from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LinearRegression
from stop_words import get_stop_words
import pandas as pd
import numpy as np
import csv


lm_model = LinearRegression()
tfidvectorizer = TfidfVectorizer(stop_words=get_stop_words('polish'))
    
    
train_nm = ['start_date', 'end_date', 'title', 'sort_title', 'data']
train_nm_test = ['data']

dataset = []
processed = []
new_text = ""
    
train_file = pd.read_csv('train/train.tsv', sep="\t", names=train_nm)
print('DONE20!')
date = (train_file['start_date'] + train_file['end_date']) / 2
print('DONE22!')
vectorizer= tfidvectorizer.fit_transform(train_file['data'])
print('DONE24!')
lm_model.fit(vectorizer, date)
print('DONE26!')
   
dev_0 = pd.read_csv("dev-0/in.tsv", error_bad_lines = False, header = None, sep = "\t", quoting=csv.QUOTE_NONE)
dev_1 = pd.read_csv("dev-1/in.tsv", error_bad_lines = False, header = None, sep = "\t", quoting=csv.QUOTE_NONE,)
test = pd.read_csv("test-A/in.tsv", names = train_nm, sep = "\t")
print('DONE31!')


test_file= tfidvectorizer.transform(test['data'])
test_file_predict = lm_model.predict(test_file)
with open('test-A/out.tsv', 'w') as file:
    for i in test_file_predict:
        file.write("%f\n" % i)

print('DONE38!')