3_MNIST.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a2dc43b5",
   "metadata": {},
   "source": [
    "# MNIST\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25026c49",
   "metadata": {},
   "source": [
    "In this project we will be using The MNISt dataset, which is a set of 70,000 small images of digits handwritten by high  school  students  and  employees  of  the  US  Cen‐sus Bureau. Each image is labeled with the digit it represents. This set has been stud‐ied  so  much  that  it  is  often  called  the  “hello  world”  of  Machine  Learning.\n",
    "</br>\n",
    "Scikit-Learn provides many helper functions to download popular datasets. MNIST isone of them. The following code fetches the MNIST dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "04eb600e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['data', 'target', 'frame', 'categories', 'feature_names', 'target_names', 'DESCR', 'details', 'url'])"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_openml\n",
    "\n",
    "mnist = fetch_openml('mnist_784', version=1, as_frame=False)\n",
    "mnist.keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16ea9206",
   "metadata": {},
   "source": [
    "Datasets loaded by Scikit-Learn generally have a similar dictionary structure, includ‐ing the following:\n",
    "- A DESCR key describing the dataset\n",
    "- A  data  key  containing  an  array  with  one  row  per  instance  and  one  column  perfeature\n",
    "- A target key containing an array with the labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5f70e693",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(70000, 784)\n",
      "(70000,)\n"
     ]
    }
   ],
   "source": [
    "X, y = mnist[\"data\"], mnist[\"target\"]\n",
    "\n",
    "print(X.shape)\n",
    "print(y.shape)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afb78bb3",
   "metadata": {},
   "source": [
    "There are 70,000 images, and each image has 784 features. This is because each imageis  28  ×  28  pixels,  and  each  feature  simply  represents  one  pixel’s  intensity,  from  0 (white) to 255 (black). Let’s take a peek at one digit from the dataset. All we need to do  is  grab  an  instance’s  feature  vector,  reshape  it  to  a  28  ×  28  array,  and  display  it using Matplotlib’s *imshow()* function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "fd1dde45",
   "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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5\n"
     ]
    }
   ],
   "source": [
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "some_digit = X[0]\n",
    "some_digit_image = some_digit.reshape(28, 28)\n",
    "\n",
    "plt.imshow(some_digit_image, cmap=mpl.cm.binary)\n",
    "plt.axis(\"off\")\n",
    "plt.show()\n",
    "print(y[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c0dddd4",
   "metadata": {},
   "source": [
    "Lets plot some more data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4836c3f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_digit(data):\n",
    "    image = data.reshape(28, 28)\n",
    "    plt.imshow(image, cmap = mpl.cm.binary,\n",
    "               interpolation=\"nearest\")\n",
    "    plt.axis(\"off\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "809a2c0c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_digits(instances, images_per_row=10, **options):\n",
    "    size = 28\n",
    "    images_per_row = min(len(instances), images_per_row)\n",
    "    images = [instance.reshape(size,size) for instance in instances]\n",
    "    n_rows = (len(instances) - 1) // images_per_row + 1\n",
    "    row_images = []\n",
    "    n_empty = n_rows * images_per_row - len(instances)\n",
    "    images.append(np.zeros((size, size * n_empty)))\n",
    "    for row in range(n_rows):\n",
    "        rimages = images[row * images_per_row : (row + 1) * images_per_row]\n",
    "        row_images.append(np.concatenate(rimages, axis=1))\n",
    "    image = np.concatenate(row_images, axis=0)\n",
    "    plt.imshow(image, cmap = mpl.cm.binary, **options)\n",
    "    plt.axis(\"off\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6517b270",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(9,9))\n",
    "example_images = X[:100]\n",
    "plot_digits(example_images, images_per_row=10)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c4bc576",
   "metadata": {},
   "source": [
    "The y label is string, lets change it to integer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0d711044",
   "metadata": {},
   "outputs": [],
   "source": [
    "y = y.astype(np.uint8)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a10f05dc",
   "metadata": {},
   "source": [
    "The MNIST dataset is actually already split into a training set (the first 60,000images) and a test set (the last 10,000 images).\n",
    "</br>\n",
    "The training set is already shuffled for us, which is good because this guarantees that all cross-validation folds will be similar (we don’t want one fold to be missing some digits). Moreover, some learning algorithms are sensitive to the order of the training instances, and they perform poorly if they get many similar instances in a row. Shuffling the dataset ensures that this won’t happen"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4863f2e9",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = X[:60000], X[60000:], y[:60000], y[60000:]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa88c8b7",
   "metadata": {},
   "source": [
    "## Training a Binary Classifier\n",
    "\n",
    "Let’s  simplify  the  problem  for  now  and  only  try  to  identify  one  digit for  example, the  number  5.  This  “5-detector”  will  be  an  example  of  a  binary  classifier,  capable  of distinguishing between just two classes, 5 and not-5. Let’s create the target vectors for this classification task"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8af6b3c9",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_5 = (y_train == 5) # True for all 5s, False for all other digits\n",
    "y_test_5 = (y_test == 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f0a70b2",
   "metadata": {},
   "source": [
    "Now let’s pick a classifier and train it. A good place to start is with a Stochastic Gradient  Descent  (SGD)  classifier,  using  Scikit-Learn’s  SGDClassifier  class.  This  classifier has  the  advantage  of  being  capable  of  handling  very  large  datasets  efficiently.  This  is in  part  because  SGD  deals  with  training  instances  independently,  one  at  a  time (which also makes SGD well suited for online learning), as we will see later. Let’s create an SGDClassifier and train it on the whole training set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "54517246",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SGDClassifier(random_state=42)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "\n",
    "sgd_clf = SGDClassifier(max_iter=1000, tol=1e-3, random_state=42)\n",
    "sgd_clf.fit(X_train, y_train_5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b475d6e",
   "metadata": {},
   "source": [
    "Now we can use it to detect images of the number 5."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3e5ce81c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52f1029b",
   "metadata": {},
   "source": [
    "The prediction was indeed right as we have used *some_digit* which is variable with value 5."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df4e035d",
   "metadata": {},
   "source": [
    "## Performance Measures\n",
    "\n",
    "Evaluating a classifier is often significantly trickier than evaluating a regressor, so we will  spend  a  large  part  of  this  chapter  on  this  topic."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64ff9b74",
   "metadata": {},
   "source": [
    "### Measuring Accuracy Using Cross-Validation\n",
    "\n",
    "Occasionally we will need more control over the cross-validation process than what Scikit-Learn provides off the shelf. In these cases, we can implement cross-validation ourself. The   following   code   does   roughly   the   same   thing   as   Scikit-Learns *cross_val_score()* function, and it prints the same result.\n",
    "</br>\n",
    "The  *StratifiedKFold*  class  performs  stratified  sampling  (as  explained  in  Chapter  2) to produce folds that contain a representative ratio of each class. At each iteration the code creates a clone of the classifier, trains that clone on the training folds, and makes predictions  on  the  test  fold.  Then  it  counts  the  number  of  correct  predictions  and outputs the ratio of correct predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "65dff554",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9669\n",
      "0.91625\n",
      "0.96785\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.base import clone\n",
    "\n",
    "skfolds = StratifiedKFold(n_splits=3, shuffle=True, random_state=42)\n",
    "\n",
    "for train_index, test_index in skfolds.split(X_train, y_train_5):\n",
    "    clone_clf = clone(sgd_clf)\n",
    "    X_train_folds = X_train[train_index]\n",
    "    y_train_folds = y_train_5[train_index]\n",
    "    X_test_fold = X_train[test_index]\n",
    "    y_test_fold = y_train_5[test_index]\n",
    "\n",
    "    clone_clf.fit(X_train_folds, y_train_folds)\n",
    "    y_pred = clone_clf.predict(X_test_fold)\n",
    "    n_correct = sum(y_pred == y_test_fold)\n",
    "    print(n_correct / len(y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f054012",
   "metadata": {},
   "source": [
    "Let’s  use  the  *cross_val_score()* function  to  evaluate  our  *SGDClassifier*  model, using K-fold cross-validation with three folds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0dad0bdb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.95035, 0.96035, 0.9604 ])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "cross_val_score(sgd_clf, X_train, y_train_5, cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0cb0252",
   "metadata": {},
   "source": [
    "Lets looka at a classifier that just classifies every single image in the \"not-5\" class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "be98559c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator\n",
    "\n",
    "class Never5Classifier(BaseEstimator):\n",
    "    def fit(self, X, y=None):\n",
    "        return self\n",
    "    def predict(self, X):\n",
    "        return np.zeros((len(X), 1), dtype=bool)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0e3d80d",
   "metadata": {},
   "source": [
    "Let's find the model's accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "8776016e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.91125, 0.90855, 0.90915])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "never_5_clf = Never5Classifier()\n",
    "cross_val_score(never_5_clf, X_train, y_train_5, cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4242fe83",
   "metadata": {},
   "source": [
    "That’s  right,  it  has  over  90%  accuracy!  This  is  simply  because  only  about  10%  of  the images are 5s, so if you always guess that an image is not a 5, you will be right about 90% of the time. Beats Nostradamus.\n",
    "</br>\n",
    "This demonstrates why accuracy is generally not the preferred performance measure for  classifiers,  especially  when  you  are  dealing  with  skewed  datasets  (i.e.,  when  some classes are much more frequent than others)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09186732",
   "metadata": {},
   "source": [
    "## Confusion Matrix\n",
    "\n",
    "A much better way to evaluate the performance of a classifier is to look at the confusion matrix. The general idea is to count the number of times instances of class A are classified as class B. For example, to know the number of times the classifier confused images of 5s with 3s, you would look in the fifth row and third column of the confusion matrix. To  compute  the  confusion  matrix,  you  first  need  to  have  a  set  of  predictions  so  that they  can  be  compared  to  the  actual  targets.  We  could  make  predictions  on  the  testset,  but  let’s  keep  it  untouched  for  now  (remember  that  we  want  to  use  the  test  set only  at  the  very  end  of  our  project,  once  we  have  a  classifier  that  you  are  ready  to launch). Instead, we can use the *cross_val_predict()* function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "9466b1a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "y_train_pred = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef84e782",
   "metadata": {},
   "source": [
    "Just  like  the  *cross_val_score()*  function,  *cross_val_predict()*  performs  K-foldcross-validation, but instead of returning the evaluation scores, it returns the predictions  made  on  each  test  fold.  This  means  that  we  get  a  clean  prediction  for  each instance  in  the  training  set  (“clean”  meaning  that  the  prediction  is  made  by  a  model that never saw the data during training).\n",
    "</br>\n",
    "Now we are ready to get the confusion matrix using the *confusion_matrix()* function. Just pass   it   the   target   classes   (y_train_5)   and   the   predicted   classes(y_train_pred)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "0a30f54c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[53892,   687],\n",
       "       [ 1891,  3530]])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "confusion_matrix(y_train_5, y_train_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2a74cfc6",
   "metadata": {},
   "source": [
    "Each row in a confusion matrix represents an actual class, while each column represents a predicted class. The first row of this matrix considers non-5 images (the negative  class):  53,057  of  them  were  correctly  classified  as  non-5s  (they  are  called  truenegatives),  while  the  remaining  1,522  were  wrongly  classified  as  5s  (false  positives).The  second  row  considers  the  images  of  5s  (the  positive  class):  1,325  were  wronglyclassified as non-5s (false negatives), while the remaining 4,096 were correctly classi‐fied as 5s (true positives). A perfect classifier would have only true positives and truenegatives, so its confusion matrix would have nonzero values only on its main diago‐nal (top left to bottom right)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b643114d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[54579,     0],\n",
       "       [    0,  5421]])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_perfect_predictions = y_train_5 # pretend we reached perfection\n",
    "confusion_matrix(y_train_5, y_train_perfect_predictions)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db816c91",
   "metadata": {},
   "source": [
    "### Precision and Recall\n",
    "\n",
    "Lets look at the precison and recall of the result.\n",
    "</br>\n",
    "precision =TP/(TP+FP) where TP is the number of true positives, and FP is the number of false positives.\n",
    "</br>\n",
    "A  trivial  way  to  have  perfect  precision  is  to  make  one  single  positive  prediction  and ensure it is correct (precision = 1/1 = 100%). But this would not be very useful, since the classifier would ignore all but one positive instance. So precision is typically used along with another metric named recall, also called sensitivity or the true positive rate(TPR): this is the ratio of positive instances that are correctly detected by the classifier.\n",
    "</br>\n",
    "recall = TP/(TP+FN)\n",
    "\n",
    "FN is, ofcourse, the number of false negatives."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6d1243b6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision:  0.8370879772350012 Recall:  0.6511713705958311\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score, recall_score\n",
    "precision = precision_score(y_train_5, y_train_pred)\n",
    "recall = recall_score(y_train_5, y_train_pred)\n",
    "\n",
    "print(\"Precision: \",precision,\"Recall: \", recall )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "163c9a11",
   "metadata": {},
   "source": [
    "Now our 5-detector does not look as shiny as it did when we looked at its accuracy. When  it  claims  an  image  represents  a  5,  it  is  correct  only  83.7%  of  the  time.  Moreover, it only detects 65.11% of the 5s."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41b1bae8",
   "metadata": {},
   "source": [
    "It is often convenient to combine precision and recall into a single metric called the F1 score, in particular if you need a simple way to compare two classifiers. The F1 score is the harmonic mean of precision and recall. Whereas the regular mean treats  all  values  equally,  the  harmonic  mean  gives  much  more  weight  to  low  values.As a result, the classifier will only get a high F1 score if both recall and precision are high.\n",
    "<br/>\n",
    "f1 = 2\\*{(precision\\*recall)/(precision+recall)}\n",
    "<br/>\n",
    "To calculate f1 score we can call *fl_score* fuction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d096af8c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7325171197343846\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "\n",
    "f1 = f1_score(y_train_5, y_train_pred)\n",
    "print(f1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9daaa06",
   "metadata": {},
   "source": [
    "The F1 score favors classifiers that have similar precision and recall. This is not always what we want: in some contexts we mostly care about precision, and in other contexts we really care about recall. For example, if we trained a classifier to detect videos  that  are  safe  for  kids,  you  would  probably  prefer  a  classifier  that  rejects  many good videos (low recall) but keeps only safe ones (high precision), rather than a clas‐sifier  that  has  a  much  higher  recall  but  lets  a  few  really  bad  videos  show  up  in  your product (in such cases, you may even want to add a human pipeline to check the clas‐sifier’s  video  selection).  On  the  other  hand,  suppose  you  train  a  classifier  to  detect shoplifters  in  surveillance  images:  it  is  probably  fine  if  your  classifier  has  only  30% precision  as  long  as  it  has  99%  recall  (sure,  the  security  guards  will  get  a  few  false alerts, but almost all shoplifters will get caught).Unfortunately,  you  can’t  have  it  both  ways:  increasing  precision  reduces  recall,  and vice versa. This is called the precision/recall trade-off."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e475afd0",
   "metadata": {},
   "source": [
    "## Precision/Recall Trade-off\n",
    "\n",
    "To understand this trade off, let’s look at how the *SGDClassifier* makes its classification decisions. For each instance, it computes a score based on a decision function. If that  score  is  greater  than  a  threshold,  it  assigns  the  instance  to  the  positive  class; otherwise  it  assigns  it  to  the  negative  class.\n",
    "<br/>\n",
    "In this precision/recall trade-off, images are ranked by their classifier score,and those above the chosen decision threshold are considered positive; the higher thethreshold, the lower the recall, but (in general) the higher the precision.\n",
    "<br/>\n",
    "Scikit-Learn does not let us set the threshold directly, but it does give you access to the  decision  scores  that  it  uses  to  make  predictions. Instead  of  calling  the  classifier’s *predict()*  method,  we  can  call  its  *decision_function()*  method,  which  returns  a score  for  each  instance,  and  then  use  any  threshold  you  want  to  make  predictions based on those scores."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "1d540d6e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2164.22030239]\n",
      "[ True]\n"
     ]
    }
   ],
   "source": [
    "y_scores = sgd_clf.decision_function([some_digit])\n",
    "print(y_scores)\n",
    "\n",
    "threshold = 0\n",
    "y_some_digit_pred = (y_scores > threshold)\n",
    "print(y_some_digit_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27af5d17",
   "metadata": {},
   "source": [
    "The *SGDClassifier* uses a threshold equal to 0, so the previous code returns the samer esult as the *predict()* method (i.e., True). Let’s raise the threshold."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b1a4975f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[False]\n"
     ]
    }
   ],
   "source": [
    "threshold = 80000\n",
    "y_some_digit_pred = (y_scores > threshold)\n",
    "print(y_some_digit_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21b2c600",
   "metadata": {},
   "source": [
    "This  confirms  that  raising  the  threshold  decreases  recall.  The  image  actually  represents a 5, and the classifier detects it when the threshold is 0, but it misses it when thethreshold is increased to 8,000.\n",
    "<br/>\n",
    "To decide which threshold to use we first, use  the  *cross_val_predict()* function to get the scores of all instances in the training set, but this time specifying that we want to return decision scores instead of predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "01d10cc3",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3,\n",
    "                            method=\"decision_function\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6282cfe3",
   "metadata": {},
   "source": [
    "With these scores, use the *precision_recall_curve()* function to compute precision and recall for all possible thresholds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "a3fe2239",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "\n",
    "precisions, recalls, thresholds = precision_recall_curve(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba47a1fa",
   "metadata": {},
   "source": [
    "Finally, using Matplotlib to plot precision and recall as functions of the threshold value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9d47287d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_precision_recall_vs_threshold(precisions, recalls, thresholds):\n",
    "    plt.plot(thresholds, precisions[:-1], \"b--\", label=\"Precision\", linewidth=2)\n",
    "    plt.plot(thresholds, recalls[:-1], \"g-\", label=\"Recall\", linewidth=2)\n",
    "    plt.legend(loc=\"center right\", fontsize=16) # Not shown in the book\n",
    "    plt.xlabel(\"Threshold\", fontsize=16)        # Not shown\n",
    "    plt.grid(True)                              # Not shown\n",
    "    plt.axis([-50000, 50000, 0, 1])             # Not shown\n",
    "\n",
    "    \n",
    "recall_90_precision = recalls[np.argmax(precisions >= 0.90)]\n",
    "threshold_90_precision = thresholds[np.argmax(precisions >= 0.90)]\n",
    "\n",
    "\n",
    "plt.figure(figsize=(8, 4))                                                                  # Not shown\n",
    "plot_precision_recall_vs_threshold(precisions, recalls, thresholds)\n",
    "plt.plot([threshold_90_precision, threshold_90_precision], [0., 0.9], \"r:\")                 # Not shown\n",
    "plt.plot([-50000, threshold_90_precision], [0.9, 0.9], \"r:\")                                # Not shown\n",
    "plt.plot([-50000, threshold_90_precision], [recall_90_precision, recall_90_precision], \"r:\")# Not shown\n",
    "plt.plot([threshold_90_precision], [0.9], \"ro\")                                             # Not shown\n",
    "plt.plot([threshold_90_precision], [recall_90_precision], \"ro\")                             # Not shown                                             # Not shown\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "532405c2",
   "metadata": {},
   "source": [
    "The precision curve is bumpier because precision may sometimes go down  when  we  raise  the  threshold  (although  in  general  it  will  go up). On the other hand, recall can only go down when the threshold is increased, which explains why its curve looks smooth."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3343f436",
   "metadata": {},
   "source": [
    "Another  way  to  select  a  good  precision/recall  trade-off  is  to  plot  precision  directlyagainst recall."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "831bcf47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_precision_vs_recall(precisions, recalls):\n",
    "    plt.plot(recalls, precisions, \"b-\", linewidth=2)\n",
    "    plt.xlabel(\"Recall\", fontsize=16)\n",
    "    plt.ylabel(\"Precision\", fontsize=16)\n",
    "    plt.axis([0, 1, 0, 1])\n",
    "    plt.grid(True)\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "plot_precision_vs_recall(precisions, recalls)\n",
    "plt.plot([recall_90_precision, recall_90_precision], [0., 0.9], \"r:\")\n",
    "plt.plot([0.0, recall_90_precision], [0.9, 0.9], \"r:\")\n",
    "plt.plot([recall_90_precision], [0.9], \"ro\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6373e1c",
   "metadata": {},
   "source": [
    "We  can  see  that  precision  really  starts  to  fall  sharply  around  80%  recall.  We  will probably  want  to  select  a  precision/recall  trade-off  just  before  that  drop for example, at around 60% recall. But of course, the choice depends on our project. Suppose we decide to aim for 90% precision. We look up the first plot and find that We need to use a threshold of about 8,000. To be more precise we can search for the lowest threshold that gives us at least 90% precision (np.argmax() will give the first index of the maximum value, which in this case means the first True value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b06a92f2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3370.0194991439557\n"
     ]
    }
   ],
   "source": [
    "threshold_90_precision = thresholds[np.argmax(precisions >= 0.90)]\n",
    "print(threshold_90_precision)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b075eade",
   "metadata": {},
   "source": [
    "To  make  predictions  (on  the  training  set  for  now),  instead  of  calling  the  classifier’s *predict()* method, we can run this code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "0c82dc2f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[False False False ...  True False False]\n"
     ]
    }
   ],
   "source": [
    "y_train_pred_90 = (y_scores >= threshold_90_precision)\n",
    "print(y_train_pred_90)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "758e75a8",
   "metadata": {},
   "source": [
    "Let’s check these predictions’ precision and recall."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8bd2dca7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9000345901072293"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision_score(y_train_5, y_train_pred_90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "ddafb46e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4799852425751706"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(y_train_5, y_train_pred_90)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb6e0509",
   "metadata": {},
   "source": [
    "Great, we have a 90% precision classifier! As we can see, it is fairly easy to create a classifier with virtually any precision we want: just set a high enough threshold, and we're  done.  But  wait,  not  so  fast.  A  high-precision  classifier  is  not  very  useful  if  its recall is too low.\n",
    "<br/>\n",
    "If  someone  says,  “Let’s  reach  99%  precision,”  you  should  ask,  “At what recall?”"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f322103",
   "metadata": {},
   "source": [
    "## The ROC Curve\n",
    "\n",
    "The  receiver  operating  characteristic  (ROC)  curve  is  another  common  tool  used  with binary  classifiers.  It  is  very  similar  to  the  precision/recall  curve,  but  instead  of  plotting precision versus recall, the ROC curve plots the true positive rate (another name for recall) against the false positive rate (FPR). The FPR is the ratio of negative instances  that  are  incorrectly  classified  as  positive.  It  is  equal  to  1  –  the  true  negative  rate(TNR), which is the ratio of negative instances that are correctly classified as negative.The TNR is also called specificity. Hence, the ROC curve plots sensitivity (recall) versus 1 – specificity.\n",
    "<br/>\n",
    "To  plot  the  ROC  curve,  We  first  use  the  *roc_curve()*  function  to  compute  the  TPR and FPR for various threshold values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "1f54675b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "\n",
    "fpr, tpr, thresholds = roc_curve(y_train_5, y_scores)\n",
    "\n",
    "def plot_roc_curve(fpr, tpr, label=None):\n",
    "    plt.plot(fpr, tpr, linewidth=2, label=label)\n",
    "    plt.plot([0, 1], [0, 1], 'k--') # dashed diagonal\n",
    "    plt.axis([0, 1, 0, 1])                                    # Not shown in the book\n",
    "    plt.xlabel('False Positive Rate (Fall-Out)', fontsize=16) # Not shown\n",
    "    plt.ylabel('True Positive Rate (Recall)', fontsize=16)    # Not shown\n",
    "    plt.grid(True)                                            # Not shown\n",
    "\n",
    "plt.figure(figsize=(8, 6))                                    # Not shown\n",
    "plot_roc_curve(fpr, tpr)\n",
    "fpr_90 = fpr[np.argmax(tpr >= recall_90_precision)]           # Not shown\n",
    "plt.plot([fpr_90, fpr_90], [0., recall_90_precision], \"r:\")   # Not shown\n",
    "plt.plot([0.0, fpr_90], [recall_90_precision, recall_90_precision], \"r:\")  # Not shown\n",
    "plt.plot([fpr_90], [recall_90_precision], \"ro\")               # Not shown                                  # Not shown\n",
    "plt.show()\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d468d0d4",
   "metadata": {},
   "source": [
    "Once  again  there  is  a  trade-off:  the  higher  the  recall  (TPR),  the  more  false  positives(FPR)  the  classifier  produces.  The  dotted  line  represents  the  ROC  curve  of  a  purely random classifier; a good classifier stays as far away from that line as possible (toward the top-left corner).\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5172fbf3",
   "metadata": {},
   "source": [
    "One way to compare classifiers is to measure the area under the curve (AUC). A perfect classifier will have a ROC AUC equal to 1, whereas a purely random classifier will have a ROC AUC equal to 0.5. Scikit-Learn provides a function to compute the ROCAUC."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "c6ed7f36",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9604938554008616"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "roc_auc_score(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3c997cd",
   "metadata": {},
   "source": [
    "Since  the  ROC  curve  is  so  similar  to  the  precision/recall  (PR)curve, you may wonder how to decide which one to use. As a rule of  thumb,  you  should  prefer  the  PR  curve  whenever  the  positive class  is  rare  or  when  you  care  more  about  the  false  positives  than the  false  negatives.  Otherwise,  use  the  ROC  curve.  For  example, looking at the previous ROC curve (and the ROC AUC score), you may  think  that  the  classifier  is  really  good.  But  this  is  mostly because  there  are  few  positives  (5s)  compared  to  the  negatives(non-5s). In contrast, the PR curve makes it clear that the classifierhas  room  for  improvement  (the  curve  could  be  closer  to  the  top-left corner)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4377789",
   "metadata": {},
   "source": [
    "Let’s  now  train  a  *RandomForestClassifier*  and  compare  its  ROC  curve  and  ROCAUC  score  to  those  of  the  SGDClassifier.  First,  you  need  to  get  scores  for  each instance  in  the  training  set.  But  due  to  the  way  it  works, the  *RandomForestClassifier*  class  does  not  have  a  *decision_function()*  method.  Instead, it Classification\n",
    "has  a  *predict_proba()*  method.  Scikit-Learn  classifiers  generally  have  one  or  the other,  or  both.  The  *predict_proba()*  method  returns  an  array  containing  a  row  perinstance  and  a  column  per  class,  each  containing  the  probability  that  the  giveninstance belongs to the given class (e.g., 70% chance that the image represents a 5)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "10df8e39",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "forest_clf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "y_probas_forest = cross_val_predict(forest_clf, X_train, y_train_5, cv=3,\n",
    "                                    method=\"predict_proba\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c34db1a",
   "metadata": {},
   "source": [
    "The  *roc_curve()* function  expects  labels  and  scores,  but  instead  of  scores  we  can give it class probabilities. Let’s use the positive class’s probability as the score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "7e1fa18b",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores_forest = y_probas_forest[:, 1] # score = proba of positive class\n",
    "fpr_forest, tpr_forest, thresholds_forest = roc_curve(y_train_5, y_scores_forest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "21ec8c9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "recall_for_forest = tpr_forest[np.argmax(fpr_forest >= fpr_90)]\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, \"b:\", linewidth=2, label=\"SGD\")\n",
    "plot_roc_curve(fpr_forest, tpr_forest, \"Random Forest\")\n",
    "plt.plot([fpr_90, fpr_90], [0., recall_90_precision], \"r:\")\n",
    "plt.plot([0.0, fpr_90], [recall_90_precision, recall_90_precision], \"r:\")\n",
    "plt.plot([fpr_90], [recall_90_precision], \"ro\")\n",
    "plt.plot([fpr_90, fpr_90], [0., recall_for_forest], \"r:\")\n",
    "plt.plot([fpr_90], [recall_for_forest], \"ro\")\n",
    "plt.grid(True)\n",
    "plt.legend(loc=\"lower right\", fontsize=16)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b65df4e",
   "metadata": {},
   "source": [
    "The RandomForestClassifier’s ROC curve looks much better  than  the  SGDClassifier’s:  it  comes  much  closer  to  the  top-left  corner.  As  a result, its ROC AUC score is also significantly better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "660d193f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9983436731328145"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_train_5, y_scores_forest)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01d59644",
   "metadata": {},
   "source": [
    "Finally, let's print the accuracy and precision."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "dd3613f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9905083315756169"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_pred_forest = cross_val_predict(forest_clf, X_train, y_train_5, cv=3)\n",
    "precision_score(y_train_5, y_train_pred_forest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "ced66404",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8662608374838591"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(y_train_5, y_train_pred_forest)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d440786",
   "metadata": {},
   "source": [
    "we now know how to train binary classifiers, choose the appropriate metric for our task, evaluate our classifiers using cross-validation, select the precision/recall trade-off that fits our needs, and use ROC curves and ROC AUC scores to compare vari‐ous models. Now let’s try to detect more than just the 5s."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bbbf9f7",
   "metadata": {},
   "source": [
    "## Multiclass Classification\n",
    "\n",
    "Whereas  binary  classifiers  distinguish  between  two  classes,  multiclass classifiers  (also called multinomial classifiers) can distinguish between more than two classes. Some algorithms (such as SGD classifiers, Random Forest classifiers, and naive Bayes classifiers)  are  capable  of  handling  multiple  classes  natively.  Others  (such  as  LogisticRegression or Support Vector Machine classifiers) are strictly binary classifiers. However, there are various strategies that you can use to perform multiclass classificationwith multiple binary classifiers. One way to create a system that can classify the digit images into 10 classes (from 0 to 9)  is  to  train  10  binary  classifiers,  one  for  each  digit  (a  0-detector,  a  1-detector,  a  2-detector, and so on). Then when we want to classify an image, we get the decision score from each classifier for that image and we select the class whose classifier outputs the highest score. This is called the one-versus-the-rest (OvR) strategy (also calledone-versus-all). Another  strategy  is  to  train  a  binary  classifier  for  every  pair  of  digits:  one  to  distinguish  0s  and  1s,  another  to  distinguish  0s  and  2s,  another  for  1s  and  2s,  and  so  on. This  is  called  the  one-versus-one  (OvO)  strategy.  If  there  are  N  classes,  we  need  totrain  N  ×  (N  –  1)  /  2  classifiers.  For  the  MNIST  problem,  this  means  training  45 binary  classifiers!  When  we  want  to  classify  an  image,  we  have  to  run  the  image through  all  45  classifiers  and  see  which  class  wins  the  most  duels.  The  main  advantage of OvO is that each classifier only needs to be trained on the part of the trainingset for the two classes that it must distinguish. Some  algorithms  (such  as  Support  Vector  Machine  classifiers)  scale  poorly  with  the size  of  the  training  set.  For  these  algorithms  OvO  is  preferred  because  it  is  faster  to train many classifiers on small training sets than to train few classifiers on large training sets. For most binary classification algorithms, however, OvR is preferred."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80296745",
   "metadata": {},
   "source": [
    "Scikit-Learn detects when we try to use a binary classification algorithm for a multi‐class  classification  task,  and  it  automatically  runs  OvR  or  OvO,  depending  on  the algorithm.  Let’s  try  this  with  a  Support  Vector  Machine  classifier, using the *sklearn.svm.SVC* class:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "efc5d58f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "\n",
    "svm_clf = SVC(gamma=\"auto\", random_state=42)\n",
    "svm_clf.fit(X_train[:1000], y_train[:1000])\n",
    "svm_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2dd2f667",
   "metadata": {},
   "source": [
    "That  was  easy!  This  code  trains  the  SVC  on  the  training  set  using  the  original  target classes   from   0   to   9   (y_train),   instead   of   the   5-versus-the-rest   target   classes(y_train_5). Then it makes a prediction (a correct one in this case). Under the hood, Scikit-Learn actually used the OvO strategy: it trained 45 binary classifiers, got their decision scores for the image, and selected the class that won the most duels.If  we  call  the  *decision_function()*  method,  we  will  see  that  it  returns  10  scores per instance (instead of just 1). That’s one score per class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "8976a585",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 2.81585438,  7.09167958,  3.82972099,  0.79365551,  5.8885703 ,\n",
       "         9.29718395,  1.79862509,  8.10392157, -0.228207  ,  4.83753243]])"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "some_digit_scores = svm_clf.decision_function([some_digit])\n",
    "some_digit_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "aa08c92c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.argmax(some_digit_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "37f6c636",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=uint8)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_clf.classes_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "04d2e5ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_clf.classes_[5]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d0db6448",
   "metadata": {},
   "source": [
    "When  a  classifier  is  trained,  it  stores  the  list  of  target  classes  in  its *classes_* attribute, ordered by value. In this case, the index of each class  in  the  *classes_*  array  conveniently  matches  the  class  itself (e.g., the class at index 5 happens to be class 5), but in general we won’t be so lucky."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51c425f8",
   "metadata": {},
   "source": [
    "If  we  want  to  force  Scikit-Learn  to  use  one-versus-one  or  one-versus-the-rest,  we can  use  the  *OneVsOneClassifier*  or  *OneVsRestClassifier*  classes.  Simply  create  an instance  and  pass  a  classifier  to  its  constructor  (it  does  not  even  have  to  be  a  binaryclassifier).  For  example,  this  code  creates  a  multiclass  classifier  using  the  OvR  strat‐egy, based on an SVC:\n",
    "<br/>\n",
    "**!!! The code below will take very long time to compute so they are commented out but can be executed whenever necessary (only a small sample is taken for faster exection**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "b07a26b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OneVsRestClassifier(estimator=SVC(gamma='auto', random_state=42))"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.multiclass import OneVsRestClassifier\n",
    "ovr_clf = OneVsRestClassifier(SVC(gamma=\"auto\", random_state=42))\n",
    "ovr_clf.fit(X_train[:100], y_train[:100])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "be78b588",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ovr_clf.predict([some_digit])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "b223f44d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ovr_clf.estimators_)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b77e99a4",
   "metadata": {},
   "source": [
    "Training an *SGDClassifier* (or a RandomForestClassifier) is just as easy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "7e11beff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.fit(X_train[:100], y_train[:100])\n",
    "sgd_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4c5ff8b",
   "metadata": {},
   "source": [
    "This time Scikit-Learn did not have to run OvR or OvO because SGD classifiers candirectly  classify  instances  into  multiple  classes.  The  decision_function()  methodnow returns one value per class. Let’s look at the score that the SGD classifier assignedto each class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "7866d5d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-20722091.53681121, -18169387.23610712, -15721140.16708742,\n",
       "         -1092999.6438692 , -39289471.34882002,   6294279.54476251,\n",
       "         -3878406.8911707 , -24314511.98221525, -14573277.07882209,\n",
       "        -23894139.33910285]])"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.decision_function([some_digit])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37252ceb",
   "metadata": {},
   "source": [
    "We can see that the classifier is fairly confident about its prediction: almost all scores are largely negative, while class 5 has a score of 2412.5. The model has a slight doubtregarding class 3, which gets a score of 573.5. Now of course we want to evaluate this classifier.  As  usual,  we  can  use  cross-validation.  Use  the  *cross_val_score()*  function to evaluate the SGDClassifier’s accuracy. **Only a sample of the data is taken so the values int this paragraph may differ with the output.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "0c094d64",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.73529412, 0.75757576, 0.84848485])"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(sgd_clf, X_train[:100], y_train[:100], cv=3, scoring=\"accuracy\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0b233a1",
   "metadata": {},
   "source": [
    "It gets over 84% on all test folds. If we used a random classifier, we would get 10% accuracy, so this is not such a bad score, but we can still do much better. Simply scaling the input increases accuracy above 89%."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "be9f0ff6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.70588235, 0.66666667, 0.81818182])"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()\n",
    "X_train_scaled = scaler.fit_transform(X_train.astype(np.float64))\n",
    "cross_val_score(sgd_clf, X_train_scaled[:100], y_train[:100], cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12dd0604",
   "metadata": {},
   "source": [
    "## Error Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a5c5e57",
   "metadata": {},
   "source": [
    "We will explore  data  preparation  options,  try out multiple  models  (shortlisting  the  best  ones  and  fine-tuning  their  hyperparametersusing  GridSearchCV),  and  automate  as  much  as  possible.  Here,  we  will  assume  that we have found a promising model and we want to find ways to improve it. One way to do this is to analyze the types of errors it makes."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93472e1f",
   "metadata": {},
   "source": [
    "First,   look   at   the   confusion   matrix.   We   need   to   make   predictions   using   the *cross_val_predict()* function, then call the *confusion_matrix()* function, just like we did earlier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "29d711d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[11,  0,  1,  0,  0,  0,  1,  0,  0,  0],\n",
       "       [ 0, 12,  0,  0,  0,  0,  2,  0,  0,  0],\n",
       "       [ 0,  0,  4,  2,  0,  0,  0,  0,  0,  0],\n",
       "       [ 1,  0,  0,  8,  1,  0,  0,  0,  1,  0],\n",
       "       [ 0,  1,  0,  1,  5,  2,  0,  1,  0,  1],\n",
       "       [ 0,  0,  0,  2,  0,  3,  0,  0,  0,  0],\n",
       "       [ 0,  0,  0,  0,  0,  0, 11,  0,  0,  0],\n",
       "       [ 0,  1,  0,  0,  1,  0,  0,  8,  0,  0],\n",
       "       [ 2,  0,  0,  0,  0,  0,  0,  0,  6,  0],\n",
       "       [ 0,  0,  0,  1,  0,  0,  1,  3,  1,  5]])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_pred = cross_val_predict(sgd_clf, X_train_scaled[:100], y_train[:100], cv=3)\n",
    "conf_mx = confusion_matrix(y_train[:100], y_train_pred)\n",
    "conf_mx"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d202a82b",
   "metadata": {},
   "source": [
    "That’s a lot of numbers. It’s often more convenient to look at an image representation of the confusion matrix, using Matplotlib’s *matshow()* function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "0c85c2bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.matshow(conf_mx, cmap=plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63d891ec",
   "metadata": {},
   "source": [
    "This  confusion  matrix  looks  pretty  good,  since  most  images  are  on  the  main  diagonal, which means that they were classified correctly. The 5s look slightly darker than the other digits, which could mean that there are fewer images of 5s in the dataset or that  the  classifier  does  not  perform  as  well  on  5s  as  on  other  digits.  In  fact,  we  can verify that both are the case. Let’s focus the plot on the errors. First, we need to divide each value in the confusionmatrix by the number of images in the corresponding class so that we can compare error  rates  instead  of  absolute  numbers  of  errors  (which  would  make  abundant classes look unfairly bad)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "929e92a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "row_sums = conf_mx.sum(axis=0, keepdims=True)\n",
    "norm_conf_mx = conf_mx / row_sums"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c85989b",
   "metadata": {},
   "source": [
    "Fill the diagonal with zeros to keep only the errors, and plot the result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "3c0ae9c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "np.fill_diagonal(norm_conf_mx, 0)\n",
    "plt.matshow(norm_conf_mx, cmap=plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fee0603",
   "metadata": {},
   "source": [
    "We can clearly see the kinds of errors the classifier makes. Remember that rows represent actual classes, while columns represent predicted classes. The column for class 3  is  quite  bright,  which  tells  us  that  many  images  get  misclassified  as  3s.  However,the  row  for  class  6  is  not  that  bad,  telling  us  that  actual  6s  in  general  get  properlyclassified  as  6s.  As  we  can  see,  the  confusion  matrix  is  not  necessarily  symmetrical. We can also see that 3s and 4s often get confused (in both directions)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "46e8a770",
   "metadata": {},
   "source": [
    "Analyzing  individual  errors  can  also  be  a  good  way  to  gain  insights  on  what  our classifier  is  doing  and  why  it  is  failing,  but  it  is  more  difficult  and  time-consuming. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd44f708",
   "metadata": {},
   "source": [
    "## Multilabel Classification"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a06a057",
   "metadata": {},
   "source": [
    "Until now each instance has always been assigned to just one class. In some cases we may want classifier to output multiple classes for each instance. Consider a face-recognition  classifier:  what  should  it  do  if  it  recognizes  several  people  in  the  same picture?  It  should  attach  one  tag  per  person  it  recognizes.  Say  the  classifier  has  been trained  to  recognize  three  faces,  Alice,  Bob,  and  Charlie.  Then  when  the  classifier  is shown  a  picture  of  Alice  and  Charlie,  it  should  output  [1,  0,  1]  (meaning  “Alice  yes, Bob no, Charlie yes”). Such a classification system that outputs multiple binary tags is called a multilabel classification system."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "f7047482",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "y_train_large = (y_train >= 7)\n",
    "y_train_odd = (y_train % 2 == 1)\n",
    "y_multilabel = np.c_[y_train_large, y_train_odd]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "20e07707",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_clf = KNeighborsClassifier()\n",
    "knn_clf.fit(X_train, y_multilabel)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2c908db",
   "metadata": {},
   "source": [
    "This  code  creates  a  *y_multilabel*  array  containing  two  target  labels  for  each  digit image: the first indicates whether or not the digit is large (7, 8, or 9), and the second indicates  whether  or  not  it  is  odd.  The  next  lines  create  a  *KNeighborsClassifier* instance  (which  supports  multilabel  classification,  though  not  all  classifiers  do),  and we  train  it  using  the  multiple  targets  array.  Now  we  can  make  a  prediction,  and notice that it outputs two labels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "939b4f17",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False,  True]])"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "586ffbb1",
   "metadata": {},
   "source": [
    "And it gets it right! The digit 5 is indeed not large (False) and odd (True)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "65ef9f4b",
   "metadata": {},
   "source": [
    "There are many ways to evaluate a multilabel classifier, and selecting the right metricreally  depends  on  your  project.  One  approach  is  to  measure  the  F1  score  for  each individual  label  (or  any  other  binary  classifier  metric  discussed  earlier),  then  simply compute the average score. This code computes the average F1 score across all labels:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "d097c383",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_knn_pred = cross_val_predict(knn_clf, X_train, y_multilabel, cv=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "363144f0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.976410265560605"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(y_multilabel, y_train_knn_pred, average=\"macro\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8bcc86e",
   "metadata": {},
   "source": [
    "This  assumes  that  all  labels  are  equally  important,  however,  which  may  not  be  the case.  In  particular,  if  we  have  many  more  pictures  of  Alice  than  of  Bob  or  Charlie, we  may  want  to  give  more  weight  to  the  classifier’s  score  on  pictures  of  Alice.  One simple  option  is  to  give  each  label  a  weight  equal  to  its  support  (i.e.,  the  number  of instances  with  that  target  label).  To  do  this,  simply  set  average= \"weighted\"  in  the preceding code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "eac358a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "#f1_score(y_multilabel, y_train_knn_pred, average=\"weighted\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1d22ec2",
   "metadata": {},
   "source": [
    "## Multioutput Classification\n",
    "\n",
    "The last type of classification task we are going to discuss here is called multioutput–multiclass classification (or simply multioutput classification). It is simply a generalization  of  multilabel  classification  where  each  label  can  be  multiclass  (i.e.,  it  can  have more than two possible values). \n",
    "<br/>\n",
    "To  illustrate  this,  let’s  build  a  system  that  removes  noise  from  images.  It  will  take  as input  a  noisy  digit  image,  and  it  will  (hopefully)  output  a  clean  digit  image,  represented  as  an  array  of  pixel  intensities,  just  like  the  MNIST  images.  Notice  that  theclassifier’s output is multilabel (one label per pixel) and each label can have multiplevalues  (pixel  intensity  ranges  from  0  to  255).  It  is  thus  an  example  of  a  multioutputclassification system."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9fe44ba2",
   "metadata": {},
   "source": [
    "Let’s  start  by  creating  the  training  and  test  sets  by  taking  the  MNIST  images  and adding  noise  to  their  pixel  intensities  with  NumPy’s  *randint()*  function.  The  target images will be the original images."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "3f83e6ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "noise = np.random.randint(0, 100, (len(X_train), 784))\n",
    "X_train_mod = X_train + noise\n",
    "noise = np.random.randint(0, 100, (len(X_test), 784))\n",
    "X_test_mod = X_test + noise\n",
    "y_train_mod = X_train\n",
    "y_test_mod = X_test"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb53aeca",
   "metadata": {},
   "source": [
    "Let’s take a peek at an image from the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "850d2b1a",
   "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"
    }
   ],
   "source": [
    "some_index = 9999\n",
    "plt.subplot(121); plot_digit(X_test_mod[some_index])\n",
    "plt.subplot(122); plot_digit(y_test_mod[some_index])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc76d676",
   "metadata": {},
   "source": [
    "On the left is the noisy input image, and on the right is the clean target image. Now let’s train the classifier and make it clean this image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "ffa8179a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_clf.fit(X_train_mod, y_train_mod)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "a7b64c75",
   "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"
    }
   ],
   "source": [
    "clean_digit = knn_clf.predict([X_test_mod[some_index]])\n",
    "plot_digit(clean_digit)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "389c7cc4",
   "metadata": {},
   "source": [
    "Looks  close  enough  to  the  target!  This  concludes  our  tour  of  classification.  We now know how to select good metrics for classification tasks, pick the appropriate  precision/recall  trade-off,  compare  classifiers,  and  more  generally  build  good classification systems for a variety of tasks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "63e359aa",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.6 64-bit ('venv': venv)",
   "language": "python",
   "name": "python396jvsc74a57bd0c9b08adf87d7192cfe942781fd54415438870c456d89c858e0a76f55356c9d48"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}