import streamlit as st
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import plotly.express as px
from PIL import Image

st.set_page_config(
    page_title="Heart Failure Classification",
    layout="wide",
    initial_sidebar_state="expanded",
)

# dataset
dataset = "https://raw.githubusercontent.com/ediashta/p1-ftds-020-rmt-g3-ediashta/main/h8dsft_P1G3_Ediashta.csv"
data = pd.read_csv(dataset)

cat_col = ["anaemia", "diabetes", "high_blood_pressure", "sex", "smoking"]
num_col = [
    "age",
    "creatinine_phosphokinase",
    "ejection_fraction",
    "platelets",
    "serum_creatinine",
    "serum_sodium",
    "time",
]

data_ord = data[cat_col]
data_num = data[num_col]


def dataset():
    # distribution plot
    st.title("Heart Failure Classification")
    st.subheader("Dataset")
    st.dataframe(data)

    # column description
    st.subheader("Column Description")

    col1, col2 = st.columns(2)
    col1.markdown("* **age** : Umur ")
    col1.markdown("* **anaemia** : Penurunan Sel Darah Merah (**boolean**)")
    col1.markdown(
        "* **creatinine_phosphokinase** : Tingkat Enzim CPK pada darah (**mcg/L**)"
    )
    col1.markdown("* **diabetes** : Status diabetes (**boolean**)")
    col1.markdown(
        "* **ejection_fraction** : Persentase darah keluar dari jantung pada tiap kontraksi (**percentage**)"
    )
    col1.markdown("* **high_blood_pressure** : Status Hipertensi (**boolean**)")
    col1.markdown("* **platelets** : Platelets pada darah (**kiloplatelets/mL**) ")
    col2.markdown(
        "* **serum_creatinine** : Tingkat serum kreatinin pada darah (**mg/dL**) "
    )
    col2.markdown("* **serum_sodium** : Tingkat serum sodium pada darah (**mEq/L**)")
    col2.markdown("* **sex** : Jenis Kelamin (**binary**)")
    col2.markdown("* **smoking** : Status merokok (**boolean**)")
    col2.markdown("* **time** : Periode follow-up (**days**)")
    col2.markdown("* **DEATH_event** : Status kematian saat follow-up (**boolean**)")


def distribution():
    # distribution plot
    st.title("Heart Failure Classification")
    st.subheader("Distribution")

    col1, col2 = st.columns(2)

    # Death Event Bar Plot
    fig = plt.figure(figsize=(10, 6))
    col1.write("Death Event Barplot")
    sns.countplot(data=data, x="DEATH_EVENT", palette="coolwarm")
    col1.pyplot(fig)

    # Death Event Bar Plot
    fig = plt.figure(figsize=(10, 6))
    col2.write("Age Distribution")
    sns.histplot(
        data=data, x="age", kde=True, bins=30, hue="DEATH_EVENT", palette="coolwarm"
    )
    col2.pyplot(fig)


def boxplot():
    # distribution plot
    st.title("Heart Failure Classification")
    st.subheader("Boxplot")

    fig, axs = plt.subplots(2, 4, figsize=(20, 10))

    for i, x in enumerate(data_num):
        if i < 4:
            axs[0, i].set_title(x)
            axs[0, i].boxplot(data[x])
        elif 7 >= i >= 4:
            axs[1, i - 4].set_title(x)
            axs[1, i - 4].boxplot(data[x])
            axs[1, 3].axis("off")

    fig.tight_layout()

    st.pyplot(fig)


def corr_matrix():
    # distribution plot
    st.title("Heart Failure Classification")
    st.subheader("Correlation")


if __name__ == "__main__":
    run()