miayulia/Lendingkart-FinalProject

1import streamlit as st2import pandas as pd3import numpy  as np4import joblib5import pickle6from sklearn.cluster import KMeans7# import streamlit_extras8# from streamlit_extras.colored_header import colored_header9 10st.image('lendingkart_owler_20190913_181413_original.jpg')11st.title('Lendingkart')12st.subheader('Think Cash, Think Lendingkart!')13# colored_header(14#     label="Lendingkart",15#     description='Think Cash, Think Lendingkart!',16#     color_name="orange-70",17# )18@st.cache_data19def fetch_data():20    df = pd.read_csv('loan_approval_dataset.csv')21    return df22 23df = fetch_data()24df.columns = df.columns.str.replace(' ', '')25 26classification, clustering = st.tabs(['Loan Approval', 'Customers Cluster'])27with classification :28 29    st.write('Predicting customers loan approval based criterias below:')30 31    cibil_score =st.number_input('cibil_score', value=0)32    loan_term = st.select_slider('loan_term', [2,4,6,8,10,12,14,16,18,20])33 34    data = {35        'cibil_score': cibil_score,36        'loan_term': loan_term,37    }38    input = pd.DataFrame(data, index=[0])39    st.subheader('Summary :')40    st.write(input)41 42    # load files43    classification = joblib.load("classification.pkl")44 45    if st.button('Loan Approval'):46        prediction = classification.predict(input)47 48        if prediction == 0:49            prediction = 'Approved'50            st.write('Approved')51            st.success('This Customer is Approved! Please input the proposal documents below.')52            st.balloons()53            st.file_uploader('Upload Here')54        else:55            prediction = 'Rejected'56            st.write('Rejected')57            st.warning('Sorry the customer is rejected. Offer them with another program.')58 59# Create a unique key for each widget60widget_keys = {61    'income_annum': 'widget_income_annum',62    'loan_amount': 'widget_loan_amount',63    'cibil_score': 'widget_cibil_score',64    'residential_assets_value': 'widget_residential_assets_value',65    'commercial_assets_value': 'widget_commercial_assets_value',66    'luxury_assets_value': 'widget_luxury_assets_value',67    'bank_asset_value': 'widget_bank_asset_value'68}69 70with clustering:    71    st.write('Predicting customers class based criterias below:')72 73    income_annum = st.number_input('income_annum', value=0, key=widget_keys['income_annum'])74    loan_amount = st.number_input('loan_amount', value=0, key=widget_keys['loan_amount'])75    cibil_score = st.number_input('cibil_score', value=0, key=widget_keys['cibil_score'])76    residential_assets_value = st.number_input('residential_assets_value', value=0, key=widget_keys['residential_assets_value'])77    commercial_assets_value = st.number_input('commercial_assets_value', value=0, key=widget_keys['commercial_assets_value'])78    luxury_assets_value = st.number_input('luxury_assets_value', value=0, key=widget_keys['luxury_assets_value'])79    bank_asset_value = st.number_input('bank_asset_value', value=0, key=widget_keys['bank_asset_value'])80 81    data = {82        'income_annum': income_annum,83        'loan_amount': loan_amount,84        'cibil_score': cibil_score,85        'residential_assets_value': residential_assets_value,86        'commercial_assets_value': commercial_assets_value,87        'luxury_assets_value': luxury_assets_value,88        'bank_asset_value': bank_asset_value89    }90    input1 = pd.DataFrame(data, index=[0])91 92    load_model1 = joblib.load("clustering.pkl")93    if st.button('Cluster'):94        prediction1 = load_model1.predict(input1)95 96        if prediction1 == 0:97            prediction = 'Class B'98        else:99            prediction = 'Class B'100 101        st.write('This customer is from:')102        st.write(prediction)103