#!/usr/bin/env python
# coding: utf-8

# # Car prediction model training and serving with streamlit

#import libraries
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import r2_score,mean_squared_error
from sklearn.pipeline import Pipeline
from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import StandardScaler,OneHotEncoder



#Load data
df=pd.read_csv('cars.csv')

X=df.drop('Price',axis=1)
y=df[['Price']]



X_train,X_test,y_train,y_test=train_test_split(X,y,
                                               test_size=0.2,
                                               random_state=42)




preproccer=ColumnTransformer(transformers=[('num',StandardScaler(),
                                           ['Mileage','Cylinder','Liter','Doors']),
                            ('cat',OneHotEncoder(),['Make','Model','Trim','Type'])])




model=LinearRegression()
pipe=Pipeline(steps=[('preprocessor',preproccer),
                    ('model',model)])

## Train with all data

pipe.fit(X,y)
#y_pred=pipe.predict(X_test)
#mean_squared_error(y_test,y_pred)**0.5,r2_score(y_test,y_pred)

import streamlit as st
def price(make,model,trim,mileage,car_type,cylinder,liter,doors,cruise,sound,leather):
	input_data=pd.DataFrame({
		'Make':[make],
		'Model':[model],
		'Trim':[trim],
		'Mileage':[mileage],
		'Type':[car_type],
		'Car_type':[car_type],
		'Cylinder':[cylinder],
		'Liter':[liter],
		'Doors':[doors],
		'Cruise':[cruise],
		'Sound':[sound],
		'Leather':[leather]
		})
	prediction=pipe.predict(input_data)[0]
	return prediction

st.title("Car Price Prediction :racing_car: \n by @sbgonenc")
st.write("Select Car Specs")
make=st.selectbox("Brand",df['Make'].unique())
model=st.selectbox("Model",df[df['Make']==make]['Model'].unique())
trim=st.selectbox("Trim",df[(df['Make']==make) & (df['Model']==model)]['Trim'].unique())
mileage=st.number_input("Milage",1 ,60000)
car_type=st.selectbox("Type",df[(df['Make']==make) & (df['Model']==model) & (df['Trim']==trim )]['Type'].unique())
cylinder=st.selectbox("Cylinders",df['Cylinder'].unique())
liter=st.number_input("Liter",1,6)
doors=st.selectbox("Doors",df['Doors'].unique())
cruise=st.radio("Cruise",[True,False])
sound=st.radio("Sound System",[True,False])
leather=st.radio("Leather",[True,False])
if st.button("Prediction"):
	pred=price(make,model,trim,mileage,car_type,cylinder,liter,doors,cruise,sound,leather)

	st.write("Predicted Price :oncoming_automobile:  $",round(pred[0],2))