Unsupervised Learning with Python.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'points' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-2-3cb54bd3c6bf>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     31\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     32\u001b[0m \u001b[1;31m# Fit model to points\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 33\u001b[1;33m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mpoints\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     34\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     35\u001b[0m \u001b[1;31m# Determine the cluster labels of new_points: labels\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'points' is not defined"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Among well-known unsupervised learning techniques : \n",
    "- t-SNE (which also serves dimension reduction purposes)\n",
    "- k-Means\n",
    "- hierarchical clustering\n",
    "\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "How many clusters?\n",
    "You are given an array points of size 300x2, where each row gives the (x, y) co-ordinates of a point on a map.\n",
    "Make a scatter plot of these points, and use the scatter plot to guess how many clusters there are.\n",
    "\n",
    "matplotlib.pyplot has already been imported as plt. In the IPython Shell:\n",
    "\n",
    "Create an array called xs that contains the values of points[:,0] - that is, column 0 of points.\n",
    "Create an array called ys that contains the values of points[:,1] - that is, column 1 of points.\n",
    "Make a scatter plot by passing xs and ys to the plt.scatter() function.\n",
    "Call the plt.show() function to show your plot.\n",
    "How many clusters do you see? 3\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Clustering 2D points\n",
    "From the scatter plot of the previous exercise, you saw that the points seem to separate into 3 clusters.\n",
    "You'll now create a KMeans model to find 3 clusters, and fit it to the data points from the previous exercise.\n",
    "After the model has been fit, you'll obtain the cluster labels for some new points using the .predict() method.\n",
    "\n",
    "You are given the array points from the previous exercise, and also an array new_points.\n",
    "\"\"\"\n",
    "\n",
    "# Import KMeans\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "# Create a KMeans instance with 3 clusters: model\n",
    "model = KMeans(n_clusters=3)\n",
    "\n",
    "# Fit model to points\n",
    "model.fit(points)\n",
    "\n",
    "# Determine the cluster labels of new_points: labels\n",
    "labels = model.predict(new_points)\n",
    "\n",
    "# Print cluster labels of new_points\n",
    "print(labels)\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "You've successfully performed k-Means clustering and predicted the labels of new points. \n",
    "But it is not easy to inspect the clustering by just looking at the printed labels. A visualization\n",
    "would be far more useful. In the next exercise, you'll inspect your clustering with a scatter plot!\n",
    "\"\"\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 2 0 2 1 2 2 2 1 2]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'\\nData is random, so results is ugly.\\n'"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Inspect your clustering\n",
    "Let's now inspect the clustering you performed in the previous exercise!\n",
    "A solution to the previous exercise has already run, so new_points is an array of points and labels is the\n",
    "array of their cluster labels.\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Data is random, so results is ugly.\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "import numpy as np\n",
    "import random\n",
    "\n",
    "n_col = 2\n",
    "n_row = 10\n",
    "\n",
    "array1 = [[random.random() for i in range(n_col)] for j in range(n_row)]\n",
    "array_col1 = [row[1] for row in array]\n",
    "array_col0 = [row[0] for row in array]\n",
    "\n",
    "array2 = [[random.random() for i in range(n_col)] for j in range(n_row)]\n",
    "\n",
    "# Import KMeans\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "# Create a KMeans instance with 3 clusters: model\n",
    "model = KMeans(n_clusters=3)\n",
    "\n",
    "# Fit model to points\n",
    "model.fit(array1)\n",
    "\n",
    "# Determine the cluster labels of new_points: labels\n",
    "labels = model.predict(array2)\n",
    "print(labels)\n",
    "\n",
    "\n",
    "\n",
    "# Import pyplot\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Assign the columns of new_points: xs and ys\n",
    "# xs = array_col0[:, 0]\n",
    "# ys = array_col1[:, 1]\n",
    "\n",
    "xs = array_col0\n",
    "ys = array_col1\n",
    "\n",
    "# Make a scatter plot of xs and ys, using labels to define the colors\n",
    "plt.scatter(x=xs , y=ys, c=labels, alpha=.5)\n",
    "\n",
    "# Assign the cluster centers: centroids\n",
    "centroids = model.cluster_centers_\n",
    "\n",
    "# Assign the columns of centroids: centroids_x, centroids_y\n",
    "centroids_x = centroids[:,0]\n",
    "centroids_y = centroids[:,1]\n",
    "\n",
    "# Make a scatter plot of centroids_x and centroids_y\n",
    "plt.scatter(x=centroids_x , y=centroids_y, marker='D', s=50)\n",
    "plt.show()\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Data is random, so results is ugly.\n",
    "\n",
    "The clustering looks great! But how can you be sure that 3 clusters is the correct choice? In other words,\n",
    "how can you evaluate the quality of a clustering? Tune into the next video in which Ben will explain how to\n",
    "evaluate a clustering!\n",
    "\"\"\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   15.26  14.84   0.871  5.763  3.312  2.221   5.22  1\n",
      "0  14.88  14.57  0.8811  5.554  3.333  1.018  4.956  1\n",
      "1  14.29  14.09  0.9050  5.291  3.337  2.699  4.825  1\n",
      "2  13.84  13.94  0.8955  5.324  3.379  2.259  4.805  1\n",
      "3  16.14  14.99  0.9034  5.658  3.562  1.355  5.175  1\n",
      "4  14.38  14.21  0.8951  5.386  3.312  2.462  4.956  1\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Evaluating a clustering\n",
    "\n",
    "Cross-tabulation : make sure a cluster corresponds correctly to a data distribution\n",
    "Measure the quality of clustering.\n",
    "Inertia : how far points are to the centroids (the lower the better)\n",
    "How many clusters to pick ? > Elbow plot\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "How many clusters of grain?\n",
    "In the video, you learned how to choose a good number of clusters for a dataset using the k-means inertia graph.\n",
    "You are given an array samples containing the measurements (such as area, perimeter, length, and several others) \n",
    "of samples of grain. What's a good number of clusters in this case?\n",
    "\n",
    "KMeans and PyPlot (plt) have already been imported for you.\n",
    "\"\"\"\n",
    "\n",
    "import sys, os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains/'\n",
    "os.chdir(path)\n",
    "\n",
    "samples = pd.read_csv('seeds.csv')\n",
    "print(samples.head())\n",
    "\n",
    "ks = range(1, 6)\n",
    "inertias = []\n",
    "\n",
    "for k in ks:\n",
    "    # Create a KMeans instance with k clusters: model\n",
    "    model = KMeans(n_clusters = k)\n",
    "    \n",
    "    # Fit model to samples\n",
    "    model.fit(samples)\n",
    "    \n",
    "    # Append the inertia to the list of inertias\n",
    "    inertias.append(model.inertia_)\n",
    "    \n",
    "# Plot ks vs inertias\n",
    "plt.plot(ks, inertias, '-o')\n",
    "plt.xlabel('number of clusters, k')\n",
    "plt.ylabel('inertia')\n",
    "plt.xticks(ks)\n",
    "plt.show()\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "The inertia decreases very slowly from 3 clusters to 4, so it looks like 3 clusters would be a good choice for this data.\n",
    "\"\"\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "209\n",
      "209\n",
      "varieties  Canadian wheat  Kama wheat  Rosa wheat\n",
      "labels                                           \n",
      "0                       0          63          10\n",
      "1                      70           5           0\n",
      "2                       0           1          60\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Program Files\\Anaconda3\\lib\\site-packages\\sklearn\\preprocessing\\data.py:645: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "C:\\Program Files\\Anaconda3\\lib\\site-packages\\sklearn\\base.py:464: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.fit(X, **fit_params).transform(X)\n",
      "C:\\Program Files\\Anaconda3\\lib\\site-packages\\sklearn\\pipeline.py:331: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  Xt = transform.transform(Xt)\n"
     ]
    },
    {
     "ename": "NameError",
     "evalue": "name 'species' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-70-726c30e753f5>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     55\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     56\u001b[0m \u001b[1;31m# Create a DataFrame with labels and species as columns: df\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 57\u001b[1;33m \u001b[0mdf\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDataFrame\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m{\u001b[0m\u001b[1;34m'labels'\u001b[0m \u001b[1;33m:\u001b[0m \u001b[0mlabels\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'species'\u001b[0m \u001b[1;33m:\u001b[0m \u001b[0mspecies\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     58\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     59\u001b[0m \u001b[1;31m# Create crosstab: ct\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'species' is not defined"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Evaluating the grain clustering\n",
    "In the previous exercise, you observed from the inertia plot that 3 is a good number of clusters for the grain data.\n",
    "In fact, the grain samples come from a mix of 3 different grain varieties: \"Kama\", \"Rosa\" and \"Canadian\". \n",
    "In this exercise, cluster the grain samples into three clusters, and compare the clusters to the grain varieties\n",
    "using a cross-tabulation.\n",
    "\n",
    "You have the array samples of grain samples, and a list varieties giving the grain variety for each sample. \n",
    "Pandas (pd) and KMeans have already been imported for you.\n",
    "\"\"\"\n",
    "\n",
    "import sys, os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains/'\n",
    "os.chdir(path)\n",
    "samples = pd.read_csv('seeds.csv')\n",
    "varieties = ['Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat']\n",
    "\n",
    "print(len(samples))\n",
    "print(len(varieties))\n",
    "\n",
    "# Create a KMeans model with 3 clusters: model\n",
    "model = KMeans(n_clusters=3)\n",
    "\n",
    "# Use fit_predict to fit model and obtain cluster labels: labels\n",
    "labels = model.fit_predict(samples)\n",
    "\n",
    "# Create a DataFrame with labels and varieties as columns: df\n",
    "df = pd.DataFrame({'labels': labels, 'varieties': varieties})\n",
    "\n",
    "# Create crosstab: ct\n",
    "ct = pd.crosstab(df['labels'], df['varieties'])\n",
    "\n",
    "# Display ct\n",
    "print(ct)\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "The cross-tabulation shows that the 3 varieties of grain separate really well into 3 clusters. \n",
    "But depending on the type of data you are working with, the clustering may not always be this good.\n",
    "Is there anything you can do in such situations to improve your clustering? You'll find out in the next video!\n",
    "\"\"\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "species  Bream  Pike  Roach  Smelt\n",
      "labels                            \n",
      "0           10     7      4      6\n",
      "1            6     2      3      2\n",
      "2            7     2      5      2\n",
      "3           11     6      8      4\n",
      "    labels                           companies\n",
      "44       0                        Schlumberger\n",
      "42       0                   Royal Dutch Shell\n",
      "57       0                               Exxon\n",
      "49       0                               Total\n",
      "10       0                      ConocoPhillips\n",
      "12       0                             Chevron\n",
      "53       0                       Valero Energy\n",
      "35       1                            Navistar\n",
      "45       1                                Sony\n",
      "48       1                              Toyota\n",
      "22       1                                  HP\n",
      "21       1                               Honda\n",
      "34       1                          Mitsubishi\n",
      "58       1                               Xerox\n",
      "15       1                                Ford\n",
      "11       1                               Cisco\n",
      "7        1                               Canon\n",
      "2        1                              Amazon\n",
      "43       1                                 SAP\n",
      "39       2                              Pfizer\n",
      "56       2                            Wal-Mart\n",
      "25       2                   Johnson & Johnson\n",
      "54       2                            Walgreen\n",
      "16       3                   General Electrics\n",
      "55       3                         Wells Fargo\n",
      "18       3                       Goldman Sachs\n",
      "5        3                     Bank of America\n",
      "3        3                    American express\n",
      "26       3                      JPMorgan Chase\n",
      "1        3                                 AIG\n",
      "36       4                    Northrop Grumman\n",
      "29       4                     Lookheed Martin\n",
      "4        4                              Boeing\n",
      "50       5  Taiwan Semiconductor Manufacturing\n",
      "51       5                   Texas instruments\n",
      "47       6                            Symantec\n",
      "59       6                               Yahoo\n",
      "23       6                                 IBM\n",
      "32       6                                  3M\n",
      "8        6                         Caterpillar\n",
      "13       6                   DuPont de Nemours\n",
      "31       6                           McDonalds\n",
      "30       6                          MasterCard\n",
      "14       6                                Dell\n",
      "24       6                               Intel\n",
      "33       6                           Microsoft\n",
      "20       6                          Home Depot\n",
      "6        7            British American Tobacco\n",
      "52       7                            Unilever\n",
      "19       7                     GlaxoSmithKline\n",
      "37       7                            Novartis\n",
      "41       7                       Philip Morris\n",
      "46       7                      Sanofi-Aventis\n",
      "27       8                      Kimberly-Clark\n",
      "28       8                           Coca Cola\n",
      "9        8                   Colgate-Palmolive\n",
      "40       8                      Procter Gamble\n",
      "38       8                               Pepsi\n",
      "17       9                     Google/Alphabet\n",
      "0        9                               Apple\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Program Files\\Anaconda3\\lib\\site-packages\\sklearn\\preprocessing\\data.py:645: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "C:\\Program Files\\Anaconda3\\lib\\site-packages\\sklearn\\base.py:464: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.fit(X, **fit_params).transform(X)\n",
      "C:\\Program Files\\Anaconda3\\lib\\site-packages\\sklearn\\pipeline.py:331: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  Xt = transform.transform(Xt)\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Transforming features for better clustering\n",
    "\n",
    "Scaling fish data for clustering\n",
    "You are given an array samples giving measurements of fish. Each row represents an individual fish. \n",
    "The measurements, such as weight in grams, length in centimeters, and the percentage ratio of height to length,\n",
    "have very different scales. In order to cluster this data effectively, you'll need to standardize these features first.\n",
    "In this exercise, you'll build a pipeline to standardize and cluster the data.\n",
    "\"\"\"\n",
    "\n",
    "import sys, os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/'\n",
    "os.chdir(path)\n",
    "\n",
    "wine = pd.read_csv('wine.csv')\n",
    "\n",
    "fish = pd.read_csv('fish.csv')\n",
    "\n",
    "# Perform the necessary imports\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "# Create scaler: scaler\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Create KMeans instance: kmeans\n",
    "kmeans = KMeans(n_clusters = 4)\n",
    "\n",
    "# Create pipeline: pipeline\n",
    "pipeline = make_pipeline(scaler, kmeans)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Clustering the fish data\n",
    "You'll now use your standardization and clustering pipeline from the previous exercise to cluster the fish by their\n",
    "measurements, and then create a cross-tabulation to compare the cluster labels with the fish species.\n",
    "\n",
    "As before, samples is the 2D array of fish measurements. Your pipeline is available as pipeline, and the species\n",
    "of every fish sample is given by the list species.\n",
    "\"\"\"\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains/'\n",
    "os.chdir(path)\n",
    "\n",
    "samples = pd.read_csv('seeds.csv')\n",
    "species = ['Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Bream', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Roach', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Smelt', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike', 'Pike']\n",
    "\n",
    "np.random.seed(42)\n",
    "row_sample = np.random.choice(samples.index, 85, replace = False)\n",
    "samples = samples.iloc[row_sample, :]\n",
    "\n",
    "\n",
    "# samples = samples.iloc[samples.index.isin(row_sample), :]\n",
    "# print(samples)\n",
    "\n",
    "# Import pandas\n",
    "import pandas as pd\n",
    "\n",
    "# Fit the pipeline to samples\n",
    "pipeline.fit(samples)\n",
    "\n",
    "# Calculate the cluster labels: labels\n",
    "labels = pipeline.predict(samples)\n",
    "\n",
    "# Create a DataFrame with labels and species as columns: df\n",
    "df = pd.DataFrame({'labels' : labels, 'species' : species})\n",
    "\n",
    "# Create crosstab: ct\n",
    "ct = pd.crosstab(df['labels'], df['species'])\n",
    "\n",
    "# Display ct\n",
    "print(ct)\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "It looks like the fish data separates really well into 4 clusters!\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Clustering stocks using KMeans\n",
    "In this exercise, you'll cluster companies using their daily stock price movements (i.e. the dollar difference\n",
    "between the closing and opening prices for each trading day). You are given a NumPy array movements of daily \n",
    "price movements from 2010 to 2015 (obtained from Yahoo! Finance), where each row corresponds to a company, and \n",
    "each column corresponds to a trading day.\n",
    "\n",
    "Some stocks are more expensive than others. To account for this, include a Normalizer at the beginning of your\n",
    "pipeline. The Normalizer will separately transform each company's stock price to a relative scale before the \n",
    "clustering begins.\n",
    "\n",
    "Note that Normalizer() is different to StandardScaler(), which you used in the previous exercise. While StandardScaler()\n",
    "standardizes features (such as the features of the fish data from the previous exercise) by removing the mean and \n",
    "scaling to unit variance, Normalizer() rescales each sample - here, each company's stock price - independently of\n",
    "the other.\n",
    "\n",
    "KMeans and make_pipeline have already been imported for you.\n",
    "\"\"\"\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/'\n",
    "os.chdir(path)\n",
    "mov = pd.read_csv('company-stock-movements-2010-2015-incl.csv')\n",
    "\n",
    "exclude = ['Unnamed: 0']\n",
    "movements = mov.drop(exclude, axis=1)\n",
    "\n",
    "# Import Normalizer\n",
    "from sklearn.preprocessing import Normalizer\n",
    "\n",
    "# Create a normalizer: normalizer\n",
    "normalizer = Normalizer()\n",
    "\n",
    "# Create a KMeans model with 10 clusters: kmeans\n",
    "kmeans = KMeans(n_clusters = 10)\n",
    "\n",
    "# Make a pipeline chaining normalizer and kmeans: pipeline\n",
    "pipeline = make_pipeline(normalizer, kmeans)\n",
    "\n",
    "# Fit pipeline to the daily price movements\n",
    "pipeline.fit(movements)\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Which stocks move together?\n",
    "In the previous exercise, you clustered companies by their daily stock price movements. So which company have\n",
    "stock prices that tend to change in the same way? You'll now inspect the cluster labels from your clustering\n",
    "to find out.\n",
    "\n",
    "Your solution to the previous exercise has already been run. Recall that you constructed a Pipeline pipeline\n",
    "containing a KMeans model and fit it to the NumPy array movements of daily stock movements. In addition, a list\n",
    "companies of the company names is available.\n",
    "\"\"\"\n",
    "\n",
    "# Import pandas\n",
    "import pandas as pd\n",
    "\n",
    "companies = mov.iloc[:, 0]\n",
    "\n",
    "# Predict the cluster labels: labels\n",
    "labels = pipeline.predict(movements)\n",
    "\n",
    "# Create a DataFrame aligning labels and companies: df\n",
    "df = pd.DataFrame({'labels': labels, 'companies': companies})\n",
    "\n",
    "# Display df sorted by cluster label\n",
    "print(df.sort_values('labels'))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x1200 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "\"\\nYou can produce great visualizations such as this with hierarchical clustering, but it can be used for more than\\njust visualizations. You'll find out more about this in the next video!\\n\""
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Hierarchical clustering of the grain data\n",
    "In the video, you learned that the SciPy linkage() function performs hierarchical clustering on an array of samples.\n",
    "Use the linkage() function to obtain a hierarchical clustering of the grain samples, and use dendrogram() to visualize\n",
    "the result. A sample of the grain measurements is provided in the array samples, while the variety of each grain sample\n",
    "is given by the list varieties.\n",
    "\"\"\"\n",
    "\n",
    "import os, sys\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/'\n",
    "# os.chdir(path)\n",
    "# eurovision = pd.read_csv('eurovision-2016.csv')\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains/'\n",
    "os.chdir(path)\n",
    "samples = pd.read_csv('seeds.csv')\n",
    "varieties = ['Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat']\n",
    "\n",
    "\n",
    "# Perform the necessary imports\n",
    "from scipy.cluster.hierarchy import linkage, dendrogram\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Calculate the linkage: mergings\n",
    "mergings = linkage(samples, method='complete')\n",
    "\n",
    "# Plot the dendrogram, using varieties as labels\n",
    "dendrogram(mergings,\n",
    "           labels=varieties,\n",
    "           leaf_rotation=90,\n",
    "           leaf_font_size=6,\n",
    ")\n",
    "plt.show()\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Hierarchies of stocks\n",
    "In chapter 1, you used k-means clustering to cluster companies according to their stock price movements. \n",
    "Now, you'll perform hierarchical clustering of the companies. You are given a NumPy array of price movements\n",
    "movements, where the rows correspond to companies, and a list of the company names companies. SciPy hierarchical\n",
    "clustering doesn't fit into a sklearn pipeline, so you'll need to use the normalize() function from sklearn.preprocessing\n",
    "instead of Normalizer.\n",
    "\n",
    "linkage and dendrogram have already been imported from scipy.cluster.hierarchy, and PyPlot has been imported as plt.\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\n",
    "# Import normalize\n",
    "from sklearn.preprocessing import normalize\n",
    "from matplotlib.pyplot import figure\n",
    "figure(figsize=(15, 15), dpi=80)\n",
    "\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/'\n",
    "os.chdir(path)\n",
    "mov = pd.read_csv('company-stock-movements-2010-2015-incl.csv')\n",
    "companies = mov.iloc[:, 0]\n",
    "\n",
    "exclude = ['Unnamed: 0']\n",
    "movements = mov.drop(exclude, axis=1)\n",
    "\n",
    "# Normalize the movements: normalized_movements\n",
    "normalized_movements = normalize(movements)\n",
    "\n",
    "# Calculate the linkage: mergings\n",
    "mergings = linkage(normalized_movements, method='complete')\n",
    "\n",
    "# Plot the dendrogram\n",
    "dendrogram(mergings,\n",
    "           #labels=companies,\n",
    "           leaf_rotation=90,\n",
    "           leaf_font_size=2,\n",
    ")\n",
    "plt.show()\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "You can produce great visualizations such as this with hierarchical clustering, but it can be used for more than\n",
    "just visualizations. You'll find out more about this in the next video!\n",
    "\"\"\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'numpy.ndarray'>\n",
      "['Albania' 'Albania' 'Albania' ... 'United Kingdom' 'United Kingdom'\n",
      " 'United Kingdom']\n",
      "42\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'pandas.core.frame.DataFrame'>\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Which clusters are closest?\n",
    "In the video, you learned that the linkage method defines how the distance between clusters is measured. \n",
    "In complete linkage, the distance between clusters is the distance between the furthest points of the clusters.\n",
    "In single linkage, the distance between clusters is the distance between the closest points of the clusters.\n",
    "\n",
    "Consider the three clusters in the diagram. Which of the following statements are true?\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Different linkage, different hierarchical clustering!\n",
    "In the video, you saw a hierarchical clustering of the voting countries at the Eurovision song contest using\n",
    "'complete' linkage. Now, perform a hierarchical clustering of the voting countries with 'single' linkage, and\n",
    "compare the resulting dendrogram with the one in the video. Different linkage, different hierarchical clustering!\n",
    "\n",
    "You are given an array samples. Each row corresponds to a voting country, and each column corresponds to a \n",
    "performance that was voted for. The list country_names gives the name of each voting country. This dataset was\n",
    "obtained from Eurovision.\n",
    "\"\"\"\n",
    "\n",
    "# Perform the necessary imports\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import os, sys\n",
    "import numpy as np\n",
    "from scipy.cluster.hierarchy import linkage, dendrogram\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/'\n",
    "os.chdir(path)\n",
    "\n",
    "dataset = pd.read_csv('eurovision-2016.csv')\n",
    "country_names = dataset.iloc[:, 0].values\n",
    "print(type(country_names))\n",
    "print(country_names)\n",
    "\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains'\n",
    "os.chdir(path)\n",
    "seeds = pd.read_csv('seeds.csv')\n",
    "\n",
    "n_rows_rand = np.random.choice(seeds.index, size = 42)\n",
    "print(len(n_rows_rand))\n",
    "\n",
    "print(type(n_rows_rand))\n",
    "print(type(seeds))\n",
    "\n",
    "samples = seeds.iloc[n_rows_rand, :]\n",
    "country_names = ['Albania', 'Armenia', 'Australia', 'Austria', 'Azerbaijan', 'Belarus', 'Belgium', 'Bosnia & Herzegovina', 'Bulgaria', 'Croatia', 'Cyprus', 'Czech Republic', 'Denmark', 'Estonia', 'F.Y.R. Macedonia', 'Finland', 'France', 'Georgia', 'Germany', 'Greece', 'Hungary', 'Iceland', 'Ireland', 'Israel', 'Italy', 'Latvia', 'Lithuania', 'Malta', 'Moldova', 'Montenegro', 'Norway', 'Poland', 'Russia', 'San Marino', 'Serbia', 'Slovenia', 'Spain', 'Sweden', 'Switzerland', 'The Netherlands', 'Ukraine', 'United Kingdom']\n",
    "\n",
    "# Calculate the linkage: mergings\n",
    "mergings = linkage(samples, method='single')\n",
    "\n",
    "# Plot the dendrogram\n",
    "dendrogram(\n",
    "    mergings,\n",
    "    leaf_rotation=90,\n",
    "    leaf_font_size=6,\n",
    "    labels=country_names\n",
    ")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41 42\n",
      "varieties  Canadian wheat  Kama wheat  Rosa wheat\n",
      "labels                                           \n",
      "1                      14          14          14\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "\"\\nyou've now mastered the fundamentals of k-Means and agglomerative hierarchical clustering. \\nNext, you'll learn about t-SNE, which is a powerful tool for visualizing high dimensional data.\\n\""
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Extracting the cluster labels\n",
    "In the previous exercise, you saw that the intermediate clustering of the grain samples at height 6 has 3 clusters.\n",
    "Now, use the fcluster() function to extract the cluster labels for this intermediate clustering, and compare the\n",
    "labels with the grain varieties using a cross-tabulation.\n",
    "\n",
    "The hierarchical clustering has already been performed and mergings is the result of the linkage() function.\n",
    "The list varieties gives the variety of each grain sample.\n",
    "\"\"\"\n",
    "\n",
    "# Perform the necessary imports\n",
    "import pandas as pd\n",
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.cluster.hierarchy import fcluster\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains'\n",
    "os.chdir(path)\n",
    "\n",
    "seeds = pd.read_csv('seeds.csv')\n",
    "n_rows_rand = np.random.choice(seeds.index, size = 42)\n",
    "samples = seeds.iloc[n_rows_rand, :]\n",
    "\n",
    "# Calculate the linkage: mergings\n",
    "mergings = linkage(samples, method='single')\n",
    "varieties = ['Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Kama wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Rosa wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat', 'Canadian wheat']\n",
    "print(len(mergings), len(varieties))\n",
    "\n",
    "# Perform a flat hierarchical clustering using fcluster to extract labels: labels\n",
    "labels = fcluster(mergings, 6, criterion='distance')\n",
    "\n",
    "# Create a DataFrame with labels and varieties as columns: df\n",
    "df = pd.DataFrame({'labels': labels, 'varieties': varieties})\n",
    "\n",
    "# Create crosstab: ct\n",
    "ct = pd.crosstab(df['labels'], df['varieties'])\n",
    "\n",
    "# Display ct\n",
    "print(ct)\n",
    "\n",
    "\"\"\"\n",
    "you've now mastered the fundamentals of k-Means and agglomerative hierarchical clustering. \n",
    "Next, you'll learn about t-SNE, which is a powerful tool for visualizing high dimensional data.\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "2-Dimensional mapping with t-SNE.\n",
    "t-SNE is great for understanding, inspecting and visualizing a dataset.\n",
    "Can help visually compete, validate or invalidate K-Means clustering inertia plot (and elbow plot).\n",
    "\n",
    "t-SNE is okay for moderate amount of data, problem being its axis have no real interpretable meaning, they change every time\n",
    "+ t-SNE has only a fit_transform() method (cannot add new data samples in the middle of the process, have to start over).\n",
    "i\"\"\"\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "t-SNE visualization of grain dataset\n",
    "In the video, you saw t-SNE applied to the iris dataset. In this exercise, you'll apply t-SNE to the grain\n",
    "samples data and inspect the resulting t-SNE features using a scatter plot. You are given an array samples\n",
    "of grain samples and a list variety_numbers giving the variety number of each grain sample.\n",
    "\"\"\"\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains'\n",
    "os.chdir(path)\n",
    "\n",
    "seeds = pd.read_csv('seeds.csv')\n",
    "n_rows_rand = np.random.choice(seeds.index, size = 210)\n",
    "samples = seeds.iloc[n_rows_rand, :]\n",
    "\n",
    "\n",
    "# Import TSNE\n",
    "from sklearn.manifold import TSNE\n",
    "\n",
    "# Create a TSNE instance: model\n",
    "model = TSNE(learning_rate=200)\n",
    "\n",
    "# Apply fit_transform to samples: tsne_features\n",
    "tsne_features = model.fit_transform(samples)\n",
    "\n",
    "# Select the 0th feature: xs\n",
    "xs = tsne_features[:,0]\n",
    "\n",
    "# Select the 1st feature: ys\n",
    "ys = tsne_features[:,1]\n",
    "\n",
    "variety_numbers = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3]\n",
    "# Scatter plot, coloring by variety_numbers\n",
    "plt.scatter(xs, ys, c=variety_numbers)\n",
    "plt.show()\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "/!\\ Remember data here is sampled, so the result here has no meaning.\n",
    "As you can see, the t-SNE visualization manages to separate the 2 varieties of grain samples. \n",
    "But how will it perform on the stock data? You'll find out in the next exercise!\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "A t-SNE map of the stock market\n",
    "t-SNE provides great visualizations when the individual samples can be labeled. In this exercise, \n",
    "you'll apply t-SNE to the company stock price data. A scatter plot of the resulting t-SNE features, \n",
    "labeled by the company names, gives you a map of the stock market! The stock price movements for each\n",
    "company are available as the array normalized_movements (these have already been normalized for you).\n",
    "The list companies gives the name of each company. PyPlot (plt) has been imported for you.\n",
    "\"\"\"\n",
    "\n",
    "# Import TSNE\n",
    "from sklearn.manifold import TSNE\n",
    "\n",
    "# Create a TSNE instance: model\n",
    "model = TSNE(learning_rate = 50)\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/'\n",
    "os.chdir(path)\n",
    "mov = pd.read_csv('company-stock-movements-2010-2015-incl.csv')\n",
    "exclude = ['Unnamed: 0']\n",
    "normalized_movements = mov.drop(exclude, axis=1)\n",
    "\n",
    "companies = mov.iloc[:, 0]\n",
    "\n",
    "# Apply fit_transform to normalized_movements: tsne_features\n",
    "tsne_features = model.fit_transform(normalized_movements)\n",
    "\n",
    "# Select the 0th feature: xs\n",
    "xs = tsne_features[:,0]\n",
    "\n",
    "# Select the 1th feature: ys\n",
    "ys = tsne_features[:,1]\n",
    "\n",
    "# Scatter plot\n",
    "plt.scatter(xs, ys, alpha=.5)\n",
    "\n",
    "# Annotate the points\n",
    "for x, y, company in zip(xs, ys, companies):\n",
    "    plt.annotate(company, (x, y), fontsize=5, alpha=0.75)\n",
    "plt.show()\n",
    "\n",
    "\"\"\"\n",
    "It's visualizations such as this that make t-SNE such a powerful tool for extracting quick insights from high\n",
    "dimensional data.\n",
    "\"\"\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   3.312  5.763\n",
      "0  3.333  5.554\n",
      "1  3.337  5.291\n",
      "2  3.379  5.324\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8604405325922295\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-1.2462176168916384e-16\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "PCA for dimensionality reduction, prior to using an unsupervised learning algorithm (to eliminate noisy features).\n",
    "PCA rotates the data samples so they are aligned with the axes\n",
    "PCA shifts data samples so that their mean is = 0\n",
    "No information loss\n",
    "And data is normalized and stored in more efficient performance-wise data types\n",
    "\n",
    "fit() is a method that \"learns\" from data based on a method, but it does not transform the data / process the data.\n",
    "transform() is a method that \"processes\" the data into its new form, after fitting. It can be applied to new data !\n",
    "(based on what we learned in the fit() part).\n",
    "\n",
    "fit_transform() does both in the same pass.\n",
    ";\n",
    "Remember : PCA features are not correlated.\n",
    "PCA produces principal components that resume data the best (components_)\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Correlated data in nature\n",
    "You are given an array grains giving the width and length of samples of grain. You suspect that width \n",
    "and length will be correlated. To confirm this, make a scatter plot of width vs length and measure their\n",
    "Pearson correlation.\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "# Perform the necessary imports\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import pearsonr\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os, sys\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains/'\n",
    "os.chdir(path)\n",
    "grains = pd.read_csv('seeds-width-vs-length.csv')\n",
    "print(grains.iloc[0:3, :])\n",
    "\n",
    "# Assign the 0th column of grains: width\n",
    "width = grains.iloc[:, 0]\n",
    "\n",
    "# Assign the 1st column of grains: length\n",
    "length = grains.iloc[:, 1]\n",
    "\n",
    "\n",
    "# Scatter plot width vs length\n",
    "plt.scatter(width, length)\n",
    "plt.axis('equal')\n",
    "plt.show()\n",
    "\n",
    "# Calculate the Pearson correlation\n",
    "correlation, pvalue = pearsonr(width, length)\n",
    "\n",
    "# Display the correlation\n",
    "print(correlation)\n",
    "\n",
    "\"\"\"\n",
    "As you would expect, the width and length of the grain samples are highly correlated.\n",
    "\n",
    "Now, decorrelating the grain measurements with PCA\n",
    "You observed in the previous exercise that the width and length measurements of the grain are correlated. \n",
    "Now, you'll use PCA to decorrelate these measurements, then plot the decorrelated points and measure their \n",
    "Pearson correlation.\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "# Import PCA\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# Create PCA instance: model\n",
    "model = PCA()\n",
    "\n",
    "# Apply the fit_transform method of model to grains: pca_features\n",
    "pca_features = PCA().fit_transform(grains)\n",
    "\n",
    "# Assign 0th column of pca_features: xs\n",
    "xs = pca_features[:,0]\n",
    "\n",
    "# Assign 1st column of pca_features: ys\n",
    "ys = pca_features[:,1]\n",
    "\n",
    "# Scatter plot xs vs ys\n",
    "plt.scatter(xs, ys)\n",
    "plt.axis('equal')\n",
    "plt.show()\n",
    "\n",
    "# Calculate the Pearson correlation of xs and ys\n",
    "correlation, pvalue = pearsonr(xs, ys)\n",
    "\n",
    "# Display the correlation\n",
    "print(correlation)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   3.312  5.763\n",
      "0  3.333  5.554\n",
      "1  3.337  5.291\n",
      "2  3.379  5.324\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Program Files\\Anaconda3\\lib\\site-packages\\sklearn\\preprocessing\\data.py:645: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "C:\\Program Files\\Anaconda3\\lib\\site-packages\\sklearn\\base.py:464: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.fit(X, **fit_params).transform(X)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Notion of intrinsic dimension : the number of features (and which) required to approximate the entire dataset.\n",
    "This is the essential idea behind dimensionality reduction : how much a dataset can be compressed.\n",
    "\n",
    "In PCA, we identify and count the number of features that have high Variance.\n",
    "\"\"\"\n",
    "\n",
    "\"\"\"\n",
    "The first principal component\n",
    "The first principal component of the data is the direction in which the data varies the most. \n",
    "In this exercise, your job is to use PCA to find the first principal component of the length and \n",
    "width measurements of the grain samples, and represent it as an arrow on the scatter plot.\n",
    "\n",
    "The array grains gives the length and width of the grain samples. PyPlot (plt) and PCA have already\n",
    "been imported for you.\n",
    "\"\"\"\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import pearsonr\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os, sys\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Grains/'\n",
    "os.chdir(path)\n",
    "grains = pd.read_csv('seeds-width-vs-length.csv')\n",
    "print(grains.iloc[0:3, :])\n",
    "\n",
    "# Assign the 0th column of grains: width\n",
    "width = grains.iloc[:, 0]\n",
    "\n",
    "# Assign the 1st column of grains: length\n",
    "length = grains.iloc[:, 1]\n",
    "\n",
    "\n",
    "# Make a scatter plot of the untransformed points\n",
    "plt.scatter(width, length)\n",
    "\n",
    "# Create a PCA instance: model\n",
    "model = PCA()\n",
    "\n",
    "# Fit model to points\n",
    "model.fit(grains)\n",
    "\n",
    "# Get the mean of the grain samples: mean\n",
    "mean = model.mean_\n",
    "\n",
    "# Get the first principal component: first_pc\n",
    "first_pc = model.components_[0, :]\n",
    "\n",
    "# Plot first_pc as an arrow, starting at mean\n",
    "plt.arrow(mean[0], mean[1], first_pc[0], first_pc[1], color='black', width=0.04)\n",
    "\n",
    "# Keep axes on same scale\n",
    "plt.axis('equal')\n",
    "plt.show()\n",
    "\n",
    "\"\"\"\n",
    "This is the direction in which the grain data varies the most.\n",
    "\"\"\"\n",
    "\n",
    "\"\"\"\n",
    "Variance of the PCA features\n",
    "The fish dataset is 6-dimensional. But what is its intrinsic dimension? Make a plot of the variances of the\n",
    "PCA features to find out. As before, samples is a 2D array, where each row represents a fish. You'll need\n",
    "to standardize the features first.\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "# Perform the necessary imports\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.pipeline import make_pipeline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Create scaler: scaler\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Create a PCA instance: pca\n",
    "pca = PCA()\n",
    "\n",
    "# Create pipeline: pipeline\n",
    "pipeline = make_pipeline(scaler, pca)\n",
    "\n",
    "# Fit the pipeline to 'samples'\n",
    "pipeline.fit(samples)\n",
    "\n",
    "# Plot the explained variances\n",
    "features = range(pca.n_components_)\n",
    "plt.bar(features, pca.explained_variance_)\n",
    "plt.xlabel('PCA feature')\n",
    "plt.ylabel('variance')\n",
    "plt.xticks(features)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(84, 2)\n",
      "[[0.51785612 0.         0.         0.68091856 0.51785612 0.        ]\n",
      " [0.         0.         0.51785612 0.         0.51785612 0.68091856]\n",
      " [0.51785612 0.68091856 0.51785612 0.         0.         0.        ]]\n",
      "['cats', 'chase', 'dogs', 'meow', 'say', 'woof']\n",
      "    label                                        article\n",
      "49      0                                       Lymphoma\n",
      "48      0                                     Gabapentin\n",
      "47      0                                          Fever\n",
      "46      0                                     Prednisone\n",
      "45      0                                    Hepatitis C\n",
      "44      0                                           Gout\n",
      "43      0                                       Leukemia\n",
      "42      0                                    Doxycycline\n",
      "41      0                                    Hepatitis B\n",
      "40      0                                    Tonsillitis\n",
      "19      1  2007 United Nations Climate Change Conference\n",
      "18      1  2010 United Nations Climate Change Conference\n",
      "17      1  Greenhouse gas emissions by the United States\n",
      "16      1                                        350.org\n",
      "15      1                                 Kyoto Protocol\n",
      "11      1       Nationally Appropriate Mitigation Action\n",
      "13      1                               Connie Hedegaard\n",
      "12      1                                   Nigel Lawson\n",
      "10      1                                 Global warming\n",
      "14      1                                 Climate change\n",
      "26      2                                     Mila Kunis\n",
      "28      2                                  Anne Hathaway\n",
      "27      2                                 Dakota Fanning\n",
      "25      2                                  Russell Crowe\n",
      "29      2                               Jennifer Aniston\n",
      "23      2                           Catherine Zeta-Jones\n",
      "22      2                              Denzel Washington\n",
      "21      2                             Michael Fassbender\n",
      "20      2                                 Angelina Jolie\n",
      "24      2                                   Jessica Biel\n",
      "38      3                                         Neymar\n",
      "30      3                  France national football team\n",
      "31      3                              Cristiano Ronaldo\n",
      "32      3                                   Arsenal F.C.\n",
      "33      3                                 Radamel Falcao\n",
      "34      3                             Zlatan Ibrahimović\n",
      "35      3                Colombia national football team\n",
      "36      3              2014 FIFA World Cup qualification\n",
      "37      3                                       Football\n",
      "39      3                                  Franck Ribéry\n",
      "0       4                                       HTTP 404\n",
      "9       4                                       LinkedIn\n",
      "8       4                                        Firefox\n",
      "7       4                                  Social search\n",
      "6       4                    Hypertext Transfer Protocol\n",
      "5       4                                         Tumblr\n",
      "4       4                                  Google Search\n",
      "3       4                                    HTTP cookie\n",
      "2       4                              Internet Explorer\n",
      "1       4                                 Alexa Internet\n",
      "57      5                          Red Hot Chili Peppers\n",
      "56      5                                       Skrillex\n",
      "55      5                                  Black Sabbath\n",
      "54      5                                 Arctic Monkeys\n",
      "50      5                                   Chad Kroeger\n",
      "52      5                                     The Wanted\n",
      "51      5                                     Nate Ruess\n",
      "58      5                                         Sepsis\n",
      "53      5                                   Stevie Nicks\n",
      "59      5                                    Adam Levine\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Dimension reduction of the fish measurements\n",
    "In a previous exercise, you saw that 2 was a reasonable choice for the \"intrinsic dimension\" of the fish measurements.\n",
    "Now use PCA for dimensionality reduction of the fish measurements, retaining only the 2 most important components.\n",
    "The fish measurements have already been scaled for you, and are available as scaled_samples.\n",
    "\"\"\"\n",
    "\n",
    "# Perform the necessary imports\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.pipeline import make_pipeline\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import pearsonr\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os, sys\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/'\n",
    "os.chdir(path)\n",
    "\n",
    "fish = pd.read_csv('fish.csv')\n",
    "species = fish.iloc[:, 0]\n",
    "samples = fish.drop(fish.columns[0], axis = 1, inplace = False)\n",
    "\n",
    "# Create scaler: scaler\n",
    "scaler = StandardScaler()\n",
    "pca = PCA(n_components = 2)\n",
    "\n",
    "scaled_samples = scaler.fit_transform(samples)\n",
    "\n",
    "# pipeline = make_pipeline(scaler, pca)\n",
    "# pipeline.fit(samples)\n",
    "\n",
    "pca.fit(scaled_samples)\n",
    "# Transform the scaled samples: pca_features\n",
    "pca_features = pca.transform(scaled_samples)\n",
    "\n",
    "# Print the shape of pca_features\n",
    "print(pca_features.shape)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Clustering Words, texts.\n",
    "Be careful of the formatting of the input data here : sparse array CSR matrix, numpy arrays.\n",
    "tf = frequency of words in document\n",
    "idf = reduces influence of frequent words\n",
    "\n",
    "\n",
    "A tf-idf word-frequency array\n",
    "In this exercise, you'll create a tf-idf word frequency array for a toy collection of documents. \n",
    "For this, use the TfidfVectorizer from sklearn. It transforms a list of documents into a word frequency\n",
    "array, which it outputs as a csr_matrix. It has fit() and transform() methods like other sklearn objects.\n",
    "You are given a list documents of toy documents about pets. Its contents have been printed in the IPython Shell.\n",
    "\"\"\"\n",
    "documents = ['cats say meow', 'dogs say woof', 'dogs chase cats']\n",
    "\n",
    "# Import TfidfVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "# Create a TfidfVectorizer: tfidf\n",
    "tfidf = TfidfVectorizer() \n",
    "\n",
    "# Apply fit_transform to document: csr_mat\n",
    "csr_mat = tfidf.fit_transform(documents)\n",
    "\n",
    "# Print result of toarray() method\n",
    "print(csr_mat.toarray())\n",
    "\n",
    "# Get the words: words\n",
    "words = tfidf.get_feature_names()\n",
    "\n",
    "# Print words\n",
    "print(words)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "You'll now move to clustering Wikipedia articles!\n",
    "\"\"\"\n",
    "\n",
    "# Perform the necessary imports\n",
    "from sklearn.decomposition import TruncatedSVD\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from scipy.sparse import csr_matrix\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "# Create a TruncatedSVD instance: svd\n",
    "svd = TruncatedSVD(n_components=50)\n",
    "# Create a KMeans instance: kmeans\n",
    "kmeans = KMeans(n_clusters=6)\n",
    "# Create a pipeline: pipeline\n",
    "pipeline = make_pipeline(svd, kmeans)\n",
    "# Import pandas\n",
    "import pandas as pd\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Wikipedia articles/'\n",
    "os.chdir(path)\n",
    "documents = pd.read_csv('wikipedia-vectors.csv', index_col=0)\n",
    "\n",
    "# articles = articles.drop(articles.columns[0], axis=1)\n",
    "titles = documents.columns\n",
    "\n",
    "# Create a TfidfVectorizer: tfidf\n",
    "articles = csr_matrix(documents.values).T\n",
    "\n",
    "# Fit the pipeline to articles\n",
    "pipeline.fit(articles)\n",
    "\n",
    "# Calculate the cluster labels: labels\n",
    "labels = pipeline.predict(articles)\n",
    "\n",
    "# Create a DataFrame aligning labels and titles: df\n",
    "df = pd.DataFrame({'label': labels, 'article': titles})\n",
    "\n",
    "# Display df sorted by cluster label\n",
    "print(df.sort_values(by='label'))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NMF : non negative matrix factorization\n",
    "\n",
    "(because it uses the concept of product of matrices that do not contain Zeros)\n",
    "It is like PCA, but instead, it is interpretable, easy to explain; although requires that all samples & features\n",
    "are != than 0 (such as word frequencies arrays, image encoded in arrays, audio spectrograms...) = always a frequency,\n",
    "at least = 1\n",
    "\n",
    "tf = frequency of words in document\n",
    "idf = reduces influence of frequent words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.sparse import csr_matrix\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Wikipedia articles/'\n",
    "os.chdir(path)\n",
    "\n",
    "documents = pd.read_csv('wikipedia-vectors.csv', index_col=0)\n",
    "titles = documents.columns\n",
    "articles = csr_matrix(documents.values).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.   0.   0.   0.   0.   0.44]\n",
      " [0.   0.   0.   0.   0.   0.57]\n",
      " [0.   0.   0.   0.   0.   0.4 ]\n",
      " [0.   0.   0.   0.   0.   0.38]\n",
      " [0.   0.   0.   0.   0.   0.49]\n",
      " [0.01 0.01 0.01 0.03 0.   0.33]\n",
      " [0.   0.   0.02 0.   0.01 0.36]\n",
      " [0.   0.   0.   0.   0.   0.49]\n",
      " [0.02 0.01 0.   0.02 0.03 0.48]\n",
      " [0.01 0.03 0.03 0.07 0.02 0.34]\n",
      " [0.   0.   0.53 0.   0.03 0.  ]\n",
      " [0.   0.   0.36 0.   0.   0.  ]\n",
      " [0.01 0.01 0.31 0.06 0.01 0.02]\n",
      " [0.   0.01 0.34 0.01 0.   0.  ]\n",
      " [0.   0.   0.43 0.   0.04 0.  ]\n",
      " [0.   0.   0.48 0.   0.   0.  ]\n",
      " [0.01 0.02 0.38 0.03 0.   0.01]\n",
      " [0.   0.   0.48 0.   0.   0.  ]\n",
      " [0.   0.01 0.55 0.   0.   0.  ]\n",
      " [0.   0.   0.47 0.   0.   0.  ]\n",
      " [0.   0.01 0.02 0.52 0.06 0.01]\n",
      " [0.   0.   0.   0.51 0.   0.  ]\n",
      " [0.   0.01 0.   0.42 0.   0.  ]\n",
      " [0.   0.   0.   0.44 0.   0.  ]\n",
      " [0.   0.   0.   0.5  0.   0.  ]\n",
      " [0.1  0.09 0.   0.38 0.   0.01]\n",
      " [0.   0.   0.   0.57 0.   0.01]\n",
      " [0.01 0.01 0.   0.47 0.   0.01]\n",
      " [0.   0.   0.   0.58 0.   0.  ]\n",
      " [0.   0.   0.   0.53 0.01 0.01]\n",
      " [0.   0.41 0.   0.   0.   0.  ]\n",
      " [0.   0.61 0.   0.01 0.   0.  ]\n",
      " [0.01 0.27 0.   0.02 0.01 0.  ]\n",
      " [0.   0.64 0.   0.   0.   0.  ]\n",
      " [0.   0.61 0.   0.   0.   0.  ]\n",
      " [0.   0.34 0.   0.   0.   0.  ]\n",
      " [0.01 0.32 0.02 0.   0.01 0.  ]\n",
      " [0.01 0.21 0.01 0.05 0.02 0.01]\n",
      " [0.01 0.47 0.   0.02 0.   0.  ]\n",
      " [0.   0.64 0.   0.   0.   0.  ]\n",
      " [0.   0.   0.   0.   0.48 0.  ]\n",
      " [0.   0.   0.   0.   0.49 0.  ]\n",
      " [0.   0.   0.   0.   0.38 0.01]\n",
      " [0.   0.   0.   0.01 0.54 0.  ]\n",
      " [0.   0.   0.01 0.   0.42 0.  ]\n",
      " [0.   0.   0.   0.   0.51 0.  ]\n",
      " [0.   0.   0.   0.   0.37 0.  ]\n",
      " [0.   0.   0.04 0.   0.23 0.  ]\n",
      " [0.01 0.   0.02 0.01 0.33 0.04]\n",
      " [0.   0.   0.   0.   0.42 0.  ]\n",
      " [0.31 0.   0.   0.   0.   0.  ]\n",
      " [0.37 0.   0.   0.   0.   0.  ]\n",
      " [0.4  0.03 0.   0.02 0.   0.02]\n",
      " [0.38 0.   0.   0.04 0.   0.01]\n",
      " [0.44 0.   0.   0.   0.   0.  ]\n",
      " [0.46 0.   0.   0.   0.   0.  ]\n",
      " [0.28 0.   0.   0.05 0.   0.02]\n",
      " [0.45 0.   0.   0.   0.01 0.  ]\n",
      " [0.29 0.01 0.01 0.01 0.19 0.01]\n",
      " [0.38 0.01 0.   0.1  0.01 0.  ]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.decomposition import NMF\n",
    "\n",
    "# Create an NMF instance: model\n",
    "model = NMF(n_components=6)\n",
    "\n",
    "# Fit the model to articles\n",
    "model.fit(articles)\n",
    "\n",
    "# Transform the articles: nmf_features\n",
    "nmf_features = model.transform(articles)\n",
    "\n",
    "# Print the NMF features\n",
    "print(nmf_features.round(2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NMF features of the Wikipedia articles\n",
    "Now you will explore the NMF features you created in the previous exercise. A solution to the previous exercise has been pre-loaded, so the array ```nmf_features``` is available. Also available is a list titles giving the ```title``` of each Wikipedia article.\n",
    "\n",
    "When investigating the features, notice that for both actors, the NMF feature 3 has by far the highest value. This means that both articles are reconstructed using mainly the 3rd NMF component. In the next video, you'll see why: NMF components represent topics (for instance, acting!)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    0.003845\n",
      "1    0.000000\n",
      "2    0.000000\n",
      "3    0.575643\n",
      "4    0.000000\n",
      "5    0.000000\n",
      "Name: Anne Hathaway, dtype: float64\n",
      "0    0.000000\n",
      "1    0.005601\n",
      "2    0.000000\n",
      "3    0.422330\n",
      "4    0.000000\n",
      "5    0.000000\n",
      "Name: Denzel Washington, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Create a pandas DataFrame: df\n",
    "df = pd.DataFrame(nmf_features, index=titles)\n",
    "\n",
    "# Print the row for 'Anne Hathaway'\n",
    "print(df.loc['Anne Hathaway'])\n",
    "\n",
    "# Print the row for 'Denzel Washington'\n",
    "print(df.loc['Denzel Washington'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notice that for both actors, the NMF feature 3 has by far the highest value. This means that both articles are reconstructed using mainly the 3rd NMF component. In the next video, you'll see why: NMF components represent topics (for instance, acting!)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NMF reconstructs samples\n",
    "In this exercise, you'll check your understanding of how NMF reconstructs samples from its components using the NMF feature values. On the right are the components of an NMF model. If the NMF feature values of a sample are ```[2, 1]```, then which of the following is most likely to represent the original sample? A pen and paper will help here! You have to apply the same technique Ben used in the video to reconstruct the sample ```[0.1203 0.1764 0.3195 0.141]```."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2.2, 1.1, 2.1])"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_feature = np.array([2, 1])\n",
    "components = np.array([[1. , 0.5, 0. ],\n",
    "                       [0.2, 0.1, 2.1]])\n",
    "\n",
    "np.matmul(sample_feature.T, components)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## NMF learns interpretable parts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NMF learns topics of documents\n",
    "In the video, you learned when NMF is applied to documents, the components correspond to topics of documents, and the NMF features reconstruct the documents from the topics. Verify this for yourself for the NMF model that you built earlier using the Wikipedia articles. Previously, you saw that the 3rd NMF feature value was high for the articles about actors Anne Hathaway and Denzel Washington. In this exercise, identify the topic of the corresponding NMF component.\n",
    "\n",
    "The NMF model you built earlier is available as ```model```, while ```words``` is a list of the words that label the columns of the word-frequency array.\n",
    "\n",
    "After you are done, take a moment to recognise the topic that the articles about Anne Hathaway and Denzel Washington have in common!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Preprocess"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [],
   "source": [
    "words = []\n",
    "\n",
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/Wikipedia articles/'\n",
    "os.chdir(path)\n",
    "with open('wikipedia-vocabulary-utf8.txt') as f:\n",
    "    words = f.read().splitlines()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6, 13125)\n",
      "film       0.627950\n",
      "award      0.253161\n",
      "starred    0.245313\n",
      "role       0.211476\n",
      "actress    0.186420\n",
      "Name: 3, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Import pandas\n",
    "import pandas as pd\n",
    "\n",
    "# Create a DataFrame: components_df\n",
    "components_df = pd.DataFrame(model.components_, columns=words)\n",
    "\n",
    "# Print the shape of the DataFrame\n",
    "print(components_df.shape)\n",
    "\n",
    "# Select row 3: component\n",
    "component = components_df.iloc[3]\n",
    "\n",
    "# Print result of nlargest\n",
    "print(component.nlargest())\n",
    "\n",
    "# Take a moment to recognise the topics that the articles about Anne Hathaway and Denzel Washington have in common!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Explore the LED digits dataset\n",
    "In the following exercises, you'll use NMF to decompose grayscale images into their commonly occurring patterns. Firstly, explore the image dataset and see how it is encoded as an array. You are given 100 images as a 2D array ```samples```, where each row represents a single 13x8 image. The images in your dataset are pictures of a LED digital display."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Preprocess"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>94</th>\n",
       "      <th>95</th>\n",
       "      <th>96</th>\n",
       "      <th>97</th>\n",
       "      <th>98</th>\n",
       "      <th>99</th>\n",
       "      <th>100</th>\n",
       "      <th>101</th>\n",
       "      <th>102</th>\n",
       "      <th>103</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 104 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   0    1    2    3    4    5    6    7    8    9    ...  94   95   96   97   \\\n",
       "0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0  0.0   \n",
       "1  0.0  0.0  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",
       "2  0.0  0.0  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",
       "3  0.0  0.0  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",
       "4  0.0  0.0  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",
       "\n",
       "   98   99   100  101  102  103  \n",
       "0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "1  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "2  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "3  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "4  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "\n",
       "[5 rows x 104 columns]"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = 'C:/Users/T0230575/source/repos/DataCamp__/Python/Unsupervised Learning with Python/data/'\n",
    "os.chdir(path)\n",
    "\n",
    "df = pd.read_csv('lcd-digits.csv', header=None)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [],
   "source": [
    "samples = df.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.\n",
      " 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0.\n",
      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0.\n",
      " 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
      " 0. 0. 0. 0. 0. 0. 0. 0.]\n",
      "[[0. 0. 0. 0. 0. 0. 0. 0.]\n",
      " [0. 0. 1. 1. 1. 1. 0. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 0. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 0. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 0. 0.]]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x1926d2beeb8>"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Select the 0th row: digit\n",
    "digit = samples[0]\n",
    "\n",
    "# Print digit\n",
    "print(digit)\n",
    "\n",
    "# Reshape digit to a 13x8 array: bitmap\n",
    "bitmap = digit.reshape(13, 8)\n",
    "\n",
    "# Print bitmap\n",
    "print(bitmap)\n",
    "\n",
    "# Use plt.imshow to display bitmap\n",
    "plt.imshow(bitmap, cmap='gray', interpolation='nearest')\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NMF learns the parts of images\n",
    "Now use what you've learned about NMF to decompose the digits dataset. You are again given the digit images as a 2D array ```samples```. This time, you are also provided with a function ```show_as_image()``` that displays the image encoded by any 1D array:\n",
    "```python\n",
    "def show_as_image(sample):\n",
    "    bitmap = sample.reshape((13, 8))\n",
    "    plt.figure()\n",
    "    plt.imshow(bitmap, cmap='gray', interpolation='nearest')\n",
    "    plt.colorbar()\n",
    "    plt.show()\n",
    "```\n",
    "\n",
    "You'll explore this dataset further in the next exercise and see for yourself how NMF can learn the parts of images."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_as_image(sample):\n",
    "    bitmap = sample.reshape((13, 8))\n",
    "    plt.figure()\n",
    "    plt.imshow(bitmap, cmap='gray', interpolation='nearest')\n",
    "    plt.colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[4.76823559e-01 0.00000000e+00 0.00000000e+00 5.90605054e-01\n",
      " 4.81559442e-01 0.00000000e+00 7.37557191e-16]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOAAAAD8CAYAAABuOagBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAE4tJREFUeJzt3X2wXVV5x/Hvr4lAeZNgfEkJr9OMI74RyoA2HUGREJQSO7XTpKLR4mTqiK+dWtAZ7eA/tnbUOiqaYkQrBmmEmjpRuCNQbDU0CaAYAxjjC9dEQwwiFUbmwtM/9rq6c7jnnH3O2fesc/b9fWb23HP221pJ7pO199prP0sRgZnl8Xu5K2A2lzkAzTJyAJpl5AA0y8gBaJaRA9AsIwegWUYOQLOMHIBmGc0fZmGSPOzGZrI/Ip7e78ErVqyI/fv3V9p3+/btN0bEin7LqttQA9CsjR8PcvD+/fvZtm1bpX0lLRykrLo5AK0RxnVMswPQGuGJJ57IXYW+OABt7EXE2LaAA/WCSloh6V5JuyRdVlelzHo1HYTdlm4kHS/pFkk7Je2Q9LYZ9pGkj6bf++9IOr20bY2k76dlTbfy+m4BJc0DPg6cB0wCWyVtiojv9XtOs37V2AJOAX8bEXdIOgrYLmmi5ff6AmBJWs4CrgTOknQs8D7gDCDSsZsi4sF2hQ3SAp4J7IqI3RHxGHAtsHKA85n1ra4WMCL2RsQd6fPDwE7guJbdVgKfi8IW4BhJi4DzgYmIOJCCbgLo+MhjkHvA44D7S98nKf43MBu6HlrAhZLKzyzWRcS6mXaUdBKwFLi9ZdNMv/vHdVjf1iABqBnWPelvQdJaYO0A5Zh1FBG99ILuj4gzuu0k6UjgS8DbI+JXrZtnqkaH9W0Ncgk6CRxf+r4Y2POk0iPWRcQZVf7QZv2q6xIUQNJTKILvmoi4foZd2v3uV4qJskECcCuwRNLJkg4BVgGbBjifWd9q7AUV8GlgZ0R8qM1um4DXpd7QFwEPRcRe4EZguaQFkhYAy9O6tvq+BI2IKUmXpgLmAesjYke/5zMbRI29oMuA1wJ3S7orrXs3cEIq55PAZuAVwC7gEeANadsBSe+naJwAroiIA50KG+hBfERsTpUxy6bOB/ER8d/MfC9X3ieAN7fZth5YX7U8j4SxRvBQNLOMxnUomgPQxt44jwWdcwE4rv9Qs6Ho8GuGcf13nXMBaM3kADTLyAFolkmPQ9FGigPQGsEtoFlGDkCzjByAZhk5AM0ycSeMWWZuAc0ycgCaZeQANMvEg7HNMnMAmmVUVy+opPXAhcC+iHjeDNv/DnhN+jofeA7w9JSO4kfAw8DjwFSVRGSeoNMaocasaFfTIZluRHwwIk6LiNOAy4H/asn78tK0vVIWQLeANvZqzglzW0rIW8VqYMMg5c25AGzSS6j2O8O+B5R0OEVLeWm5GsBNaSboT7XLuF025wLQmmk2UtN38afA/7Rcfi6LiD2SngFMSLonIm7rdBIHoDVCDwFYKTV9BatoufyMiD3p5z5JN1BMYNQxAN0JY2NveixolaUOkp4KnA18ubTuiDSdGZKOoMiK/d1u53ILaI1Q1z2gpA3AORSXqpMU8/09JZXxybTbnwE3RcSvS4c+E7gh9THMB74QEV/rVt4gE3QeD3wOeBbwBMW19L/0ez6zQdTYC7q6wj5XUzyuKK/bDbyw1/IGaQGrzCRqNhRzbiRMmg1mb/r8sKTpmUQdgDZ0cy4AyzrMJOoJOm3WzekXcrvMJEp6xrIu7Tue/03ZyJuTLWCFmUTNhmLOBWDFmUTNhmJcA3CQB/HTM4m+TNJdaXlFTfUyq6zqmxCjGKSD9IJ2nUnUbFhGMbiq8EgYa4Q52wtqNgrcApplMqr3d1U4AK0RHIBmGTkAzTJyAJplMqfHgpqNAreAZhmNawA6J4w1Ql1D0SStl7RP0oz5XCSdI+mh0vDL95a2rZB0r6Rdki6rUm+3gNYINbaAVwMfo0i30s43IuLC8gpJ84CPA+cBk8BWSZu6ZYhwANrYq7MTpsfM2GVnArtSbhgkXQuspEuGCF+CWiMM+W2IF0v6tqSvSnpuWncccH9pn8m0riO3gNYIQ8yMfQdwYkT8X3r97j+AJcz8ZlDXSjkArRGGlRm7nHYlIjZL+oSkhRQt3vGlXRcDe7qdzwFoY2+Yg7ElPQv4eUSEpDMpbuN+AfwSWCLpZOCnFKnr/6rb+RyA1ghDzIz9auBNkqaAR4FVURQ+JelS4EZgHrA+InZ0K88BaI1QYy9ox8zYEfExiscUM23bDGzupTwHoDXCuI6EcQDa2PMLuWaZOQDNMnIAmmXkADTLZJxfyB14LKikeZLulPSVOipk1o85lxm75G3ATuDoGs5l1pdRDK4qBmoBJS0GXglcVU91zPozV1vAjwDvAo6qoS5mfRvF4Kqi7xZQ0oXAvojY3mW/tZK2tbwCYlab6U6YKsuoGaQFXAZclN6JOgw4WtLnI+Li8k6eIdeGYc61gBFxeUQsjoiTKF69uLk1+MyGZa7eA5qNhFEMripqCcCIuBW4tY5zmfVqVFu3KtwCWiM4AM0yGsUeziqcltDGXtUOmJoyY79G0nfS8k1JLyxt+5Gku1PG7EqP3dwCWiMMMTP2D4GzI+JBSRdQPGI7q7T9pRGxv2phDkBrhLoCsFtm7Ij4ZunrFor0g33zJag1QqbngJcAXy1XA7hJ0nZJa6ucwC2gjb0e3wccNDM2AJJeShGAf1JavSwi9kh6BjAh6Z6IuK3TeRyA1gjDyowNIOkFFG8AXRARvyjVYU/6uU/SDRQTtnQMQF+CWiMM6xJU0gnA9cBrI+K+0vojJB01/RlYDszYk1rmFtAaYYiZsd8LPA34hCSAqdSiPhO4Ia2bD3whIr7WrTwHoDVCjb2g3TJjvxF44wzrdwMvfPIRnTkAbex5LKhZZuM6FM0BaI3gFtAsIwegWSa+BzTLzAFolpED0Cwj94KaZeJ7QLPMHIBmGTkAzTJyAJplMmcn6JR0jKSNku6RtFPSi+uqmFkv5mpq+n8BvhYRr5Z0CHB4DXUy69koBlcVfQegpKOBlwCvB4iIx4DH6qmWWW/GNQAHuQQ9BXgA+EyaI/6q9Cq+2dCN6yXoIAE4HzgduDIilgK/Bi5r3ckTdNpsG3JmbEn6qKRdKTv26aVtayR9Py1rqtR9kACcBCYj4vb0fSNFQB4kItZFxBmDZqIy66TGGXKvBlZ02H4BsCQta4ErASQdS5E/5iyKbGjvk7SgW2GDTND5M+B+Sc9Oq84Fvtfv+cwGUVcLmPJ4Huiwy0rgc1HYAhwjaRFwPjAREQci4kFggs6BDAzeC/oW4JrUA7obeMOA5zPryxDv744D7i99n0zr2q3vaKAAjIi7AF9aWlY9drAMmhlbM1Whw/qOPBLGGmGImbEngeNL3xcDe9L6c1rW39rtZM6MbY0wxMcQm4DXpd7QFwEPRcRe4EZguaQFqfNleVrXkVtAa4S6xoJWyIy9GXgFsAt4hNTvEREHJL0f2JpOdUVEdOrMARyA1gB1PmSvkBk7gDe32bYeWN9LeQ5Aa4RRHOVShQPQGsEBaJaRA9Ask3F+IdcBaI3gFtAsIwegWUYOQLOMHIBmmYzq2+5VOACtEdwLapaRW0CzjByAZpn4HtAsMwegWUYOQLOM3AtqlonvAc0yG9cAdFIma4Q6kzJJWiHp3pR+fqbpFj4s6a603Cfpl6Vtj5e2bepWlltAa4S6WkBJ84CPA+dRpBrcKmlTRPw263tEvKO0/1uApaVTPBoRp1Utzy2gjb3pF3JrmhviTGBXROxOU+5dS5GOvp3VwIZ+6z7oDLnvkLRD0nclbZB02CDnM+tXD5egC6dn60rL2pZTVU4xL+lE4GTg5tLqw9J5t0h6Vbd6DzJB53HAW4FTI+JRSdcBqyhmlzEbqhozY/eSYn4VsDEiHi+tOyEi9kg6BbhZ0t0R8YN2hQ16CTof+H1J8ymmp94z4PnM+lJjJ0y71PMzWUXL5WdE7Ek/d1Okpl/65MN+Z5DpyX4K/DPwE2AvRYrum/o9n9kgagzArcASSSenWb9WUaSjP0ialm8B8K3SugWSDk2fFwLL6DJlX98BmPLfr6S4Bv4D4AhJF8+wn2fItVlVNfgqzg84BVxKMa/DTuC6iNgh6QpJF5V2XQ1cGwef9DnANknfBm4BPlDuPZ3JII8hXg78MCIeAJB0PfDHwOdb/kDrgHVpn/F8Wmojr86haBGxmWIOiPK697Z8/4cZjvsm8PxeyhokAH8CvEjS4cCjFDPkupWzLMZ1JEzfARgRt0vaCNwBTAF3klo6s2GbcwEIEBHvo5i+ySwbD8Y2y8wBaJaR3wc0y8SXoGaZOQDNMnIAmmXkADTLyAFololnyDXLzC2gWUYOQLOMHIBmmfhBvFlmDkCzjMa1F9R5Qa0RhpwZ+/WSHihlwH5jadsaSd9Py5puZbkFtLFX5z1glczYyRcj4tKWY4+leD/2DIpUhtvTsQ+2K88toDVCjS1gr5mxy84HJiLiQAq6CWBFpwMcgNYIGTJj/7mk70jaKGk6j2jlrNrTfAlqjdBDJ0wdmbH/E9gQEb+R9DfAZ4GXVTz2IG4BbezVmReUCpmxI+IXEfGb9PVfgT+qemwrB6A1wjAzY0taVPp6EUUCXyiS+S5PGbIXAMvTurZ8CWqNUFcvaERMSZrOjD0PWD+dGRvYFhGbgLemLNlTwAHg9enYA5LeTxHEAFdExIFO5WmYIwicGdva2N7lvqyjI488Mp7//GoJqbds2TJQWXVzC2iNMK5D0breA0paL2mfpO+W1h0raSI97Z9I17tmWUy/kFvTDLlDVaUT5mqe/DDxMuDrEbEE+Hr6bpZNnUPRhqlrAEbEbRQ3mmUrKZ59kH52nYrXbDaNawD2ew/4zIjYCxAReyU9o8Y6mfVsFIOrilnvhElDfVqH+5jVZlRbtyr6DcCfS1qUWr9FwL52O3qCThuGcQ3AfkfCbAKm33VaA3y5nuqY9Wdce0G7toCSNgDnUIwin6R43+kDwHWSLqGYKfcvZrOSZt2MawvYNQAjYnWbTefWXBezvszFe0CzkeIANMvIAWiW0Sh2sFThALSx53tAs8wcgGYZOQDNMhrXAHROGGuEIWfGfqek76W0hF+XdGJp2+OljNmbWo9t5RbQxt70C7l1qJgZ+07gjIh4RNKbgH8C/jJtezQiTqtanltAa4RhZsaOiFsi4pH0dQtF+sG+OACtEWoMwF6zW18CfLX0/bCUcXuLpK4vqvsS1Bqhh06YhZK2lb6vS6/MTauc3VrSxRQTsZxdWn1CROyRdApws6S7I+IH7SrjALSx1+OD+G6p6Stlt5b0cuA9wNmlLNlExJ70c7ekW4GlQNsA9CWoNcKQM2MvBT4FXBQR+0rrF0g6NH1eCCwDWqc1O4hbQGuEunpBK2bG/iBwJPDvkgB+EhEXAc8BPiXpCYrG7QPx5HkFD+LM2DYKBspWfcghh8TChQsr7bt3715nxjarkwdjm2XmADTLyAFolpFfyDXLxPeAZpk5AM0ycgCaZeQANMtoXAOw3xlyPyjpnvRG8A2Sjpndapq1NxdnyJ0AnhcRLwDuAy6vuV5mPRnXCTr7miE3Im6KiKn0daA3gs3qMK4BWMc94F8DX2y30RN02jCMYnBVMVAASnoPMAVc024fT9Bps21UW7cq+g5ASWuAC4FzY1z/9NYY4/or2FcASloB/D3F6/iPdNvfbLaNYg9nFVUeQ2wAvgU8W9JkmhX3Y8BRwERKQPrJWa6nWUfj2glTueJ1LBTZpbx4aV22Dfp7NX/+/EpLlbIoHrvdC+wCLpth+6EUHY+7gNuBk0rbLk/r7wXO71aWkzJZI9TVApYyY18AnAqslnRqy26XAA9GxB8CHwb+MR17KkUSp+dSBPEn0vnacgBaI9R4Cdo1M3b6/tn0eSNwrorsTCuBayPiNxHxQ4qW8MxOhTkArRFqHIpWJTP2b/dJA1IeAp5W8diDDHsw9n7gxx22L0z75DYK9RiFOsBw6nHigMffSFHPKg6rITN2u30qZ9WeNtQAjIind9ouaVuMQMq4UajHKNRhlOrRSUS0jlUeRJXM2NP7TEqaDzyVYrhmpazaZb4ENTtY18zY6fua9PnVwM1pMMomYJWkQyWdDCwB/rdTYX4f0KwkqmXG/jTwb5J2UbR8q9KxOyRdR5GOfgp4c0Q83qm8oWbG7kbS2pbr8Tlbj1GowyjVo6lGKgDN5hrfA5plNDIBKGmFpHsl7ZJ0WYbyj5d0i6SdknZIetuw69BSn3mS7pT0lUzlHyNpY0o9slPSi3PUo+lG4hI0Dde5DziPoit3K7A6ukztVHMdFgGLIuIOSUcB24FXDbMOLfV5J8Xsq0dHxIUZyv8s8I2IuCr1Bh4eEb8cdj2ablRawCrDf2ZVROyNiDvS54eBnXQZxTBbJC0GXglclan8o4GXUPT2ERGPOfhmx6gEYM9DeGaTpJMopha+PVMVPgK8C8j1ktspwAPAZ9Jl8FWSjshUl0YblQDseQjPbJF0JPAl4O0R8asM5V8I7IuI7cMuu2Q+cDpwZUQsBX4NDP2+fC4YlQDseQjPbJD0FIrguyYirh92+cky4CJJP6K4FH+ZpM8PuQ6TwGRETF8BbKQISKvZqARgleE/syq9TvJpYGdEfGiYZZdFxOURsTgiTqL4e7g5Ii4ech1+Btwv6dlp1bkUozusZiMxFK3d8J8hV2MZ8Frgbkl3pXXvjojNQ67HqHgLcE36D3E38IbM9WmkkXgMYTZXjcolqNmc5AA0y8gBaJaRA9AsIwegWUYOQLOMHIBmGTkAzTL6f5IYzTO6wHGKAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Import NMF\n",
    "from sklearn.decomposition import NMF\n",
    "\n",
    "# Create an NMF model: model\n",
    "model = NMF(n_components=7)\n",
    "\n",
    "# Apply fit_transform to samples: features\n",
    "features = model.fit_transform(samples)\n",
    "\n",
    "# Call show_as_image on each component\n",
    "for component in model.components_:\n",
    "    show_as_image(component)\n",
    "    \n",
    "# Assign the 0th row of features: digit_features\n",
    "digit_features = features[0]\n",
    "\n",
    "# Print digit_features\n",
    "print(digit_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Take a moment to look through the plots and notice how NMF has expressed the digit as a sum of the components!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PCA doesn't learn parts\n",
    "Unlike NMF, PCA doesn't learn the parts of things. Its components do not correspond to topics (in the case of documents) or to parts of images, when trained on images. Verify this for yourself by inspecting the components of a PCA model fit to the dataset of LED digit images from the previous exercise. The images are available as a 2D array ```samples```. Also available is a modified version of the ```show_as_image()``` function which colors a pixel red if the value is negative.\n",
    "\n",
    "After submitting the answer, notice that the components of PCA do not represent meaningful parts of images of LED digits!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# Createa PCA instance: model\n",
    "model = PCA(n_components=7)\n",
    "\n",
    "# Apply fit_transform to samples: features\n",
    "features = model.fit_transform(samples)\n",
    "\n",
    "# Call show_as_image on each component\n",
    "for component in model.components_:\n",
    "    show_as_image(component)\n",
    "    \n",
    "# Notice that the components of PCA do not represent meaningful parts of images of LED digits!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building recommender systems using NMF\n",
    "- Finding similar articles\n",
    "    - Engineer at a large online newspaper\n",
    "    - Task: recommand articles similar to article being read by customer\n",
    "    - Similar articles should have similar topics\n",
    "- Strategy\n",
    "    - Apply NMF to the word-frequency array\n",
    "    - NMF feature values describe the topics, so similar documents have similar NMF feature values\n",
    "    - Compare NMF feature values?\n",
    "- Versions of articles\n",
    "    - Different versions of the same document have same topic proportions\n",
    "    - exact feature values may be different!\n",
    "        E.g., because one version uses many meaningless words\n",
    "    - But all versions lie on the same line through the origin\n",
    "- Cosine Similarity\n",
    "![cosine similarity](https://github.com/goodboychan/chans_jupyter/blob/master/_notebooks/image/cosine.png?raw=1)\n",
    "    - Uses the angle between the lines\n",
    "    - Higher values means more similar\n",
    "    - Maximum value is 1, when angle is 0 degrees"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Which articles are similar to 'Cristiano Ronaldo'?\n",
    "In the video, you learned how to use NMF features and the cosine similarity to find similar articles. Apply this to your NMF model for popular Wikipedia articles, by finding the articles most similar to the article about the footballer Cristiano Ronaldo. The NMF features you obtained earlier are available as ```nmf_features```, while ```titles``` is a list of the article titles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cristiano Ronaldo                1.000000\n",
      "Franck Ribéry                    0.999972\n",
      "Radamel Falcao                   0.999942\n",
      "Zlatan Ibrahimović               0.999942\n",
      "France national football team    0.999923\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import normalize\n",
    "\n",
    "# Normalize the NMF features: norm_features\n",
    "norm_features = normalize(nmf_features)\n",
    "\n",
    "# Create a DataFrame: df\n",
    "df = pd.DataFrame(norm_features, index=titles)\n",
    "\n",
    "# Select the row corresponding to 'Cristiano Ronaldo': article\n",
    "article = df.loc['Cristiano Ronaldo']\n",
    "\n",
    "# Compute the dot products: similarities\n",
    "similarities = df.dot(article)\n",
    "\n",
    "# Display thouse with the largest cosine similarity\n",
    "print(similarities.nlargest())\n",
    "\n",
    "# although you may need to know a little about football (or soccer, depending on where you're from!) \n",
    "# to be able to evaluate for yourself the quality of the computed similarities!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Recommend musical artists part I\n",
    "In this exercise and the next, you'll use what you've learned about NMF to recommend popular music artists! You are given a sparse array ```artists``` whose rows correspond to artists and whose columns correspond to users. The entries give the number of times each artist was listened to by each user.\n",
    "\n",
    "In this exercise, build a pipeline and transform the array into normalized NMF features. The first step in the pipeline, ```MaxAbsScaler```, transforms the data so that all users have the same influence on the model, regardless of how many different artists they've listened to. In the next exercise, you'll use the resulting normalized NMF features for recommendation!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Preprocess"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.sparse import coo_matrix\n",
    "\n",
    "df = pd.read_csv('./dataset/scrobbler-small-sample.csv')\n",
    "artists1 = df.sort_values(['artist_offset', 'user_offset'], ascending=[True, True])\n",
    "row_ind = np.array(artists1['artist_offset'])\n",
    "col_ind = np.array(artists1['user_offset'])\n",
    "data1 = np.array(artists1['playcount'])\n",
    "artists = coo_matrix((data1, (row_ind, col_ind)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Perform the necessary imports\n",
    "from sklearn.decomposition import NMF\n",
    "from sklearn.preprocessing import Normalizer, MaxAbsScaler\n",
    "from sklearn.pipeline import make_pipeline\n",
    "\n",
    "# Create a MaxAbsScaler: scaler\n",
    "scaler = MaxAbsScaler()\n",
    "\n",
    "# Create an NMF model: nmf\n",
    "nmf = NMF(n_components=20)\n",
    "\n",
    "# Create a Normalizer: normalizer\n",
    "normalizer = Normalizer()\n",
    "\n",
    "# Create a pipeline: pipeline\n",
    "pipeline = make_pipeline(scaler, nmf, normalizer)\n",
    "\n",
    "# Apply fit_transform to artists: norm_features\n",
    "norm_features = pipeline.fit_transform(artists)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "now that you've computed the normalized NMF features, you'll use them in the next exercise to recommend musical artists!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import pandas\n",
    "import pandas as pd\n",
    "\n",
    "# Create a DataFrame: df\n",
    "df = pd.DataFrame(norm_features, index=artist_names)\n",
    "\n",
    "# Select row of 'Bruce Springsteen': artist\n",
    "artist = df.loc['Bruce Springsteen']\n",
    "\n",
    "# Compute cosine similarities: similarities\n",
    "similarities = df.dot(artist)\n",
    "\n",
    "# Display those with highest cosine similarity\n",
    "print(similarities.nlargest())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "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
}