ngram_model_code.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import re\n",
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#UNK is used for unseen words in training vocabulary\n",
    "UNK= None\n",
    "#sentence start and end\n",
    "sent_start= \"<s>\"\n",
    "send_end= \"</s>\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def read_sentences_from_file(path):\n",
    "    with open(path, \"r\") as f:\n",
    "        #string.rstrip(\"\\n\") removes string portion after \\n or new line\n",
    "        #re.split( arg1, arg2): from arg2 extract arg1 pattern\n",
    "        string= [re.split(\"\\s+\", line.rstrip(\"\\n\")) for line in f]"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class unigram_language_model:\n",
    "    def __init__(self, sentences, smoothing=False):\n",
    "        self.unigram_frequencies= dict()\n",
    "        self.corpus_length= 0\n",
    "        for sentence in sentences:\n",
    "            for word in sentence:\n",
    "                #dict.get(key,val) , if the key is not present in the dict this func will assign \"val\" value to key and if it is already present then it returns the value of the key\n",
    "                self.unigram_frequencies[word]= self.unigram_frequencies.get(word, 0) + 1\n",
    "                if(word != sent_start and word != sent_end):\n",
    "                    self.corpus_length+=1\n",
    "        # subtract 2 from the total no. of unique words becoz sent_start and sent_end are also there\n",
    "        self.unique_words= len(self.unigram_frequencies) -2\n",
    "        self.smoothing= smoothing\n",
    "        \n",
    "    def calculate_unigram_probability(self, word):\n",
    "        word_probability_numerator= self.unigram_frequencies.get(word, 0)\n",
    "        word_probability_denominator= self.corpus_length\n",
    "        if(self.smoothing):\n",
    "            #add 1 smoothing is used \n",
    "            #1 is added to each word count in numerator and for denominator total unique words length is used\n",
    "            word_probability_numerator += 1\n",
    "            word_probability_denominator += self.unique_words +1 # here 1 is for UNK- unseen word\n",
    "        return float(word_probability_numerator)/float(word_probability_denominator)\n",
    "    \n",
    "    def calculate_sentence_probability(self, sentence, normalize_probability=True):\n",
    "        sentence_probability_log_sum=0\n",
    "        for word in sentence:\n",
    "            if(word != sent_start and word !=sent_end):\n",
    "                word_probability= self.calculate_unigram_probability(word)\n",
    "                sentence_probability_log_sum += math.log(word_probability, 2)\n",
    "        return math.pow(2, sentence_probability_log_sum) if(normalize_probability) else sentence_probability_log_sum\n",
    "    \n",
    "    def sorted_vocabulary(self):\n",
    "        full_vocab= list(self.unigram_frequencies.keys())\n",
    "        full_vocab.remove(sent_start)\n",
    "        full_vocab.remove(sent_end)\n",
    "        full_vocab.sort()\n",
    "        full_vocab.append(UNK)\n",
    "        full_vocab.append(sent_start)\n",
    "        full_vocab.append(sent_end)\n",
    "        return full_vocab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class bigram_language_model(unigram_language_model):\n",
    "    def __init__(self, sentences, smoothing=False):\n",
    "        self.bigram_frequencies= dict()\n",
    "        self.unique_bigrams= set()\n",
    "        for sentence in sentences:\n",
    "            previous_word= None\n",
    "            for word in sentence:\n",
    "                if(previous_word != None):\n",
    "                    self.bigram_frequencies[(previous_word, word)] = self.bigram_frequencies.get((previous_word, word), 0) + 1\n",
    "                    if(previous_word != sent_start and word != sent_end):\n",
    "                        self.unique_bigrams.add((previous_word, word))\n",
    "                previous_word= word\n",
    "        self.unique_bigram_words= len(self.unigram_frequencies)\n",
    "        \n",
    "    def calculate_bigram_probability(self, previous_word, word):\n",
    "        bigram_word_probability_numerator= self.bigram_frequencies.get((previous_word, word), 0)\n",
    "        bigram_word_probability_denominator= self.unigram_frequencies.get(previous_word, 0)\n",
    "        if(self.smoothing):\n",
    "            bigram_word_probability_numerator+=1\n",
    "            bigram_word_probability_denominator+= self.unique_bigram_words\n",
    "        return 0.0 if(bigram_word_probability_numerator==0 or bigram_word_probability_denominator==0) else float(bigram_word_probability_numerator)/float(bigram_word_probability_denominator)\n",
    "    \n",
    "    def calculate_bigram_sentence_probability(self, sentence, normalize_probability=True):\n",
    "        bigram_sentence_probability_log_sum= 0\n",
    "        previous_word= None\n",
    "        for word in sentence:\n",
    "            if(previous_word != None):\n",
    "                bigram_sentence_probability = self.calculate_bigram_probability(previous_word, word)\n",
    "                bigram_sentence_probability_log_sum += math.log(bigram_sentence_probability, 2)\n",
    "            previous_word= word\n",
    "        return pow(2, bigram_sentence_probability_log_sum) if(normalize_probability) else bigram_sentence_probability_log_sum"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Now, we calculate number of unigrams and bigrams\n",
    "def calculate_number_of_unigrams(sentences):\n",
    "    unigram_count=0\n",
    "    for sentence in sentences:\n",
    "        # remove 2 for <s> and </s>\n",
    "        unigram_count += len(sentence) -2\n",
    "    return unigram_count\n",
    "\n",
    "def calculate_number_of_bigrams(sentences):\n",
    "    bigram_count=0\n",
    "    for sentence in sentences:\n",
    "        #remove 1 for number of bigrams in sentence\n",
    "        bigram_count += len(sentence) -1\n",
    "    return bigram_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Calculate perplexity values\n",
    "def calculate_unigram_perplexity(model, sentences):\n",
    "    unigram_count= calculate_number_of_unigrams(sentences)\n",
    "    sentence_probability_log_sum= 0\n",
    "    for sentence in sentences:\n",
    "        try:\n",
    "            sentence_probability_log_sum -= math.log(model.calculate_sentence_probability(sentence), 2)\n",
    "        except:\n",
    "            sentence_probability_log_sum -= float('-inf')\n",
    "    return math.pow(2, sentence_probability_log_sum/unigram_count)\n",
    "\n",
    "def calculate_bigram_perplexity(model, sentences):\n",
    "    bigram_count= calculate_number_of_bigrams(sentences)\n",
    "    sentence_probability_log_sum= 0\n",
    "    for sentence in sentences:\n",
    "        try:\n",
    "            sentence_probability_log_sum -= math.log(model.calculate_bigram_sentence_probability(sentence), 2)\n",
    "        except:\n",
    "            sentence_probability_log_sum -= float('-inf')\n",
    "    return math.pow(2, sentence_probability_log_sum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# print unigram and bigram probs\n",
    "def print_unigram_probs(sorted_vocab_keys, model):\n",
    "    for vocab_key in sorted_vocab_keys:\n",
    "        if vocab_key != SENTENCE_START and vocab_key != SENTENCE_END:\n",
    "            print(\"{}: {}\".format(vocab_key if vocab_key != UNK else \"UNK\",\n",
    "                                       model.calculate_unigram_probability(vocab_key)), end=\" \")\n",
    "    print(\"\")\n",
    "\n",
    "def print_bigram_probs(sorted_vocab_keys, model):\n",
    "    print(\"\\t\\t\", end=\"\")\n",
    "    for vocab_key in sorted_vocab_keys:\n",
    "        if vocab_key != SENTENCE_START:\n",
    "            print(vocab_key if vocab_key != UNK else \"UNK\", end=\"\\t\\t\")\n",
    "    print(\"\")\n",
    "    for vocab_key in sorted_vocab_keys:\n",
    "        if vocab_key != SENTENCE_END:\n",
    "            print(vocab_key if vocab_key != UNK else \"UNK\", end=\"\\t\\t\")\n",
    "            for vocab_key_second in sorted_vocab_keys:\n",
    "                if vocab_key_second != SENTENCE_START:\n",
    "                    print(\"{0:.5f}\".format(model.calculate_bigram_probabilty(vocab_key, vocab_key_second)), end=\"\\t\\t\")\n",
    "            print(\"\")\n",
    "    print(\"\")"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "'NoneType' object is not iterable",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-9-675e5132a9f0>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      2\u001b[0m     \u001b[0mtwitter_dataset\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mread_sentences_from_file\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"sample.txt\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m     \u001b[0mtwitter_dataset_test\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mread_sentences_from_file\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"test_data.txt\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 4\u001b[1;33m     \u001b[0mtwitter_dataset_model_unsmoothed\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mbigram_language_model\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtwitter_dataset\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      5\u001b[0m     \u001b[0mtwitter_dataset_model_smoothed\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mbigram_language_model\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtwitter_dataset\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msmoothing\u001b[0m\u001b[1;33m=\u001b[0m \u001b[1;32mTrue\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      6\u001b[0m     \u001b[0msorted_vocab_keys\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mtwitter_dataset_model_unsmoothed\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msorted_vocabulary\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-5-c121df40fffa>\u001b[0m in \u001b[0;36m__init__\u001b[1;34m(self, sentences, smoothing)\u001b[0m\n\u001b[0;32m      3\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbigram_frequencies\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mdict\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0munique_bigrams\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mset\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 5\u001b[1;33m         \u001b[1;32mfor\u001b[0m \u001b[0msentence\u001b[0m \u001b[1;32min\u001b[0m \u001b[0msentences\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      6\u001b[0m             \u001b[0mprevious_word\u001b[0m\u001b[1;33m=\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      7\u001b[0m             \u001b[1;32mfor\u001b[0m \u001b[0mword\u001b[0m \u001b[1;32min\u001b[0m \u001b[0msentence\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mTypeError\u001b[0m: 'NoneType' object is not iterable"
     ]
    }
   ],
   "source": [
    "if(__name__ == \"__main__\"):\n",
    "    twitter_dataset= read_sentences_from_file(\"sample.txt\")\n",
    "    twitter_dataset_test= read_sentences_from_file(\"test_data.txt\")\n",
    "    twitter_dataset_model_unsmoothed= bigram_language_model(twitter_dataset)\n",
    "    twitter_dataset_model_smoothed= bigram_language_model(twitter_dataset, smoothing= True)\n",
    "    sorted_vocab_keys= twitter_dataset_model_unsmoothed.sorted_vocabulary()\n",
    "    \n",
    "    print(\"---------------- Toy dataset ---------------\\n\")\n",
    "    print(\"=== UNIGRAM MODEL ===\")\n",
    "    print(\"- Unsmoothed  -\")\n",
    "    print_unigram_probs(sorted_vocab_keys, twitter_dataset_model_unsmoothed)\n",
    "    print(\"\\n- Smoothed  -\")\n",
    "    print_unigram_probs(sorted_vocab_keys, twitter_dataset_model_smoothed)\n",
    "\n",
    "    print(\"\")\n",
    "\n",
    "    print(\"=== BIGRAM MODEL ===\")\n",
    "    print(\"- Unsmoothed  -\")\n",
    "    print_bigram_probs(sorted_vocab_keys, twitter_dataset_model_unsmoothed)\n",
    "    print(\"- Smoothed  -\")\n",
    "    print_bigram_probs(sorted_vocab_keys, twitter_dataset_model_smoothed)\n",
    "\n",
    "    print(\"\")\n",
    "\n",
    "    print(\"== SENTENCE PROBABILITIES == \")\n",
    "    longest_sentence_len = max([len(\" \".join(sentence)) for sentence in twitter_dataset_test]) + 5\n",
    "    print(\"sent\", \" \" * (longest_sentence_len - len(\"sent\") - 2), \"uprob\\t\\tbiprob\")\n",
    "    for sentence in twitter_dataset_test:\n",
    "        sentence_string = \" \".join(sentence)\n",
    "        print(sentence_string, end=\" \" * (longest_sentence_len - len(sentence_string)))\n",
    "        print(\"{0:.5f}\".format(twitter_dataset_model_smoothed.calculate_sentence_probability(sentence)), end=\"\\t\\t\")\n",
    "        print(\"{0:.5f}\".format(twitter_dataset_model_smoothed.calculate_bigram_sentence_probability(sentence)))        \n",
    "        \n",
    "    print(\"\")\n",
    "\n",
    "    print(\"== TEST PERPLEXITY == \")\n",
    "    print(\"unigram: \", calculate_unigram_perplexity(twitter_dataset_model_smoothed, twitter_dataset_test))\n",
    "    print(\"bigram: \", calculate_bigram_perplexity(twitter_dataset_model_smoothed, twitter_dataset_test))\n",
    "    \n",
    "    print(\"\")\n",
    "\n",
    "    actual_dataset = read_sentences_from_file(\"sample.txt\")\n",
    "    actual_dataset_test = read_sentences_from_file(\"test_data.txt\")\n",
    "    actual_dataset_model_smoothed = bigram_language_model(actual_dataset, smoothing=True)\n",
    "    print(\"---------------- Actual dataset ----------------\\n\")\n",
    "    print(\"PERPLEXITY of sample.txt\")\n",
    "    print(\"unigram: \", calculate_unigram_perplexity(actual_dataset_model_smoothed, actual_dataset))\n",
    "    print(\"bigram: \", calculate_bigram_perplexity(actual_dataset_model_smoothed, actual_dataset))\n",
    "\n",
    "    print(\"\")\n",
    "\n",
    "    print(\"PERPLEXITY of test_data.txt\")\n",
    "    print(\"unigram: \", calculate_unigram_perplexity(actual_dataset_model_smoothed, actual_dataset_test))\n",
    "    print(\"bigram: \", calculate_bigram_perplexity(actual_dataset_model_smoothed, actual_dataset_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}