("Age of the patient", 1,100)
    # Decrease of red blood cells or hemoglobin
    anaemia = st.sidebar.selectbox('Have Anaemia? (1 : True | 0 : False)',(1,0))
    # Level of the CPK enzyme in the blood
    creatinine_phosphokinase = st.sidebar.slider('How much is the Creatinine_Phosphokinase(CP) in the body?',20,8000,3000)
    # If the patient has diabetes
    diabetes = st.sidebar.selectbox('If the patient has diabetes (1 : True | 0 : False)',(1,0))
    # Heart capability
    ejection_fraction = st.sidebar.slider('What is the Ejection Fraction?',0,150,75)
    # If the patient has hypertension
    high_blood_pressure = st.sidebar.selectbox('If the patient has hypertension (1 : True | 0 : False)',(1,0))
    # Platelets in the blood
    platelets = st.sidebar.number_input("What is the Blood Platelets count? (kiloplatelets/mL)",0.0,100000.0)
    # Level of creatinine in the blood
    serum_creatinine = st.sidebar.slider('What is the level of serum creatinine in the bloodstream?',0.5,10.0,0.5)
    # Level of sodium in the blood
    serum_sodium = st.sidebar.slider('What is the level of Serum_Sodium in the Body?',50,200,50)
    # Male or Female
    sex = st.sidebar.selectbox('Gender',('Male','Female'))
    # If the patient smokes
    smoking = st.sidebar.selectbox('If the patient smokes (1 : True | 0 : False)',(1,0))
    # Follow-up period
    time = st.sidebar.slider('Follow-up period',0,400,20)
    data = {'age': age,'anaemia':anaemia,'creatinine_phosphokinase':creatinine_phosphokinase,
            'diabetes':diabetes,'ejection_fraction':ejection_fraction,
            'high_blood_pressure':high_blood_pressure,'platelets':platelets,
            'serum_creatinine':serum_creatinine,'serum_sodium':serum_sodium,
            'sex':sex,'smoking':smoking,'time':time}
    # 
    features = pd.DataFrame(data,index=[0])
    return features

input_df = user_input_features()

st.subheader('Medical Data📋')
st.write(input_df)

with open('death_pred.pkl', 'rb') as file:
    load_model = pickle.load(file)

# Access the individual models
model_xgboost = load_model['model_xgboost']

if st.button('Predict'):
    try:
        prediction = model_xgboost.predict(input_df)
        prediction_proba = model_xgboost.predict_proba(input_df)

        st.subheader('DIAGNOSIS')
        for i in range(len(prediction)):
            if prediction[i] == 0:
                st.success("The patient is in good condition")
            else:
                st.error("This Patient is at **Risk of Death**")

        st.subheader('Probability')
        st.write("0 : No | 1 : Yes")
        st.write(prediction_proba)

    except ValueError:
        st.header("Insufficient Medical Data")