embedding_gensim-udpipe.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import unicode_literals\n",
    "# -*- coding: utf-8 -*-\n",
    "STOP_WORDS = \"\"\"\n",
    " ابھی از \n",
    " اس استعمال اسی اسے البتہ الف ان اندر اور اوپر اکثر اگر\n",
    " اگرچہ اگلے ایسا ایسی ایسے اے بار بارے باوجود باہر بظاہر بعض بغیر بلکہ بن\n",
    " بھر بھریں بھی بہت بے تا تاکہ تب تحت تر تمام\n",
    " تو تک جب سوال \n",
    " جبکہ جو حالانکہ حالاں خلاف\n",
    " خود سے میں چنانچہ دیر ذریعے تھا هے ایک آپ\n",
    " سا ساتھ سامنے سب تھی تھے\n",
    " سکا سکتا سکتے سی سے شان شاید صرف صورت ضرورت ضروری طرح طرف طور علاوہ عین غیر \n",
    " لہذا لیکن لیں لیے لے مجھ مجھے مزید مقابلے مل مکمل مگر \n",
    " نا نہ نہیں نیچے واقعی والا والوں والی والے وجہ وغیرہ وہ وہاں وہی وہیں وی ویسے پاس\n",
    " پایا پر پوری پھر پیچھے چونکہ چکی\n",
    " ڈالے کئے کافی کبھی کسی کم \n",
    " کوئی کچھ کہ کہا کہہ کہیں کہے کیونکہ کیے کے گئی\n",
    " گئے گا گویا گی گے ہاں ہر ہمیشہ ہو ہوئی ہوئیں ہوئے ہوا ہوتا\n",
    " ہوتی ہوتیں ہوتے ہونا ہونگے ہونی ہونے ہوں ہی ہیں ہے یا یات یعنی یقینا یہ یہاں یہی یہیں ھیں\n",
    " ہوگی تقریباً\n",
    " کا کو کی نے ہوتو پہلے کر\n",
    " اپنا حتى كہ ميں لئیے  \n",
    " رض\n",
    "پس\n",
    "طرح\n",
    " بعد ارادہ افاقہ آخر\n",
    "\"\"\".split()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "from ufal.udpipe import Model, Pipeline, ProcessingError\n",
    "import pandas as pd \n",
    "from conllu import parse\n",
    "\n",
    "model = Model.load('E:/Software/urdu-udtb-ud-2.3-181115.udpipe')\n",
    "pipeline = Pipeline(model, 'tokenize', Pipeline.DEFAULT, Pipeline.DEFAULT, 'conllu')\n",
    "error = ProcessingError()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.0%\n",
      "2.0%\n",
      "3.0%\n",
      "4.0%\n",
      "5.0%\n",
      "6.0%\n",
      "7.0%\n",
      "8.0%\n",
      "9.0%\n",
      "10.0%\n",
      "11.0%\n",
      "12.0%\n",
      "13.0%\n",
      "14.0%\n",
      "15.0%\n",
      "16.0%\n",
      "17.0%\n",
      "18.0%\n",
      "19.0%\n",
      "20.0%\n",
      "21.0%\n",
      "22.0%\n",
      "23.0%\n",
      "24.0%\n",
      "25.0%\n",
      "26.0%\n",
      "27.0%\n",
      "28.0%\n",
      "29.0%\n",
      "30.0%\n",
      "31.0%\n",
      "32.0%\n",
      "33.0%\n",
      "34.0%\n",
      "35.0%\n",
      "36.0%\n",
      "37.0%\n",
      "38.0%\n",
      "39.0%\n",
      "40.0%\n",
      "41.0%\n",
      "42.0%\n",
      "43.0%\n",
      "44.0%\n",
      "45.0%\n",
      "46.0%\n",
      "47.0%\n",
      "48.0%\n",
      "49.0%\n",
      "50.0%\n",
      "51.0%\n",
      "52.0%\n",
      "53.0%\n",
      "54.0%\n",
      "55.0%\n",
      "56.0%\n",
      "57.0%\n",
      "58.0%\n",
      "59.0%\n",
      "60.0%\n",
      "61.0%\n",
      "62.0%\n",
      "63.0%\n",
      "64.0%\n",
      "65.0%\n",
      "66.0%\n",
      "67.0%\n",
      "68.0%\n",
      "69.0%\n",
      "70.0%\n",
      "71.0%\n",
      "72.0%\n",
      "73.0%\n",
      "74.0%\n",
      "75.0%\n",
      "76.0%\n",
      "77.0%\n",
      "78.0%\n",
      "79.0%\n",
      "80.0%\n",
      "81.0%\n",
      "82.0%\n",
      "83.0%\n",
      "84.0%\n",
      "85.0%\n",
      "86.0%\n",
      "87.0%\n",
      "88.0%\n",
      "89.0%\n",
      "90.0%\n",
      "91.0%\n",
      "92.0%\n",
      "93.0%\n",
      "94.0%\n",
      "95.0%\n",
      "96.0%\n",
      "97.0%\n",
      "98.0%\n",
      "99.0%\n",
      "100.0%\n",
      "Query count: 1000\n"
     ]
    }
   ],
   "source": [
    "import pyodbc\n",
    "from gensim.models import Word2Vec\n",
    "\n",
    "conn = pyodbc.connect(r'Driver={Microsoft Access Driver (*.mdb, *.accdb)};DBQ=E:\\Software\\EQWH V6.0 160620\\EasyHadees.mdb;')\n",
    "cursor = conn.cursor()\n",
    "cursor.execute(\"select top 1000 urdutext from ahadith where urdutext is not null and urdutext like '%وضو%'\")\n",
    "querycount = 0\n",
    "sentences = []\n",
    "#joined_sentences = []\n",
    "for row in cursor.fetchall(): \n",
    "    querycount +=1\n",
    "    x1 = row[0]  #'ابوالنعمان، ابوعوانہ، ابی بشر، یوسف بن ماہک، عبداللہ بن عمرو (رض) کہتے ہیں کہ ایک سفر میں نبی (صلی اللہ علیہ وآلہ وسلم) ہم سے پیچھے رہ گئے'\n",
    "    sanad = row[0]\n",
    "    x2= x1.split(' کہ ')\n",
    "    if(len(x2)>0):\n",
    "        sanad = ''\n",
    "        for i in range(1,len(x2)):\n",
    "            sanad += x2[i]\n",
    "    if len(sanad) == 0:\n",
    "        sanad = row[0]\n",
    "    #print(sanad)\n",
    "    #print()\n",
    "    processed = pipeline.process(sanad, error)\n",
    "    if error.occurred():\n",
    "        print(\"An error occurred when running run_udpipe: \")  \n",
    "    \n",
    "    data = parse(processed)\n",
    "    \n",
    "    wordarray = []\n",
    "    try:\n",
    "        if len(data[0]) > 0:\n",
    "            for ir in range(0,len(data[0])):\n",
    "                if data[0][ir]['upostag'] == \"NOUN\" and data[0][ir]['form'] not in STOP_WORDS:\n",
    "                     wordarray.append(data[0][ir]['form'])\n",
    "    except:\n",
    "        print(len(sanad))        \n",
    "                \n",
    "    sentences.append(wordarray)\n",
    "    #joined_sentences.append(sen)\n",
    "    if querycount%10==0:\n",
    "        print(str(querycount/10) + \"%\")\n",
    "#joined_sentences\n",
    "print(\"Query count: \"+str(querycount))\n",
    "#print(sentences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Word2Vec(vocab=2094, size=100, alpha=0.025)\n"
     ]
    }
   ],
   "source": [
    "model = Word2Vec(sentences, min_count=1)\n",
    "# summarize the loaded model\n",
    "print(model)\n",
    "# summarize vocabulary\n",
    "words = list(model.wv.vocab)\n",
    "#print(words)\n",
    "# access vector for one word\n",
    "#print(model['ٹخنوں'])\n",
    "# save model\n",
    "model.save('model.bin')\n",
    "# load model\n",
    "new_model = Word2Vec.load('model.bin')\n",
    "#print(new_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualize, Plot Word Vectors Using PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\admin\\Anaconda3\\envs\\islamic qa\\lib\\site-packages\\ipykernel_launcher.py:6: DeprecationWarning: Call to deprecated `__getitem__` (Method will be removed in 4.0.0, use self.wv.__getitem__() instead).\n",
      "  \n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x864 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "from matplotlib import pyplot\n",
    "from bidi import algorithm as bidialg\n",
    "import arabic_reshaper\n",
    "\n",
    "X = model[model.wv.vocab]\n",
    "pca = PCA(n_components=2)\n",
    "result = pca.fit_transform(X)\n",
    "\n",
    "from matplotlib.pyplot import figure\n",
    "figure(num=None, figsize=(12, 9), dpi=96, facecolor='w', edgecolor='k')\n",
    "\n",
    "pyplot.scatter(result[:, 0], result[:, 1], s=1)\n",
    "\n",
    "words = list(model.wv.vocab)\n",
    "for i, word in enumerate(words):\n",
    "    text = bidialg.get_display( arabic_reshaper.reshape(word))\n",
    "    #text = bidialg.get_display(u''+word+'')\n",
    "    #print(text,'  ',word)\n",
    "    pyplot.annotate(text, xy=(result[i, 0], result[i, 1]))\n",
    "\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\mazhar.SFC\\AppData\\Local\\Continuum\\anaconda3\\envs\\islamicqa\\lib\\site-packages\\ipykernel_launcher.py:1: DeprecationWarning: Call to deprecated `similarity` (Method will be removed in 4.0.0, use self.wv.similarity() instead).\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.9995203332746023"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.similarity('مسح','وضو')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\mazhar.SFC\\AppData\\Local\\Continuum\\anaconda3\\envs\\islamicqa\\lib\\site-packages\\ipykernel_launcher.py:2: DeprecationWarning: Call to deprecated `most_similar` (Method will be removed in 4.0.0, use self.wv.most_similar() instead).\n",
      "  \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[('نماز', 0.9998999238014221),\n",
       " ('رض', 0.9998710751533508),\n",
       " ('پانی', 0.9998375773429871),\n",
       " ('پس', 0.9998369216918945),\n",
       " ('روایت', 0.9998307228088379),\n",
       " ('طرح', 0.9998277425765991),\n",
       " ('شخص', 0.99982750415802),\n",
       " ('ہاتھ', 0.9998089075088501),\n",
       " ('حکم', 0.9998078942298889),\n",
       " ('بیان', 0.9997956156730652),\n",
       " ('لوگوں', 0.9997929334640503),\n",
       " ('نبی', 0.9997862577438354),\n",
       " ('عرض', 0.999780535697937),\n",
       " ('وقت', 0.9997740387916565),\n",
       " ('دن', 0.9997681379318237),\n",
       " ('دعا', 0.9997678399085999),\n",
       " ('رات', 0.9997600317001343),\n",
       " ('ادا', 0.9997538328170776),\n",
       " ('دریافت', 0.9997473955154419),\n",
       " ('حاجت', 0.9997448921203613),\n",
       " ('حدیث', 0.9997448325157166),\n",
       " ('سر', 0.9997414350509644),\n",
       " ('جگہ', 0.9997376203536987),\n",
       " ('بات', 0.9997340440750122),\n",
       " ('شروع', 0.9997265338897705),\n",
       " ('ذکر', 0.9997261166572571),\n",
       " ('لوگ', 0.9997224807739258),\n",
       " ('آدمی', 0.999721884727478),\n",
       " ('غسل', 0.9997127652168274),\n",
       " ('تشریف', 0.9997121095657349),\n",
       " ('گھر', 0.9997004270553589),\n",
       " ('ہاتھوں', 0.9996985793113708),\n",
       " ('عائشہ', 0.9996979236602783),\n",
       " ('منہ', 0.999697744846344),\n",
       " ('مرتبہ', 0.9996960163116455),\n",
       " ('مسجد', 0.9996930360794067),\n",
       " ('گناہ', 0.9996883869171143),\n",
       " ('خدمت', 0.999687135219574),\n",
       " ('صحابہ', 0.9996869564056396),\n",
       " ('رکعتیں', 0.9996715784072876),\n",
       " ('عن', 0.9996660947799683),\n",
       " ('ارشاد', 0.9996496438980103),\n",
       " ('عورت', 0.9996410012245178),\n",
       " ('جواب', 0.9996381402015686),\n",
       " ('مسئلہ', 0.9996241927146912),\n",
       " ('بار', 0.9996093511581421),\n",
       " ('عمر', 0.9996041059494019),\n",
       " ('حالت', 0.9996019601821899),\n",
       " ('ابوبکر', 0.9995893239974976),\n",
       " ('اذان', 0.9995839595794678),\n",
       " ('آیت', 0.99957275390625),\n",
       " ('والد', 0.9995713829994202),\n",
       " ('طریقہ', 0.9995682239532471),\n",
       " ('نقل', 0.999566376209259),\n",
       " ('پاؤں', 0.9995497465133667),\n",
       " ('واسطے', 0.9995473623275757),\n",
       " ('باب', 0.9995421171188354),\n",
       " ('طرف', 0.9995378255844116),\n",
       " ('اہل', 0.9995301365852356),\n",
       " ('ارادہ', 0.9995224475860596),\n",
       " ('دل', 0.9995211362838745),\n",
       " ('مقام', 0.9995208382606506),\n",
       " ('مسح', 0.9995203018188477),\n",
       " ('رکعت', 0.9995194673538208),\n",
       " ('قول', 0.9995137453079224),\n",
       " ('مشکیزے', 0.9995065927505493),\n",
       " ('حصہ', 0.9994879364967346),\n",
       " ('عمل', 0.9994679689407349),\n",
       " ('جنت', 0.9994600415229797),\n",
       " ('کام', 0.9994550347328186),\n",
       " ('گوشت', 0.9994524717330933),\n",
       " ('آواز', 0.999439537525177),\n",
       " ('نام', 0.9994329214096069),\n",
       " ('داخل', 0.9994319081306458),\n",
       " ('راوی', 0.9994276165962219),\n",
       " ('ظہر', 0.9994171261787415),\n",
       " ('مذی', 0.9994164705276489),\n",
       " ('برتن', 0.9994063377380371),\n",
       " ('حیض', 0.9993823170661926),\n",
       " ('زمین', 0.9993798136711121),\n",
       " ('امام', 0.9993699193000793),\n",
       " ('علم', 0.9993657469749451),\n",
       " ('عصر', 0.999363362789154),\n",
       " ('سفر', 0.9993396997451782),\n",
       " ('خون', 0.9993305206298828),\n",
       " ('چہرہ', 0.9993272423744202),\n",
       " ('حسن', 0.9993253350257874),\n",
       " ('صبح', 0.9993233680725098),\n",
       " ('جنابت', 0.999274730682373),\n",
       " ('آگ', 0.9992575645446777),\n",
       " ('کھانا', 0.9992541074752808),\n",
       " ('سجدہ', 0.9992303848266602),\n",
       " ('رح', 0.9992285966873169),\n",
       " ('قسم', 0.9992092251777649),\n",
       " ('گرہ', 0.9992050528526306),\n",
       " ('چیز', 0.9992036819458008),\n",
       " ('موزوں', 0.9991942644119263),\n",
       " ('کلی', 0.9991884827613831),\n",
       " ('نور', 0.9991729259490967),\n",
       " ('بیوی', 0.9991603493690491)]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#model = Word2Vec.load('model.bin')\n",
    "model.most_similar(positive='وضو', topn=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": 2
}