218417985_assignment2_solution.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "----\n",
    "\n",
    "### <span style=\"color:#0b486b\">SIT744 Practical Machine Learning for Data Science</span>\n",
    "\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "# <span style=\"color:#0b486b\">Assignment Two: Deep Neural Networks, Representation Learning, and Text Analytics</span>\n",
    "### Due: <span style=\"color:red\">24:00pm 24 September 2019</span>  (Tuesday)\n",
    "\n",
    "#### <span style=\"color:red\">Important note:</span> This is an **individual** assignment. It contributes **40%** to your final mark. Read the assignment instruction carefully."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook has been prepared for your to complete Assignment 2. The theme of this assignment is about practical machine learning knowledge and skills in deep neural networks, word embedding and text analytics. Some sections have been partially completed to help you get\n",
    "started. **The total marks for this notebook are 80 marks, which will be re-scaled to 40 marks in the grade.**.\n",
    "\n",
    "* Before you start, read the entire notebook carefully once to understand what you need to do. <br><br>\n",
    "\n",
    "* For each cell marked with **#YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL**, there will be places where you **must** supply your own codes when instructed. <br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <span style=\"color:#0b486b\">Instruction</span>\n",
    "\n",
    "This assignment contains **two** parts \n",
    "\n",
    "* Part 1: Deep Feedforward Neural Network **[45 points]**\n",
    "* Part 2: Word2Vec, text analytics and application **[35 points]**\n",
    "\n",
    "**Hint**: This assignment was essentially designed based on the lectures and practical lab sessions covered from Week 5 to 9. You are strongly recommended to go through these contents thoroughly which might help you to complete this assignment."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <span style=\"color:#0b486b\">What to submit</span>\n",
    "\n",
    "This assignment is to be completed individually and submitted to CloudDeakin. **By the due date, you are required to submit the following files to the corresponding Assignment in CloudDeakin**:\n",
    "\n",
    "1.\t<span style=\"color:red\">**[YourID]_assignment2_solution.ipynp**</span>:  **this is your Python notebook solution source file**.\n",
    "1.\t<span style=\"color:red\">**[YourID]_assingment2_output.html**</span>: **this is the output of your Python notebook solution *exported* in html format**.\n",
    "1.\t<span style=\"color:red\">Any extra files needed to complete your assignment</span> (e.g., images used in your answers).\n",
    "\n",
    "For example, if your student ID is: 123456, you will then need to submit the following files:\n",
    "* 123456_assignment2_solution.ipynp\n",
    "* 123456_assignment2_output.html\n",
    "* any extra files or subfolder you might have (this can be named according to your preference).\n",
    "\n",
    "<span style=\"color:red\">Please proceed to the content below to complete your assignment!</span>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <span style=\"color:#0b486b\">Part 1: Deep Feedforward Neural Network </span>\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[Total mark for this part: 45 points]**<span></div>\n",
    "\n",
    "The first part of this assignment is for you to demonstrate the knowledge in deep learning that you have acquired from the lectures and practical lab materials. Most of the contents in this assignment are drawn from the practical materials in week 5, 6 and 7 for deep neural networks. Going through these materials before attempting this assignment is highly recommended."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Run the following cell to create necessary subfolders for this assignment. You must **not** modify these codes and **must** run it first*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create necessary subfolders to store immediate files for this assignment.\n",
    "\n",
    "import os\n",
    "if not os.path.exists(\"./models/dnn0\"):\n",
    "    os.makedirs(\"models/dnn0\")\n",
    "    \n",
    "#custom function for pretty printing\n",
    "\n",
    "def printDash(n=50):\n",
    "    print('='*n)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first part of this assignment is to apply DNN to recognize letters from A-Z. You have played with MNIST dataset in your pracs and this should have given a good sense of how to apply DNN on images for recognition task. \n",
    "\n",
    "In this assignment, you are going to work with the **notMNIST** dataset for *letter recognition task*. The dataset contains 10 classes of letters A-J taken from different fonts. You will see some examples at the visualization task in the next part. A short blog about the data can be found [here](http://yaroslavvb.blogspot.com.au/2011/09/notmnist-dataset.html).\n",
    "\n",
    "Here we only consider a small subset which can be found at [this link](http://yaroslavvb.com/upload/notMNIST/notMNIST_small.mat). This file has been already downloaded and stored in subfolder `datasets` of this assignment folder. The file is in *Matlab* format, thus our first task is to:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 1.1**</span>. Load the data into *`numpy array`* format of two variables:\n",
    "* *`x`*: storing features with dimension `[num_samples, width, height]` (`num_samples`: number of samples, `width`: image width, `height`: image height), and\n",
    "* *`y`*: storing labels with dimension `num_samples`. \n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[3 points]**</span></div>\n",
    "\n",
    "Enter the missing codes in the following cell to complete this question."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['__header__', '__version__', '__globals__', 'images', 'labels'])\n",
      "(18724,)\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "from prettytable import PrettyTable\n",
    "import numpy as np\n",
    "import scipy.io as sio\n",
    "import pandas as pd\n",
    "from pprint import pprint\n",
    "\n",
    "data = sio.matlab.loadmat(\"datasets/notMNIST_small.mat\")\n",
    "#print(type(data))\n",
    "#print(data.images)\n",
    "#print(data['images'])\n",
    "print(data.keys())\n",
    "x, y = data['images'],data['labels']\n",
    "x = np.rollaxis(x, axis=2)\n",
    "print(y.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 1.2**</span>. Print out the total number of data points, and the *unique* labels in this dataset.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[3 points]**</span></div> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data points =  18724\n",
      "Unique labels =  [0. 1. 2. 3. 4. 5. 6. 7. 8. 9.]\n"
     ]
    }
   ],
   "source": [
    "#YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "print(\"Data points = \",format(x.shape[0]))\n",
    "print(\"Unique labels = \",format(np.unique(y)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 1.3**</span>. Display 100  images in the form of `10x10` matrix, each row showing 10 *random* images of a label. You might decide to use the function `display_images` provided at the beginning of this assignment, or you can write your own codes.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[4 points]**</span></div> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# this function is a utility to display images from the dataset\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "def display_images(images, shape):\n",
    "    fig = plt.figure(figsize=shape)\n",
    "    fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)\n",
    "    for i in range(np.prod(shape)):\n",
    "        p = fig.add_subplot(shape[0], shape[1], i+1, xticks=[], yticks=[])\n",
    "        p.imshow(images[i], cmap=plt.cm.bone)                "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 100 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "\n",
    "unique_labels = np.unique(y)\n",
    "images = []\n",
    "for l in unique_labels:    \n",
    "    idx = np.where(y == l)[0]\n",
    "    idx = idx[np.random.permutation(len(idx))[:10]]    \n",
    "    for i in idx:\n",
    "        images += [x[i]]\n",
    "\n",
    "display_images(images, shape=(10, 10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 1.4**</span>. Use the *deep feedforward neural network* as the classifier to perform images classification task in a *single split training and testing*.\n",
    "\n",
    "The total marks for this question is <span style=\"color:red\">**[35 points]**</span>, with the following detailed breakdown sub-questions:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"color:red\">**(a)**</span> **Write your code to reshape the variable `x` storing features from `[num_samples, width, height]` dimension to `[num_samples, num_features]` with `num_features = width x height`. \n",
    "** (*Hint*: you might want to use the `reshape()` function)**<div style=\"text-align: right\"> <span style=\"color:red\">**[3 points]**</span> </div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(18724, 784)\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "r,c = x[0].shape\n",
    "x = np.reshape(x,(y.shape[0],r*c))\n",
    "print(x.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In training the DNN, scaling data is important. The pixel intensities of images are in the range of [0, 255], which makes the neural network difficult to learn."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Rescale the input data into the range of [0, 1]**\n",
    "<div style=\"text-align: right\"> \n",
    "<span style=\"color:red\">**[2 points]**</span> </div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "After scaling:\n",
      "Min = 0.0\n",
      "Max = 1.0\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "x_scale = MinMaxScaler().fit(x).transform(x)\n",
    "print(\"After scaling:\")\n",
    "print(\"Min =\",format(x_scale.min()))\n",
    "print(\"Max =\",format(x_scale.max()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"color:red\">**(b)**</span> **Split the data into two subsets: 70% for training and 30% for testing. Note that you must use [*Stratified-Shuffle-Split*](http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.StratifiedShuffleSplit.html) to make sure training and testing are balanced and randomly shuffled before learning the model.**\n",
    "<div style=\"text-align: right\"> <span style=\"color:red\">**[5 points]**</span> </div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18724\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "ss_split=StratifiedShuffleSplit(n_splits=1,test_size=0.3,random_state=11795)\n",
    "\n",
    "for train_index, test_index in ss_split.split(x_scale,y):\n",
    "#    print(\"TRAIN:\", train_index, \"TEST:\", test_index)\n",
    "    print(train_index.shape[0]+test_index.shape[0])\n",
    "    X_train, X_test = x_scale[train_index], x_scale[test_index]\n",
    "    y_train, y_test = y[train_index], y[test_index]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"color:red\">**(c)**</span> **Construct a deep feedforward neural network with the following architecture:**\n",
    "\n",
    "* An input layer followed by *two* hidden layers, each with *500* hidden units, and an *output* layer;\n",
    "* *ReLU* activations for neurons in each hidden layer;\n",
    "* Training with gradient descent optimizer with learning rate **0.0011**, batch size 128 and 50 epochs.\n",
    "\n",
    "(*Hint*: *this question heavily relies on the knowledge you've learned from lab session in week 5 and 6. You are encouraged to revise these materials for this question*)\n",
    "\n",
    "<div style=\"text-align: right\"> <span style=\"color:red\">**[20 points]**</span> </div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL [5 marks]\n",
    "\n",
    "import tensorflow as tf\n",
    "\n",
    "tf.reset_default_graph()\n",
    "\n",
    "learn_rate = 0.0011\n",
    "num_inputs = r*c\n",
    "num_hidden1 = 500\n",
    "num_hidden2 = 500\n",
    "num_outputs = len(np.unique(y))\n",
    "\n",
    "inputs =  tf.placeholder(tf.float32,shape=[None, num_inputs],name='inputs')\n",
    "labels = tf.placeholder(tf.int64,shape=[None],name='labels')\n",
    "\n",
    "W = tf.Variable(tf.truncated_normal([num_inputs, num_outputs], stddev=0.02), name='weights')\n",
    "b = tf.Variable(tf.zeros([num_outputs]), name='biases')\n",
    "logits = tf.add(tf.matmul(inputs, W),b, name='logits')\n",
    "with tf.name_scope('evaluation'):\n",
    "    xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=labels, logits=logits, name='xentropy')\n",
    "    loss = tf.reduce_mean(xentropy, name='loss')\n",
    "    correct = tf.nn.in_top_k(logits, labels, 1)\n",
    "    accuracy = tf.reduce_mean(tf.cast(correct, tf.float32), name='accuracy')\n",
    "with tf.name_scope(\"train\"):\n",
    "    optimizer=tf.train.GradientDescentOptimizer(learn_rate)\n",
    "    training_op=optimizer.minimize(loss)\n",
    "    \n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL [3 marks]\n",
    "\n",
    "def neuron_layer(x, num_neurons, name, activation=None):\n",
    "    with tf.name_scope(name):\n",
    "        num_inputs=int(inputs.get_shape()[1])\n",
    "        std_dev=2/np.sqrt(num_inputs)\n",
    "        init=tf.truncated_normal([num_inputs,num_neurons],stddev=std_dev)\n",
    "        b = tf.Variable(tf.zeros([num_neurons]), name='biases')\n",
    "        W = tf.Variable(init,name='weights')\n",
    "        z = tf.matmul(inputs,W)+b\n",
    "    if activation == \"sigmoid\":\n",
    "        return tf.nn.sigmoid(z)\n",
    "    elif activation == \"relu\":\n",
    "        return tf.nn.relu(z)\n",
    "    else:\n",
    "        return z"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL [7 marks]\n",
    "\n",
    "with tf.name_scope(\"dnn\"):\n",
    "    hidden1 = neuron_layer(num_inputs,num_hidden1, \"hidden1\", activation=\"relu\")\n",
    "    hidden2 = neuron_layer(hidden1, num_hidden2, \"hidden2\", activation=\"relu\")\n",
    "    logits = neuron_layer(hidden2, num_outputs, \"output\")\n",
    "with tf.name_scope(\"loss\"):\n",
    "    xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=labels, logits=logits, name='xentropy')\n",
    "    loss = tf.reduce_mean(xentropy, name=\"loss\")\n",
    "    \n",
    "with tf.name_scope(\"evaluation\"):\n",
    "    xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=labels, logits=logits, name='xentropy')\n",
    "    correct = tf.nn.in_top_k(logits, labels, 1)\n",
    "    accuracy = tf.reduce_mean(tf.cast(correct, tf.float32), name='accuracy')\n",
    "    \n",
    "with tf.name_scope(\"train\"):\n",
    "    optimizer=tf.train.GradientDescentOptimizer(learn_rate)\n",
    "    grads=optimizer.compute_gradients(loss)\n",
    "    training_op=optimizer.apply_gradients(grads)\n",
    "    init = tf.global_variables_initializer()\n",
    "    saver=tf.train.Saver()\n",
    "    \n",
    "    for var in tf.trainable_variables():\n",
    "        tf.summary.histogram(var.op.name + \"/values\", var)\n",
    "        \n",
    "    for grad, var in grads:\n",
    "        if grad is not None:\n",
    "            tf.summary.histogram(var.op.name + \"/gradients\", grad)\n",
    "\n",
    "# summary\n",
    "accuracy_summary = tf.summary.scalar('accuracy',accuracy)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL [5 marks]\n",
    "\n",
    "# merge all summary\n",
    "tf.summary.histogram('hidden1/activations', hidden1)\n",
    "tf.summary.histogram('hidden2/activations', hidden2)\n",
    "merged = tf.summary.merge_all()\n",
    "\n",
    "init = tf.global_variables_initializer()\n",
    "saver = tf.train.Saver()\n",
    "\n",
    "train_writer = tf.summary.FileWriter(\"models/dnn0/train\", tf.get_default_graph())\n",
    "test_writer = tf.summary.FileWriter(\"models/dnn0/test\", tf.get_default_graph())\n",
    "\n",
    "num_epochs = 50\n",
    "batch_size = 128"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"color:red\">**(d)**</span> **You are now required write code to train the DNN.** Write codes in the following cell. <span style=\"color:red\">**[5 points]**</span> </div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-------+-------------------+------------------+\n",
      "| Epoch | Training Accuracy | Testing Accuracy |\n",
      "+-------+-------------------+------------------+\n",
      "|   1   |        0.1        |    0.25275898    |\n",
      "|   2   |        0.28       |    0.4013884     |\n",
      "|   3   |        0.48       |    0.51281595    |\n",
      "|   4   |        0.6        |    0.59202564    |\n",
      "|   5   |        0.7        |    0.6571734     |\n",
      "|   6   |        0.74       |    0.7013172     |\n",
      "|   7   |        0.78       |    0.73300105    |\n",
      "|   8   |        0.78       |    0.75934494    |\n",
      "|   9   |        0.78       |    0.7789249     |\n",
      "|   10  |        0.8        |    0.7903168     |\n",
      "|   11  |        0.8        |    0.7986828     |\n",
      "|   12  |        0.8        |    0.8054468     |\n",
      "|   13  |        0.8        |    0.81060874    |\n",
      "|   14  |        0.8        |    0.8170167     |\n",
      "|   15  |        0.8        |    0.81986475    |\n",
      "|   16  |        0.82       |    0.8250267     |\n",
      "|   17  |        0.82       |    0.8285867     |\n",
      "|   18  |        0.82       |    0.8301887     |\n",
      "|   19  |        0.82       |    0.83374864    |\n",
      "|   20  |        0.86       |    0.83588463    |\n",
      "|   21  |        0.88       |    0.83784264    |\n",
      "|   22  |        0.88       |    0.84051263    |\n",
      "|   23  |        0.88       |    0.84247065    |\n",
      "|   24  |        0.88       |    0.8438946     |\n",
      "|   25  |        0.88       |    0.8449626     |\n",
      "|   26  |        0.88       |    0.8463866     |\n",
      "|   27  |        0.88       |    0.8476326     |\n",
      "|   28  |        0.88       |    0.8485226     |\n",
      "|   29  |        0.88       |    0.8501246     |\n",
      "|   30  |        0.9        |    0.8510146     |\n",
      "|   31  |        0.9        |    0.8522606     |\n",
      "|   32  |        0.9        |    0.85297257    |\n",
      "|   33  |        0.9        |    0.85457456    |\n",
      "|   34  |        0.9        |    0.8552866     |\n",
      "|   35  |        0.9        |    0.85617656    |\n",
      "|   36  |        0.9        |    0.8568886     |\n",
      "|   37  |        0.9        |    0.8579566     |\n",
      "|   38  |        0.9        |    0.8584906     |\n",
      "|   39  |        0.9        |    0.85884655    |\n",
      "|   40  |        0.9        |    0.8595586     |\n",
      "|   41  |        0.9        |    0.85991454    |\n",
      "|   42  |        0.9        |    0.8600926     |\n",
      "|   43  |        0.9        |    0.8606266     |\n",
      "|   44  |        0.9        |    0.86080456    |\n",
      "|   45  |        0.9        |    0.8611606     |\n",
      "|   46  |        0.9        |    0.8622286     |\n",
      "|   47  |        0.9        |    0.86258453    |\n",
      "|   48  |        0.9        |    0.8627626     |\n",
      "|   49  |        0.9        |    0.8636525     |\n",
      "|   50  |        0.9        |    0.8641865     |\n",
      "+-------+-------------------+------------------+\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    \n",
    "    tbl = PrettyTable()\n",
    "    tbl.field_names = [\"Epoch\", \"Training Accuracy\", \"Testing Accuracy\"]\n",
    "    \n",
    "    for epoch in range(num_epochs):\n",
    "        for idx_start in range(0, X_train.shape[0], batch_size):\n",
    "            idx_end = (idx_start+batch_size)\n",
    "            x_batch, y_batch = X_train[idx_start:idx_end,:],y_train[idx_start:idx_end,]\n",
    "            sess.run(training_op, feed_dict={inputs: x_batch, labels: y_batch})\n",
    "            \n",
    "        summary_train, acc_train = sess.run([merged,accuracy],feed_dict={inputs: x_batch, labels: y_batch})\n",
    "        summary_test, acc_test = sess.run([merged,accuracy],feed_dict={inputs: X_test, labels: y_test})\n",
    "        \n",
    "        train_writer.add_summary(summary_train, epoch)\n",
    "        test_writer.add_summary(summary_test, epoch)\n",
    "        tbl.add_row([epoch+1, acc_train,acc_test])\n",
    "    print(tbl)\n",
    "    save_path = saver.save(sess, \"models/dnn0.ckpt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <span style=\"color:#0b486b\">Part 2: Word2Vec, Text Analytics and Application</span>\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[Total mark for this part: 35 points]**<span></div>\n",
    "\n",
    "\n",
    "In this part, you are going to use Word2Vec for document classification on [20 Newsgroups](http://www.cs.cmu.edu/afs/cs.cmu.edu/project/theo-20/www/data/news20.html) dataset. This dataset is a collection of messages collected from 20 different netnews newsgroups. One thousand messages from each of the twenty newsgroups were chosen at random and partitioned by newsgroup name. The list of newsgroups from which the messages were chosen is as follows:\n",
    "\n",
    "`\n",
    "alt.atheism\n",
    "talk.politics.guns\n",
    "talk.politics.mideast\n",
    "talk.politics.misc\n",
    "talk.religion.misc\n",
    "soc.religion.christian\n",
    "comp.sys.ibm.pc.hardware\n",
    "comp.graphics\n",
    "comp.os.ms-windows.misc\n",
    "comp.sys.mac.hardware\n",
    "comp.windows.x\n",
    "rec.autos\n",
    "rec.motorcycles\n",
    "rec.sport.baseball\n",
    "rec.sport.hockey\n",
    "sci.crypt\n",
    "sci.electronics\n",
    "sci.space\n",
    "sci.med\n",
    "misc.forsale\n",
    "`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <span style=\"color:#0b486b\"> Download the dataset and data pre-processing</span>\n",
    "\n",
    "####  <span style=\"color:red\">**Question 2.1**</span> Your first task is to run the following code to download the dataset.\n",
    "\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[1 point]**</span></div> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import fetch_20newsgroups\n",
    "newsgroups_all = fetch_20newsgroups(subset='all', remove=('headers'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 2.2**</span>. Print out the total number of documents, and the *unique* labels in this dataset.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[1 point]**</span></div> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['data', 'filenames', 'target_names', 'target', 'DESCR'])\n",
      "===========================================================================\n",
      "Total number of documents =  18846\n",
      "===========================================================================\n",
      "Unique labels are :  [ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19]\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "print(newsgroups_all.keys())\n",
    "printDash(75)\n",
    "print(\"Total number of documents = \",format(len(newsgroups_all.data)))\n",
    "printDash(75)\n",
    "y = newsgroups_all.target\n",
    "print(\"Unique labels are : \",format(np.unique(y)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 2.3**</span>. Convert the documents into a list of tokens using the function `gensim.utils.tokenize`.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[3 point]**</span></div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gensim as gs\n",
    "\n",
    "corpus_token=[]\n",
    "for i in range(len(newsgroups_all.data)):\n",
    "    #,lowercase=True,deacc=True\n",
    "    corpus_token.append(list(gs.utils.tokenize(newsgroups_all.data[i],lowercase=True,deacc=True)))\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <span style=\"color:#0b486b\"> Train the model </span>\n",
    "\n",
    "\n",
    "####  <span style=\"color:red\">**Question 2.4**</span>. Train gensim's word2vec model.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[5 points]**</span></div>\n",
    "\n",
    "* Train gensim's word2vec model with the settings of:\n",
    "    * The dimensionality of the feature vectors: `size=100`,\n",
    "    * The maximum distance between the current and predicted word within a sentence: `window=5`,\n",
    "    * Minimum frequence (ignore all words with total frequency lower than this): `min_count=5`,\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from gensim.models import Word2Vec\n",
    "\n",
    "w2v_model = Word2Vec(corpus_token, size=100, window=5, min_count=5, workers=8,sg=1) #sg is used for long text/ big documents.\n",
    "words = list(w2v_model.wv.vocab.keys())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Save the trained model to a file named \"20_newsgroups.gensim\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda3/lib/python3.7/site-packages/smart_open/smart_open_lib.py:398: UserWarning: This function is deprecated, use smart_open.open instead. See the migration notes for details: https://github.com/RaRe-Technologies/smart_open/blob/master/README.rst#migrating-to-the-new-open-function\n",
      "  'See the migration notes for details: %s' % _MIGRATION_NOTES_URL\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "model_name = 'models/20_newsgroups.gensim'\n",
    "w2v_model.save(model_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 2.5**</span>. Print out the vocabulary size (number of words in vocabulary).\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[2 points]**</span></div> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number of words =  34983\n",
      "The first 10 words in the vocabularies =  ['i', 'am', 'sure', 'some', 'bashers', 'of', 'pens', 'fans', 'are', 'pretty']\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "print(\"The number of words = \",format(len(words)))\n",
    "\n",
    "print(\"The first 10 words in the vocabularies = \",format(words[0:10]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 2.6**</span>. Using the embedding results, calculate and print out the ten most similar words to word 'law' and their corresponding similarity scores.<div style=\"text-align: right\"><span style=\"color:red\">**[3 points]**</span></div> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-------------------------+--------------------+\n",
      "| Words(similar to *Law*) |  Similarity score  |\n",
      "+-------------------------+--------------------+\n",
      "|       enforcement       | 0.8403921127319336 |\n",
      "|         abiding         | 0.7467506527900696 |\n",
      "|          giveth         | 0.6896644830703735 |\n",
      "|     unconstitutional    | 0.6697934865951538 |\n",
      "|           fca           | 0.6657330393791199 |\n",
      "|         menacing        | 0.6591196060180664 |\n",
      "|           laws          | 0.6582514047622681 |\n",
      "|          uphold         | 0.6499788761138916 |\n",
      "|          felony         | 0.6490593552589417 |\n",
      "|           ccw           | 0.6488798260688782 |\n",
      "+-------------------------+--------------------+\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "\n",
    "\n",
    "similar_words_list = w2v_model.wv.most_similar(positive = ['law'],topn=10)\n",
    "tbl = PrettyTable()\n",
    "tbl.field_names = [\"Words(similar to *Law*)\", \"Similarity score\"]\n",
    "\n",
    "for row in similar_words_list:\n",
    "    tbl.add_row([row[0],row[1]])\n",
    "    \n",
    "print(tbl)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <span style=\"color:#0b486b\"> Evaluate the embeddings using classification </span>\n",
    "\n",
    "Now we investigate the quality of embedded vectors via document classification task. We have learned the embeddings for words, but not for documents yet, thus we need to find a way to extract the document embeddings from word embeddings. We are going to try two approaches:\n",
    "* Taking the **sum** of vectors of all words in the document; or\n",
    "* Taking the **average** of vectors of all words in the document.\n",
    "\n",
    "####  <span style=\"color:red\">**Question 2.7**</span>. Extract document vectors using `sum`.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[5 points]**</span></div> \n",
    "\n",
    "* Remove all ***empty*** documents. A document is empty if it does not contain any word in the vocabulary;\n",
    "* Extract document vectors and save to variable `x`;\n",
    "* Save the corresponding labels to variable `y`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:3: DeprecationWarning: Call to deprecated `__contains__` (Method will be removed in 4.0.0, use self.wv.__contains__() instead).\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n",
      "//anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:3: DeprecationWarning: Call to deprecated `__getitem__` (Method will be removed in 4.0.0, use self.wv.__getitem__() instead).\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(18809,)\n",
      "(18809,)\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "def extract_doc2vec(doc, w2v,operator='sum'):\n",
    "    vecs = [w2v[word] for word in doc if word in w2v]\n",
    "    if len(vecs) > 0:\n",
    "        if (operator == 'avg'): #modified for question 2.10\n",
    "            vecs = np.asarray(vecs).mean(axis=0)\n",
    "        else:\n",
    "            vecs = np.asarray(vecs).sum(axis=0)\n",
    "        \n",
    "    return vecs\n",
    "\n",
    "\n",
    "#returns the index of empty documents so that it can be removed\n",
    "\n",
    "def remove_empty_docs(corpus,labels):\n",
    "    idx = []\n",
    "    for ix, i in enumerate(corpus):\n",
    "        if len(i) == 0:\n",
    "            #print(ix)\n",
    "            idx.append(ix)\n",
    "\n",
    "    corpus = np.delete(corpus,idx,0)\n",
    "    labels = np.delete(labels,idx,0)\n",
    "\n",
    "    return corpus,labels\n",
    "\n",
    "\n",
    "vec_x_sum=[extract_doc2vec(t,w2v_model) for t in corpus_token]\n",
    "\n",
    "\n",
    "vec_x_sum, y_filtered = remove_empty_docs(vec_x_sum,y)\n",
    "print(vec_x_sum.shape)\n",
    "print(y_filtered.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 2.8**</span>. Print out the number of documents retained after removing empty documents.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[1 point]**</span></div> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of documents after removing empty =  18809\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "\n",
    "print(\"Number of documents after removing empty = \",format(len(vec_x_sum)))\n",
    "\n",
    "\n",
    "y_filtered = np.array(y_filtered)\n",
    "vec_x_sum = np.array(vec_x_sum)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### <span style=\"color:red\">**Question 2.9**</span>. Split the data into two subsets: 70% for training and 30% for testing. Note that you must use [*Stratified-Shuffle-Split*](http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.StratifiedShuffleSplit.html) to make sure training and testing are balanced and randomly shuffled before learning the model.\n",
    "<div style=\"text-align: right\"> <span style=\"color:red\">**[2 points]**</span> </div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(13166,) (5643,)\n",
      "(100,)\n"
     ]
    }
   ],
   "source": [
    "# YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "\n",
    "ssts = StratifiedShuffleSplit(n_splits=1,test_size=0.3, random_state=999)\n",
    "\n",
    "for train_index, test_index in ssts.split(vec_x_sum,y_filtered):\n",
    "    y_train, y_test = y_filtered[train_index], y_filtered[test_index]\n",
    "    X_train, X_test = vec_x_sum[train_index], vec_x_sum[test_index]\n",
    "    \n",
    "print(X_train.shape,y_test.shape)\n",
    "pprint(X_train[1].shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### <span style=\"color:red\">**Question 2.10**</span>. **Use [Logistic Regression](http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) as the classifier, train and test the model using the training and test datasets from the previous step. Report the training accuracy and testing accuracy.**\n",
    "<div style=\"text-align: right\"> <span style=\"color:red\">**[2 points]**</span> </div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of null/empty data (0, 100)\n"
     ]
    }
   ],
   "source": [
    "#cleaning data\n",
    "\n",
    "x_train = pd.DataFrame(list(X_train))\n",
    "x_test  = pd.DataFrame(list(X_test))\n",
    "null_data = x_train[x_train.isnull().any(axis=1)]\n",
    "\n",
    "x_train.fillna(x_train.mean(),inplace=True)\n",
    "\n",
    "print(\"Shape of null/empty data\",format(null_data.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+----------+--------------------+\n",
      "| Accuracy |       Value        |\n",
      "+----------+--------------------+\n",
      "| Training | 0.7662919641500835 |\n",
      "| Testing  | 0.7185894027999291 |\n",
      "+----------+--------------------+\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:947: ConvergenceWarning: lbfgs failed to converge. Increase the number of iterations.\n",
      "  \"of iterations.\", ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "##### YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "from sklearn.linear_model import LogisticRegression #as LogReg\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "\n",
    "mx_iter = 99^999\n",
    "logReg = LogisticRegression(solver='lbfgs',max_iter=mx_iter,multi_class='multinomial')\n",
    "logReg.fit(x_train,y_train)\n",
    "\n",
    "tbl = PrettyTable()\n",
    "tbl.field_names = [\"Accuracy\", \"Value\"]\n",
    "\n",
    "\n",
    "y_pred=logReg.predict(x_train)\n",
    "train_accuracy_sum=accuracy_score(y_train,y_pred)\n",
    "tbl.add_row(['Training',train_accuracy_sum])\n",
    "#print(\"Train Accuracy =\",format(train_accuracy))\n",
    "\n",
    "y_pred=logReg.predict(x_test)\n",
    "test_accuracy_sum=accuracy_score(y_test,y_pred)\n",
    "tbl.add_row(['Testing',test_accuracy_sum])\n",
    "#print (\"Test Accuracy =\",format(test_accuracy))\n",
    "'''\n",
    "train_accuracy=cross_val_score(logReg,x_train,y_train,scoring='accuracy', verbose=1, n_jobs=-1,cv=5)\n",
    "#print(\"Training accuracy =\",format(np.mean(train_accuracy)))\n",
    "tbl.add_row(['Training - CV',np.mean(train_accuracy)])\n",
    "\n",
    "test_accuracy=cross_val_score(logReg,x_test,y_test, scoring='accuracy', verbose=1, n_jobs=-1,cv=5)\n",
    "#print(\"Testing accuracy =\",format(np.mean(test_accuracy)))\n",
    "tbl.add_row(['Testing - CV',np.mean(test_accuracy)])\n",
    "'''\n",
    "\n",
    "print(tbl)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  <span style=\"color:red\">**Question 2.11**</span>. Now modify the `extract_doc2vec` function above to  extract document vectors using `average`, instead of `sum`, and repeat the experiment: split the data, train and test using Logistic Regression.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[5 points]**</span></div> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:3: DeprecationWarning: Call to deprecated `__contains__` (Method will be removed in 4.0.0, use self.wv.__contains__() instead).\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n",
      "//anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:3: DeprecationWarning: Call to deprecated `__getitem__` (Method will be removed in 4.0.0, use self.wv.__getitem__() instead).\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+----------+--------------------+\n",
      "| Accuracy |       Value        |\n",
      "+----------+--------------------+\n",
      "| Training | 0.7304420476986176 |\n",
      "| Testing  | 0.7136275031011873 |\n",
      "+----------+--------------------+\n"
     ]
    }
   ],
   "source": [
    "##### YOU ARE REQUIRED TO INSERT YOUR CODES IN THIS CELL\n",
    "\n",
    "vec_x_avg=[extract_doc2vec(t,w2v_model,'avg') for t in corpus_token]\n",
    "vec_x_avg , y_filtered = remove_empty_docs(vec_x_avg,y)\n",
    "vec_x_avg = np.array(vec_x_avg)\n",
    "\n",
    "\n",
    "ssts = StratifiedShuffleSplit(n_splits=1,test_size=0.3, random_state=999)\n",
    "\n",
    "for train_index, test_index in ssts.split(vec_x_avg,y_filtered):\n",
    "    y_train, y_test = y_filtered[train_index], y_filtered[test_index]\n",
    "    X_train, X_test = vec_x_avg[train_index], vec_x_avg[test_index]\n",
    "    \n",
    "\n",
    "x_train = pd.DataFrame(list(X_train))\n",
    "x_test  = pd.DataFrame(list(X_test))\n",
    "null_data = x_train[x_train.isnull().any(axis=1)]\n",
    "\n",
    "x_train.fillna(x_train.mean(),inplace=True)\n",
    "\n",
    "\n",
    "lr = LogisticRegression(solver='lbfgs',max_iter=mx_iter,multi_class='multinomial')\n",
    "lr.fit(x_train,y_train)\n",
    "\n",
    "tbl = PrettyTable()\n",
    "tbl.field_names = [\"Accuracy\", \"Value\"]\n",
    "\n",
    "y_pred=lr.predict(x_train)\n",
    "train_accuracy_avg = accuracy_score(y_train,y_pred)\n",
    "tbl.add_row(['Training',train_accuracy_avg])\n",
    "#print(\"Train Accuracy =\",format(train_accuracy))\n",
    "\n",
    "y_pred=lr.predict(x_test)\n",
    "test_accuracy_avg=accuracy_score(y_test,y_pred)\n",
    "tbl.add_row(['Testing',test_accuracy_avg])\n",
    "#print (\"Test Accuracy =\",format(test_accuracy))\n",
    "\n",
    "'''\n",
    "train_accuracy=cross_val_score(lr,x_train,y_train,scoring='accuracy', verbose=1, n_jobs=-1,cv=5)\n",
    "#print(\"Training accuracy =\",format(np.mean(train_accuracy)))\n",
    "tbl.add_row(['Training - CV',np.mean(train_accuracy)])\n",
    "\n",
    "test_accuracy=cross_val_score(lr,x_test,y_test, scoring='accuracy', verbose=1, n_jobs=-1,cv=5)\n",
    "#print(\"Testing accuracy =\",format(np.mean(test_accuracy)))\n",
    "tbl.add_row(['Testing - CV',np.mean(test_accuracy)])\n",
    "'''\n",
    "\n",
    "print(tbl)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "####  <span style=\"color:red\">**Question 2.12**</span>. Which representation (sum or average doc vector) gives the best performance? Write your observations and any lessons learned.\n",
    "<div style=\"text-align: right\"><span style=\"color:red\">**[5 points]**</span></div> \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vector represented using *SUM* gives the best performance. \n",
    "\n",
    "\n",
    "|*Vector*|*Train*|*Test*|\n",
    "|:-------------|:-------------|:-----|\n",
    "|**SUM**|*0.7662919641500835*|*0.7185894027999291*|\n",
    "|Average|0.7304420476986176|0.7136275031011873|\n",
    "\n",
    "As the vector calculated using the sum, actually sums the whole length and then perform operations unlike, *average* which just considers the average and base the result on that. In our case, the sample size is less and hence the difference cannot be seen much, but it cannot be used in production."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "<div style=\"text-align: center\"> <span style=\"color:black\">**END OF ASSIGNMENT**</span> </div>"
   ]
  }
 ],
 "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": 4
}