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


from sklearn.linear_model import LinearRegression
import pandas as pd
import string
import csv



price = []
mileage = []
year = []
brand = []
enginetype = []
enginecapacity = []
with open("train/train.tsv") as tsv:
    for line in csv.reader(tsv, delimiter="\t"):
        price.append(line[0])
        mileage.append(line[1])
        year.append(line[2])
        brand.append(line[3])
        enginetype.append(line[4])
        enginecapacity.append(line[5])   
        
cena = pd.DataFrame(list(zip(price)))
reszta = pd.DataFrame(list(zip(mileage,year,enginecapacity)))
model = LinearRegression().fit(reszta,cena)

      
out1 = open("dev-0/out.tsv", "w")
mileage_dev = []
year_dev = []
brand_dev = []
enginetype_dev = []
enginecapacity_dev = []
with open("dev-0/in.tsv") as tsv:
    for line in csv.reader(tsv, delimiter="\t"):
        mileage_dev.append(line[0])
        year_dev.append(line[1])
        brand_dev.append(line[2])
        enginetype_dev.append(line[3])
        enginecapacity_dev.append(line[4])   
        
do_predykcji1 = pd.DataFrame(list(zip(mileage_dev,year_dev,enginecapacity_dev)))   
predict1 = model.predict(do_predykcji1)

for x in predict1:
    new1 = str(x).replace('[','')
    new1 = str(new1).replace(']','')
    out1.write(str(new1) + '\n')
out1.close()



out2 = open("test-A/out.tsv", "w")
mileage_test = []
year_test = []
brand_test = []
enginetype_test = []
enginecapacity_test = []
with open("test-A/in.tsv") as tsv:
    for line in csv.reader(tsv, delimiter="\t"):
        mileage_test.append(line[0])
        year_test.append(line[1])
        brand_test.append(line[2])
        enginetype_test.append(line[3])
        enginecapacity_test.append(line[4])   

do_predykcji2 = pd.DataFrame(list(zip(mileage_test,year_test,enginecapacity_test)))   
predict2 = model.predict(do_predykcji2)  

for y in predict2:
    new2 = str(y).replace('[','')
    new2 = str(new2).replace(']','')
    out2.write(str(new2) + '\n')
out2.close()