# import library
import streamlit as st
import pandas as pd
import numpy as np
import pickle
import json

# Load Model
with open('model.pkl', 'rb') as file:
    model = pickle.load(file)
with open('feature.txt', 'r') as file:
    feature = json.load(file)

# Function to run streamlit model predictor
def run():

   # Set Title
    st.title("Predict the Price Range of a Mobile Based on its Features")
    st.markdown('---')

    # Create a Form for Data Inference
    st.markdown('## Input Data')
    with st.form('my_form'):
        battery_power = st.number_input('Battery Power', min_value=500, max_value=8000)
        blue = st.selectbox('Has bluetooth or not? 0 = No, Yes = 1', (0,1))
        clock_speed = st.number_input('Clock Speed (Speed at Wich Microprocessor Execute Instruction)', min_value=0.5, max_value=3.0)
        dual_sim = st.selectbox('Has dual sim or not? 0 = No, Yes = 1', (0,1))
        fc = st.number_input('Front Camera mega pixels', min_value=0, max_value=40)
        four_g = st.selectbox('Has 4G or not? 0 = No, Yes = 1', (0,1))
        int_memory = st.number_input('Internal Memory in Gigabytes', min_value=2, max_value=256)
        m_dep = st.number_input('Mobile Depth in cm', min_value=0.1, max_value=1.0)
        mobile_wt = st.number_input('Weight of Mobilephone', min_value=80, max_value=300)
        n_cores = st.number_input('Number of Cores of Processor', min_value=1, max_value=10)
        pc = st.number_input('Primary Camera mega pixels', min_value=0, max_value=20)
        px_height = st.number_input('Pixel Resolution Height', min_value=0, max_value=2000)
        px_width = st.number_input('Pixel Resolution Width', min_value=500, max_value=2000)
        ram = st.number_input('RAM in Megabytes', min_value=256, max_value=4000)
        sc_h = st.number_input('Screen Height of Mobile in cm', min_value=5, max_value=20)
        sc_w = st.number_input('Screen Width of Mobile in cm', min_value=1, max_value=15)
        talk_time = st.number_input('The Longest Time for one battery charge when you use it', min_value=2, max_value=168)
        three_g = st.selectbox('Has 3G or not? 0 = No, Yes = 1', (0,1))
        touch_screen = st.selectbox('Has touch_screen or not? 0 = No, Yes = 1', (0,1))
        wifi = st.selectbox('Has wifi or not? 0 = No, Yes = 1', (0,1))

        # Create a button to make predictions
        submitted = st.form_submit_button("Predict")

    # Dataframe
    data = {'battery_power': battery_power,
            'blue': blue,
            'clock_speed': clock_speed,
            'dual_sim': dual_sim,
            'fc': fc, 
            'four_g': four_g,
            'int_memory': int_memory, 
            'm_dep': m_dep,
            'mobile_wt': mobile_wt,
            'n_cores': n_cores,
            'pc': pc,
            'px_height': px_height,
            'px_width': px_width,
            'ram': ram,
            'sc_h': sc_h,
            'sc_w': sc_w,
            'talk_time': talk_time,
            'three_g': three_g,
            'touch_screen': touch_screen,
            'wifi': wifi}

    df = pd.DataFrame([data])
    st.dataframe(df)

    if submitted:
        df_selected = df[feature]
        y_pred_inf = model.predict(df_selected)
        if y_pred_inf[0] == 0:
            st.subheader('~ The Mobile Features you Enter is in "Entry-Level" price ~')
        elif y_pred_inf[0] == 1:
            st.write('~ The Mobile Features you Enter is in "Mid-Range" price ~')
        elif y_pred_inf[0] == 2:
            st.write('~ The Mobile Features you Enter is in "High-End" price ~')
        elif y_pred_inf[0] == 3:
            st.write('~ The Mobile Features you Enter is in "Flagship" price ~')

if __name__== '__main__':
    run()