import streamlit as st
import pandas as pd
import numpy as np
import pickle
import base64
import seaborn as sns
import matplotlib.pyplot as plt

st.write("""
# Breast Cancer Prediction App


Kanker payudara adalah kanker paling umum di kalangan wanita di dunia. Ini menyumbang 25% dari semua kasus kanker, dan mempengaruhi lebih dari 2,1 Juta orang pada tahun 2015 saja. Ini dimulai ketika sel-sel di payudara mulai tumbuh di luar kendali. Sel-sel ini biasanya membentuk tumor yang dapat dilihat melalui sinar-X atau dirasakan sebagai benjolan di area payudara.

Tantangan utama untuk mendeteksinya adalah bagaimana mengklasifikasikan tumor menjadi ganas (kanker) atau jinak (non kanker).
""")

url_dataset = f'<a href="breast_cancer.csv">Download Dataset CSV File</a>'
st.markdown(url_dataset, unsafe_allow_html=True)

def user_input_features() :
    radius_mean = st.sidebar.slider('radius_mean', 6.981,28.11)
    texture_mean = st.sidebar.slider('texture_mean', 9.71,39.28)
    perimeter_mean = st.sidebar.slider('perimeter_mean', 43.79,188.5)
    area_mean = st.sidebar.slider('area_mean', 143.5,2501.0)
    smoothness_mean = st.sidebar.slider('smoothness_mean', 0.05263,0.1634)
    compactness_mean = st.sidebar.slider('compactness_mean', 0.01938,0.3454)
    concavity_mean = st.sidebar.slider('concavity_mean', 0.0,4268.0)
    concave_points_mean = st.sidebar.slider('concave_points_mean', 0.0,0.2012)
    symmetry_mean = st.sidebar.slider('symmetry_mean', 0.106,0.304)
    fractal_dimension_mean = st.sidebar.slider('fractal_dimension_mean', 0.04996,0.09744)
    radius_se = st.sidebar.slider('radius_se', 0.1115,2.873)
    texture_se = st.sidebar.slider('texture_se', 0.3602,4.885)
    perimeter_se = st.sidebar.slider('perimeter_se', 0.757,21.98)
    area_se = st.sidebar.slider('area_se', 6.802,542.2)
    smoothness_se = st.sidebar.slider('smoothness_se', 0.001713,0.03113)
    compactness_se = st.sidebar.slider('compactness_se', 0.002252,0.1354)
    concavity_se = st.sidebar.slider('concavity_se', 0.0,0.396)
    concave_points_se = st.sidebar.slider('concave_points_se', 0.0,0.05279)
    symmetry_se = st.sidebar.slider('symmetry_se', 0.007882,0.07895)
    fractal_dimension_se = st.sidebar.slider('fractal_dimension_se', 0.000895,0.02984)
    radius_worst = st.sidebar.slider('radius_worst', 7.93,36.04)
    texture_worst = st.sidebar.slider('texture_worst', 12.02,49.54)
    radius_worst = st.sidebar.slider('radius_worst', 0.1115,2.873)
    perimeter_worst = st.sidebar.slider('perimeter_worst', 50.41,251.2)
    area_worst = st.sidebar.slider('area_worst', 185.2,4254.0)
    smoothness_worst = st.sidebar.slider('smoothness_worst', 0.07117,0.2226)
    compactness_worst = st.sidebar.slider('compactness_worst', 0.02729,1.058)
    concavity_worst = st.sidebar.slider('concavity_worst', 0.0,1.252)
    concave_points_worst = st.sidebar.slider('concave_points_worst', 0.0,0.291)
    symmetry_worst = st.sidebar.slider('symmetry_worst', 0.1565,0.6638)
    fractal_dimension_worst = st.sidebar.slider('fractal_dimension_worst', 0.1565,0.2075)

    
    data = {'radius_mean':[radius_mean],
    'texture_mean':[texture_mean],
    'radius_mean':[radius_mean],
    'perimeter_mean':[perimeter_mean],
    'radius_mean':[radius_mean],
    'area_mean':[area_mean],
    'smoothness_mean':[smoothness_mean],
    'compactness_mean':[compactness_mean],
    'concavity_mean':[concavity_mean],
    'concave_points_mean':[concave_points_mean],
    'symmetry_mean':[symmetry_mean],
    'fractal_dimension_mean':[fractal_dimension_mean],
    'radius_se':[radius_se],
    'texture_se':[texture_se],
    'perimeter_se':[perimeter_se],
    'area_se':[area_se],
    'smoothness_se':[smoothness_se],
    'compactness_se':[compactness_se],
    'concavity_se':[concavity_se],
    'concave_points_se':[concave_points_se],
    'symmetry_se':[symmetry_se],
    'fractal_dimension_se':[fractal_dimension_se],
    'fractal_dimension_se':[fractal_dimension_se],
    'radius_worst':[radius_worst],
    'texture_worst':[texture_worst],
    'perimeter_worst':[perimeter_worst],
    'area_worst':[area_worst],
    'smoothness_worst':[smoothness_worst],
    'compactness_worst':[compactness_worst],
    'concavity_worst':[concavity_worst],
    'concave_points_worst':[concave_points_worst],
    'symmetry_worst':[symmetry_worst],
    'fractal_dimension_worst':[fractal_dimension_worst]
    }

    features = pd.DataFrame(data)
    return features

input_df = user_input_features()


breast_raw = pd.read_csv('breast_cancer.csv')
breast_raw.fillna(0, inplace=True)
breast = breast_raw.drop(columns=['diagnosis'])
df = pd.concat([input_df,breast],axis=0)

df = df[:1] # Selects only the first row (the user input data)
df.fillna(0, inplace=True)

features = ['radius_mean', 'texture_mean', 'perimeter_mean',
       'area_mean', 'smoothness_mean', 'compactness_mean', 'concavity_mean',
       'concave points_mean', 'symmetry_mean', 'fractal_dimension_mean',
       'radius_se', 'texture_se', 'perimeter_se', 'area_se', 'smoothness_se',
       'compactness_se', 'concavity_se', 'concave points_se', 'symmetry_se',
       'fractal_dimension_se', 'radius_worst', 'texture_worst',
       'perimeter_worst', 'area_worst', 'smoothness_worst',
       'compactness_worst', 'concavity_worst', 'concave points_worst',
       'symmetry_worst', 'fractal_dimension_worst']

df = df[features]


st.subheader('User Input features')
st.write(df)

load_clf = pickle.load(open('breast_clf.pkl', 'rb'))
prediction = load_clf.predict(df)
prediction_proba = load_clf.predict_proba(df)
breast_labels = np.array(['Benign','Malignant'])
st.subheader('Detection')
if(prediction == 'B') :
    st.write(breast_labels[0])
else :
    st.write(breast_labels[1])
st.subheader('Detection Probability')
df_proba = pd.DataFrame(data=prediction_proba, index=['Probability'], columns=breast_labels)
st.write(df_proba)