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

# In[59]:


import pandas as pd
from sklearn.linear_model import LinearRegression
from sklearn.feature_extraction.text import CountVectorizer,TfidfVectorizer
from sklearn.metrics import mean_squared_error
from sklearn.pipeline import make_pipeline


# In[60]:


colnames = ['start', 'text']
data = pd.read_csv('train/train.tsv', sep='\t', names=colnames, usecols=[0, 4])


# In[66]:


x_train = data['text']
y_train = data['start']


# In[67]:


tfidf_vectorizer=TfidfVectorizer(use_idf=True, max_df=0.95)
tfidf_vectorizer.fit_transform(x_train.values)
x_train_prepared = tfidf_vectorizer.transform(x_train.values)


# In[68]:


lr = LinearRegression()
model = lr.fit(x_train_prepared, y_train)


# In[69]:


y_dev0_exp = pd.read_csv('dev-0/expected.tsv', sep='\t', names=['text'])

f = open("dev-0/in.tsv", "r", encoding='utf-8')
lines_dev_0 = f.readlines()
x_dev0 = pd.DataFrame(lines_dev_0)
x_dev0.rename(columns = {0 : 'text'}, inplace = True)

x_dev0_prepared = tfidf_vectorizer.transform(x_dev0['text'].values)
y_dev0_pred = model.predict(x_dev0_prepared)

file = open('dev-0/out.tsv', 'w')
for y in y_dev0_pred:
    file.write(f'{y}\n')
file.close()


# In[74]:


y_dev1_exp = pd.read_csv('dev-1/expected.tsv', sep='\t', names=['text'])

f = open("dev-1/in.tsv", "r", encoding='utf-8')
lines_dev_1 = f.readlines()
x_dev1 = pd.DataFrame(lines_dev_1)
x_dev1.rename(columns = {0 : 'text'}, inplace = True)

x_dev1_prepared = tfidf_vectorizer.transform(x_dev1['text'].values)
y_dev1_pred = model.predict(x_dev1_prepared)

file = open('dev-1/out.tsv', 'w')
for y in y_dev1_pred:
    file.write(f'{y}\n')
file.close()


# In[76]:


f = open("test-A/in.tsv", "r", encoding='utf-8')
lines_test = f.readlines()
x_test = pd.DataFrame(lines_test)
x_test.rename(columns = {0 : 'text'}, inplace = True)

x_test_prepared = tfidf_vectorizer.transform(x_test['text'].values)
y_test_pred = model.predict(x_test_prepared)

file = open('test-A/out.tsv', 'w')
for y in y_test_pred:
    file.write(f'{y}\n')
file.close()