solution

.ipynb_checkpoints/Untitled-checkpoint.ipynb

@@ -0,0 +1,6 @@
+{
+ "cells": [],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 4
+}

program.py

@@ -0,0 +1,38 @@
+import pandas as pd
+import numpy as np
+import csv
+from sklearn.linear_model import LinearRegression
+from stop_words import get_stop_words
+from sklearn.feature_extraction.text import TfidfVectorizer
+ 
+train = pd.read_csv("train/train.tsv", names = ['start_date', 'end_date', 'title', 'sort_title', 'data'], sep = "\t")
+
+vectorizer = TfidfVectorizer(stop_words=get_stop_words('polish'))
+linear_reg = LinearRegression()
+
+date = (train['start_date'] + train['end_date']) / 2
+
+train_vec = vectorizer.fit_transform(train['data'])
+linear_reg.fit(train_vec, date)
+
+dev_0 = pd.read_csv("dev-0/in.tsv", error_bad_lines = False, header = None, sep = "\t", quoting=csv.QUOTE_NONE)
+pred_dev_0 = linear_reg.predict(vectorizer.transform(dev_0[0]))
+pd.DataFrame(pred_dev_0).to_csv('dev-0/out.tsv', sep = "\t", index = False, header = False)
+
+
+dev_1 = pd.read_csv("dev-1/in.tsv", error_bad_lines = False, header = None, sep = "\t", quoting=csv.QUOTE_NONE)
+pred_dev_1 = linear_reg.predict(vectorizer.transform(dev_1[0]))
+pd.DataFrame(pred_dev_1).to_csv('dev-1/out.tsv', sep = "\t", index = False, header = False)
+
+
+test = pd.read_csv("test-A/in.tsv", names = ['data'], sep = "\t")
+pred_test = linear_reg.predict(vectorizer.transform(test['data']))
+pd.DataFrame(pred_test).to_csv('test-A/out.tsv', sep = "\t", index = False, header = False)
+
+#./geval -t dev-0
+#21.8069
+#./geval -t dev-1
+#22.0247
+
+
+