From 23768372ce6a4c6c36035b084f3bf8760970b297 Mon Sep 17 00:00:00 2001
From: manu8996 <manusingh.8996@gmail.com>
Date: Wed, 21 Nov 2018 05:20:26 +0530
Subject: [PATCH] Sentence tfidf calculated

---
 Text Summarisation.ipynb | 135 ++++++++++++++++++++++++++++-----------
 1 file changed, 97 insertions(+), 38 deletions(-)

diff --git a/Text Summarisation.ipynb b/Text Summarisation.ipynb
index 347eabe..bfe51a0 100644
--- a/Text Summarisation.ipynb	
+++ b/Text Summarisation.ipynb	
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -14,7 +14,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -27,7 +27,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -36,7 +36,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -45,7 +45,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -62,7 +62,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -80,7 +80,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -127,7 +127,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -166,7 +166,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
@@ -463,7 +463,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -482,7 +482,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
@@ -500,7 +500,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {
@@ -519,7 +519,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
     {
@@ -536,7 +536,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 14,
    "metadata": {
     "scrolled": true
    },
@@ -557,7 +557,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -571,7 +571,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 16,
    "metadata": {
     "scrolled": false
    },
@@ -662,7 +662,7 @@
        " 'actually, it is immaterial to my argument whether all the\\ncomplements hang from the same node or whether they, too, are introduced\\nby binary branching, like the adjuncts.']"
       ]
      },
-     "execution_count": 19,
+     "execution_count": 16,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -673,7 +673,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -695,7 +695,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 18,
    "metadata": {
     "scrolled": false
    },
@@ -811,7 +811,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [
     {
@@ -1171,7 +1171,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [
     {
@@ -1188,7 +1188,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [
     {
@@ -1205,7 +1205,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [
     {
@@ -1222,7 +1222,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [
     {
@@ -1243,31 +1243,90 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 139,
    "metadata": {},
    "outputs": [
     {
-     "data": {
-      "text/plain": [
-       "<1x443 sparse matrix of type '<class 'numpy.float64'>'\n",
-       "\twith 19 stored elements in Compressed Sparse Row format>"
-      ]
-     },
-     "execution_count": 29,
-     "metadata": {},
-     "output_type": "execute_result"
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'scipy.sparse.csr.csr_matrix'>\n"
+     ]
     }
    ],
    "source": [
-    "X[0]"
+    "print(type(X[35,:]))"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 140,
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
+      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n"
+     ]
+    }
+   ],
+   "source": [
+    "print((X[35,:].toarray()[0]))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 141,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0.2133141289945213, 0.19268349217056602, 0.23567236548378917, 0.20918963054521372, 0.20414163476852237, 0.2142031418888816, 0.2953940191257623, 0.3190722211068253, 0.24766883854585858, 0.24359008885419822, 0.2691392393632223, 0.21792781559904986, 0.4409236974794452, 0.2003291117020829, 0.141093056522983, 0.3013860982971811, 0.2648058357549551, 0.154810820419545, 0.22094264628922314, 0.19597742106773774, 0.12235705255435296, 0.3420746571230864, 0.7068189089897328, 0.18711131675460949, 0.17564167393574218, 0.1600690831192614, 0.14707137788350813, 0.24855083679987244, 0.31842509242552686, 0.3622285702824559, 0.20744839403144236, 0.2915297848121885, 0.25615439420276653, 0.32492545860541067, 0.3372416580862453, 0, 0.21810916095415211, 0, 0.25201721754566336, 0.1168623335134955, 0.1850542899139658, 0.4340132904839581, 0.2558776048492524, 0, 0.23871620079008612, 0, 0.21505914004381577, 0.188702608551362, 0.2473621280450456, 0.17103202855623842, 0.30282806549952346, 0.447213595499958, 0.7071067811865475, 0.3616038758957919, 0.44575716620707667, 0.24363129254497012, 0.4977539498337, 0.3950854623715471, 0.4977539498337, 0.31133778533740636, 0.4977539498337, 0.4015513269236906, 0.5768478040843482, 0.5773502691896257, 0.39184457982347703, 0.3697810962297591, 0.43673030234484533, 0.5503728292507438, 0.7071067811865475, 0.3936975574505491, 0.4996381244664003, 0.3656381806345497, 0.7071067811865475, 0.37744343499577615, 0.7071067811865475, 0.21557143081821167, 0.5603565195423051, 0.41760955902435687, 0.19846903596500826, 0.22973631970082772, 0.19687072052499346]\n"
+     ]
+    }
+   ],
+   "source": [
+    "sent_tfid = []\n",
+    "for i in range(X.shape[0]):\n",
+    "    count = 0\n",
+    "    sum_row = 0\n",
+    "    #k = X[i,:].toarray()\n",
+    "    for j in range(X.shape[1]):\n",
+    "        if X[i,:].toarray()[0][j] != 0:\n",
+    "            sum_row +=  X[i,:].toarray()[0][j]\n",
+    "            count += 1\n",
+    "    try:        \n",
+    "        div_ans = sum_row / count\n",
+    "    except:\n",
+    "        div_ans = 0\n",
+    "    #print(i)\n",
+    "    sent_tfid.append(div_ans)\n",
+    "print(sent_tfid)\n",
+    "        \n",
+    "            \n",
+    "        "
+   ]
   }
  ],
  "metadata": {