{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "greenhouse-technician",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import sklearn\n",
    "import pandas as pd\n",
    "from gzip import open as open_gz\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "acoustic-dividend",
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_year(x, path_out, model):\n",
    "    \n",
    "    results = model.predict(x)\n",
    "\n",
    "    with open(path_out, 'wt') as file:\n",
    "        for r in results:\n",
    "            #if r%1==0:\n",
    "             #   r = r+0,5\n",
    "            file.write(str(r) + '\\n')\n",
    "            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "senior-harassment",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('train.tsv', 'r', encoding='utf8') as file:\n",
    "    train = pd.read_csv(file, sep='\\t', names=['Begin', 'End', 'Title', 'Publisher', 'Text'])\n",
    "    \n",
    "#with open('train/train.tsv', 'r', encoding='utf8') as file:\n",
    "#    train = pd.read_csv(file, sep='\\t', names=['Begin', 'End', 'Title', 'Publisher', 'Text'])\n",
    "train = train[0:2000]\n",
    "train_x=train['Text']    \n",
    "train_y=train['Begin']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "beneficial-traveler",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       nowią część kultury. U nas już nikt ich nie ch...\n",
       "1       hlstorja znana w okresie piramid, jak wlaśclcl...\n",
       "2       działek. Idąc dalej w swych hipotetycznych roz...\n",
       "3       w Warszawie o stosunkach domowych dziatwy szko...\n",
       "4       \\\\'iykład: \"Cywilizacyjna Koncepcja dziejów ¥e...\n",
       "                              ...                        \n",
       "1995    i' padną dobitne rozkazy. Od modlitwy nie poni...\n",
       "1996    WOJEWÖDZKI roz. 28 N'r. 7 III. IW slüsUlnlku d...\n",
       "1997    główną wagę kłaść należy na wspomniane w Bibli...\n",
       "1998    dniu 1'3 tym marca rozchwytywali broil z miejs...\n",
       "1999    ubezpieczenie społeczne, po rozpoznaniu na pos...\n",
       "Name: Text, Length: 2000, dtype: object"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "polyphonic-coach",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(memory=None,\n",
       "     steps=[('tfidfvectorizer', TfidfVectorizer(analyzer='word', binary=False, decode_error='strict',\n",
       "        dtype=<class 'numpy.float64'>, encoding='utf-8', input='content',\n",
       "        lowercase=True, max_df=1.0, max_features=None, min_df=1,\n",
       "        ngram_range=(1, 1), norm='l2', preprocessor=None, smooth...ression', LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None,\n",
       "         normalize=False))])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = make_pipeline(TfidfVectorizer(), LinearRegression())\n",
    "model.fit(train_x, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "varying-wright",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        Gazet, a tam o osobie zamformuie się. Uwiadomi...\n",
       "1        materiały, która wyniesie na rok w przybliżeni...\n",
       "2        były nawet w posiadaniu miejscowego polskiego ...\n",
       "3        Usuwanie nawarstwień... 105 powania nieudowodn...\n",
       "4        nie słyszał odC .S' źnniefAle'ObJ—A.\" \"hOdZI Ś...\n",
       "                               ...                        \n",
       "19993    wypoczęci! wzmocnieni, pełni najiepszych chęci...\n",
       "19994    ten krok draltyczny, ahby na tirazniejllzej je...\n",
       "19995    47 ust. 1 pkt 2 tej ustawy, obowiązkiem nałożo...\n",
       "19996    w, mmm ”w\" w „|. ..no-ń r. .I. Lennobrovi jako...\n",
       "19997    lat, przy czym za kolejny rok wnoszona jest do...\n",
       "Name: 0, Length: 19998, dtype: object"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with open('devin.tsv', 'r', encoding='utf8') as file:\n",
    "    x_dev = pd.read_csv(file, header=None, sep='\\t')\n",
    "x_dev = x_dev[0]   \n",
    "x_dev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "frozen-ticket",
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_year(x_dev, 'devout.tsv', model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "traditional-amount",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        1862.531349\n",
       "1        1962.791429\n",
       "2        1950.953131\n",
       "3        1965.496217\n",
       "4        1920.848072\n",
       "            ...     \n",
       "19993    1914.228771\n",
       "19994    1902.264257\n",
       "19995    2009.252595\n",
       "19996    1918.643586\n",
       "19997    1963.890277\n",
       "Name: 0, Length: 19998, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_dev = pd.read_csv('devout.tsv',header = None, sep = '/t',engine = 'python')\n",
    "y_dev = y_dev[0]\n",
    "y_dev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "cordless-maker",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_dev_exp = pd.read_csv('expected.tsv',header = None, sep = '/t',engine = 'python')\n",
    "y_dev_exp = y_dev_exp[0:19998]\n",
    "y_dev_exp = y_dev_exp[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "authorized-basics",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3631.1358243407444"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "RMSE_dev = mean_squared_error(y_dev_exp, y_dev)\n",
    "RMSE_dev "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "congressional-requirement",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_dev = pd.read_csv('dev-0/in.tsv',header = None, sep = '/t',engine = 'python')\n",
    "x_dev = x_dev[0]\n",
    "x_dev = pd.read_csv('dev-1/in.tsv',header = None, sep = '/t',engine = 'python')\n",
    "x_dev = x_dev[0]\n",
    "x_test = pd.read_csv('test-A/in.tsv',header = None, sep = '/t',engine = 'python')\n",
    "x_test = x_test[0]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "close-clinton",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "official-sweet",
   "metadata": {},
   "outputs": [],
   "source": [
    "#evaluation(x_dev,'dev-0/out.tsv', model)\n",
    "#evaluation(x_test,'test-A/out.tsv', model)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}