{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "6xoz29ZbhlWk"
   },
   "source": [
    "Authors: Aurélien Garivier and Yohann De Castro\n",
    "\n",
    "# Dimension reduction for the MNIST dataset\n",
    "\n",
    "\n",
    "The Mnist dataset provides a (somewhat old but) very classical example of Machine Learning. The goal is simple and useful: recognize hand-written digits. This was an important challenge for the postal companies in the 1980s.  \n",
    "\n",
    "It is available in sklearn (and on the webpage of <a href=\"http://yann.lecun.com/exdb/mnist/\">Yann LeCun</a>, a very famous researcher in Machine Learning in general and neural networks in particular). \n",
    "\n",
    "</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "pygIPffThlWm",
    "outputId": "a375cb2f-1adf-46d9-8373-f97bac3dcd8c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "# loading a lot of facilities for numerical computations and graphs\n",
    "%pylab inline \n",
    "pylab.rcParams['figure.figsize'] = (20, 6) # to have larger plot\n",
    "seed(240979) #initialize random number generator\n",
    "\n",
    "# set working directory\n",
    "workingdir = './'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "moFV5JWfhlWq"
   },
   "source": [
    "## 1. Loading the data set\n",
    "\n",
    "### 1.1 Description of the data set\n",
    "The MNIST data contains $n\\_tot=70000$ images of size $dimX\\times dimY=28\\times28$, and their *labels* (a digit between $0$ and $9$). \n",
    "The images are stored as lines of the matrix *X*. \n",
    "Each image can be reshaped as a matrix $im \\in \\mathcal{M}_{dimX, dimY}(\\mathbb{R})$, where $im[i,j]$ is the intensity level of pixel (i,j): 0 means 'white', 255 means 'black'.\n",
    "\n",
    "The labels, stored in *y*, are numbers (here, we cast them to integers) between $0$ and $9$. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 89
    },
    "colab_type": "code",
    "id": "weoZ7FrOhlWr",
    "outputId": "02082a1d-4f9b-4911-ce1d-fa19b8536caa"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 70000 images of size 28x28\n",
      "Number of classes: 10\n",
      "Classes:\n",
      "['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_openml\n",
    "X, y = fetch_openml('mnist_784', version=1, return_X_y=True)\n",
    "\n",
    "X = X.values\n",
    "y = y.values\n",
    "\n",
    "# defining general variables for use throuhout the notebook  \n",
    "n_tot, p = shape(X)\n",
    "dimX = int(sqrt(p))\n",
    "dimY = dimX \n",
    "nc = len(unique(y)) # number of classes\n",
    "\n",
    "print(\"Loaded %d images of size %dx%d\"%(n_tot, dimX, dimY))\n",
    "print(\"Number of classes: %d\"%(nc))\n",
    "print(\"Classes:\")\n",
    "print(sorted(unique(y)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Ncg6bV2JhlWt"
   },
   "source": [
    "### 1.2 What does the data look like?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 776
    },
    "colab_type": "code",
    "id": "5XxvcmyghlWt",
    "outputId": "af5c91b3-b863-43c6-9632-2c47245079d3"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Showing one image \n",
    "figure(figsize = (7,7))\n",
    "ii = random.randint(0, n_tot-1)\n",
    "plt.imshow(reshape(X[ii], [dimX, dimY], order='C'), cmap=\"Greys\", interpolation=\"none\") # or \"nearest\"\n",
    "ax = plt.gca()\n",
    "ax.set_xticks(arange(-0.5, dimX, 1))\n",
    "ax.set_yticks(arange(-0.5, dimY, 1))\n",
    "ax.set_xticklabels(arange(0, dimX+1, 1))\n",
    "ax.set_yticklabels(arange(0, dimY+1, 1))\n",
    "ax.grid(color='r', linestyle='--', linewidth=1)\n",
    "plt.show()\n",
    "\n",
    "# Showing a few images in the dataset with their labels\n",
    "nbRows = 5\n",
    "nbCols = 15\n",
    "I = [[0 for k in range(nbCols)] for j in range(nbRows)]\n",
    "I[0] = [random.randint(0, n_tot-1) for k in range(nbCols)]\n",
    "I[0][0] = ii\n",
    "M = np.concatenate([reshape(X[i], [dimX,dimY], order='C') for i in I[0]], axis=1)\n",
    "\n",
    "for j in range(1, nbRows):\n",
    "    I[j] = [random.randint(0, n_tot-1) for k in range(nbCols)]\n",
    "    M = concatenate([M, np.concatenate([np.reshape(X[i], [dimX, dimY], order='C') for i in I[j]], axis=1)], axis=0)\n",
    "plt.imshow(M, cmap=\"Greys\", interpolation=\"none\")\n",
    "for j in range(nbRows):\n",
    "    for k in range(nbCols):\n",
    "        plt.text(dimX*k+2, dimX*j+5, y[I[j][k]], fontsize=14, color=\"red\")\n",
    "ax = plt.gca()\n",
    "ax.set_xticks(np.arange(-0.5, dimX*nbCols, dimX))\n",
    "ax.set_yticks(np.arange(-0.5, dimY*nbRows, dimY))\n",
    "ax.set_xticklabels([])\n",
    "ax.set_yticklabels([])\n",
    "ax.grid(color='k', linestyle='-', linewidth=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "8G3Fk6kAhlWw"
   },
   "source": [
    "## Nearest-neighbor classifier\n",
    "Distance-based methods consider some notion of *distance* in the feature space, and assume that if to feature vectors are close, then the corresponding targets are likely to be equal.\n",
    "\n",
    "### 7.1 What distance?\n",
    "The first problem, of course, is to provide the feature space with a relevant distance. There are a few default choices, and in particular the Euclidian distance (o $L^2$ distance) in $\\mathbb{R}^p$ :\n",
    "$$d(x,y) = \\sqrt{\\sum_{i=1}^p (x_i-y_i)^2}$$\n",
    "\n",
    "Is this distance really relevant on images? \n",
    "A better alternative would be the *earthmoving distance*, but it is much more costly in computation time and we will see that the $L^2$ distance already performs well.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "LpWC93U5hlWw"
   },
   "source": [
    "If you code it by yourself, the result will be good butthe computation time ridiculously high...\n",
    "\n",
    "The classifiers of scikitlearn are usually well coded, and much more efficient than a few \"quick and dirty\" lines.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 53
    },
    "colab_type": "code",
    "id": "nhZAupUJhlWx",
    "outputId": "0593d4e3-4b82-4312-f8a7-eb38a7631c7d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proportion of misclassified images in testing set: 0.0542\n",
      "elapsed: 1.35652 seconds\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "\n",
    "n_train = 10000\n",
    "I = random.choice(range(n_tot), n_train, replace=False)\n",
    "trainingX = X[I, :] \n",
    "trainingTarget = y[I]\n",
    "\n",
    "# testing set: we may not use them, except at the end to assert the quality of the classifier\n",
    "mask = np.ones(n_tot, dtype=bool) #np.ones_like(a,dtype=bool)\n",
    "mask[I] = False\n",
    "testingX = X[mask, :]\n",
    "testingTarget = y[mask]\n",
    "\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "k = 7\n",
    "nearestNeighborsClassifier = KNeighborsClassifier(n_neighbors=k) # L2 distance is the default\n",
    "nearestNeighborsClassifier.fit(trainingX, trainingTarget)\n",
    "\n",
    "J = random.choice(range(len(testingTarget)), 5000, replace=False)\n",
    "\n",
    "start_time = time.time()\n",
    "pred = nearestNeighborsClassifier.predict(testingX[J])\n",
    "print(\"Proportion of misclassified images in testing set: %g\" %(mean(pred != testingTarget[J])))\n",
    "print(\"elapsed: %g seconds\"%(time.time()-start_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "iOgqpr3khlWz"
   },
   "source": [
    "# Using PCA to reduce dimensionality"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 53
    },
    "colab_type": "code",
    "id": "epsS__4rhlW0",
    "outputId": "0ce978b0-494b-4539-a6c2-24023cd6104f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PCA projection: Proportion of misclassified images in testing set: 0.0622\n",
      "elapsed: 0.710548 seconds\n"
     ]
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(n_components=15)\n",
    "trainingXred = pca.fit_transform(trainingX)\n",
    "\n",
    "nearestNeighborsClassifierRed = KNeighborsClassifier(n_neighbors=k) # L2 distance is the default\n",
    "nearestNeighborsClassifierRed.fit(trainingXred, trainingTarget)\n",
    "\n",
    "\n",
    "testingXred = pca.transform(testingX)\n",
    "start_time = time.time()\n",
    "predRed = nearestNeighborsClassifierRed.predict(testingXred[J])\n",
    "print(\"PCA projection: Proportion of misclassified images in testing set: %g\" %(mean(predRed != testingTarget[J])))\n",
    "print(\"elapsed: %g seconds\"%(time.time()-start_time))\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "42X8h-6mhlW2"
   },
   "source": [
    "# Using Random Projections to reduce dimensionality\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 53
    },
    "colab_type": "code",
    "id": "NhdgF1U5hlW3",
    "outputId": "0e940d4d-abfd-4064-dfd2-887e7e545013"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random projection: Proportion of misclassified images in testing set: 0.0974\n",
      "elapsed: 0.897575 seconds\n"
     ]
    }
   ],
   "source": [
    "d = 50\n",
    "W = random.normal(0, 1/sqrt(d), dimX*dimY*d).reshape((d, dimX*dimY))\n",
    "trainingXrproj = [dot(W, x) for x in trainingX]\n",
    "\n",
    "nearestNeighborsClassifierRproj = KNeighborsClassifier(n_neighbors=k) # L2 distance is the default\n",
    "nearestNeighborsClassifierRproj.fit(trainingXrproj, trainingTarget)\n",
    "\n",
    "\n",
    "testingXrproj = np.array([dot(W, x) for x in testingX])\n",
    "start_time = time.time()\n",
    "predRproj = nearestNeighborsClassifierRproj.predict(testingXrproj[J])\n",
    "print(\"Random projection: Proportion of misclassified images in testing set: %g\" %(mean(predRproj != testingTarget[J])))\n",
    "print(\"elapsed: %g seconds\"%(time.time()-start_time))\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "CLbpuDoWhlW5"
   },
   "source": [
    "## Questions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "jR6_Y-UlhlW5"
   },
   "source": [
    "### Q1. Plot the main eigenvectors (as images) and try to interpret them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 770
    },
    "colab_type": "code",
    "id": "ORi330Zxsu_v",
    "outputId": "29fd050d-34d6-492c-f613-faa1dc2b8119",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PCA explained variance is: [0.09844255 0.07207938 0.06179311 0.05295605 0.04942005 0.04397757\n",
      " 0.03189804 0.02919198 0.02766595 0.02318207 0.02120448 0.02074016\n",
      " 0.01743203 0.01692201 0.01552079]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 25 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# Plot 25 first eigenvectors\n",
    "pca25 = PCA(n_components=25)\n",
    "pca25.fit(trainingX)\n",
    "\n",
    "print(\"PCA explained variance is:\", pca.explained_variance_ratio_)\n",
    "\n",
    "fig, axs = plt.subplots(5,5, sharex=True, figsize=(12, 12))\n",
    "for i in range(25):\n",
    "  col = i % 5\n",
    "  col = int(col)\n",
    "  row = int((i - col) / 5)\n",
    "  axs[row,col].set_title('PCA eigenvec. %g' %(i+1))\n",
    "  axs[row,col].imshow(reshape(pca25.components_[i], [dimX, dimY], order='C'), cmap=\"Greys\", interpolation=\"none\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "NK2nezchhlW6"
   },
   "source": [
    "### Q2. For both PCA and random projections, plot the estimated classification performance as a function of the dimension $d$ in the projected space. Comment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 471
    },
    "colab_type": "code",
    "id": "SqfNBdaZtgJv",
    "outputId": "40e2fd53-5df1-4df9-bdfc-40ef77ff719b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Last explained variance ratio is: 0.0010296525807260868\n",
      "Last Classification Error PCA_Red is: 0.0502\n",
      "Last Classification Error Random_Red is: 0.0698\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# maximal dimension\n",
    "\n",
    "## PCA \n",
    "from sklearn.decomposition import PCA\n",
    "pcafull = PCA(n_components=dimX*dimY)\n",
    "pcafull.fit(trainingX)\n",
    "\n",
    "# PCA red classification performances \n",
    "\n",
    "Ds = [1+5*j for j in range(20)]\n",
    "N = len(Ds)\n",
    "cut = max(Ds)\n",
    "ClassificationErrorRedPCA=np.zeros(N)\n",
    "\n",
    "# Explained variance\n",
    "print('Last explained variance ratio is:', pcafull.explained_variance_ratio_[cut])\n",
    "\n",
    "\n",
    "for j in range(N):\n",
    "  d = Ds[j]\n",
    "  pca = PCA(n_components = d+1)\n",
    "  trainingXred = pca.fit_transform(trainingX)\n",
    "  nearestNeighborsClassifierRed = KNeighborsClassifier(n_neighbors=k) # L2 distance is the default\n",
    "  nearestNeighborsClassifierRed.fit(trainingXred, trainingTarget)\n",
    "  testingXred = pca.transform(testingX)\n",
    "  predRed = nearestNeighborsClassifierRed.predict(testingXred[J])\n",
    "  ClassificationErrorRedPCA[j] = mean(predRed != testingTarget[J])\n",
    "\n",
    "print('Last Classification Error PCA_Red is:', ClassificationErrorRedPCA[N-1])\n",
    "\n",
    "## Random Projection\n",
    "ClassificationErrorRedRandom=np.zeros(N)\n",
    "\n",
    "# Random marix\n",
    "Wn = random.normal(0, 1/sqrt(cut), dimX*dimY*cut).reshape((cut, dimX*dimY))\n",
    "\n",
    "for j in range(N):\n",
    "  d = Ds[j]\n",
    "  W = Wn[:(d+1),:] # the normalization is not correct but it does not matter for knn which relies on relative distances\n",
    "  trainingXrproj = [dot(W, x) for x in trainingX]\n",
    "  nearestNeighborsClassifierRproj = KNeighborsClassifier(n_neighbors=k) # L2 distance is the default\n",
    "  nearestNeighborsClassifierRproj.fit(trainingXrproj, trainingTarget)\n",
    "  testingXrproj = np.array([dot(W, x) for x in testingX])\n",
    "  predRproj = nearestNeighborsClassifierRproj.predict(testingXrproj[J])\n",
    "  ClassificationErrorRedRandom[j] = mean(predRproj != testingTarget[J])\n",
    "\n",
    "print('Last Classification Error Random_Red is:', ClassificationErrorRedRandom[N-1])\n",
    "\n",
    "# plot\n",
    "fig = plt.figure()\n",
    "ylim((0,0.8))\n",
    "grid(linestyle='--', linewidth=0.3)\n",
    "plt.plot(Ds, ClassificationErrorRedPCA)\n",
    "plt.plot(Ds, ClassificationErrorRedRandom)\n",
    "#plt.plot(pcafull.explained_variance_ratio_[:cut])\n",
    "plt.legend(['Error PCA', 'Error Random'])#, 'Explained Variance PCA'])\n",
    "fig.suptitle('Classification Error Dimension Reduction')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ZoIw9hO_hlW6"
   },
   "source": [
    "### You are given the same time budget required for the prediction of $n_{test}=5000$ examples with the $k$-nearest neighbors computed on $n=10000$ examples above. You must still use the same classifier, but you may use more examples in the training set and use dimensionality reduction. \n",
    "\n",
    "### Q3. Can you improve the prediction performance?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 89
    },
    "colab_type": "code",
    "id": "0X1CjPKYtg4R",
    "outputId": "6315e3d4-427f-41a7-b720-74fdba1f7cc1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PCA projection: Proportion of misclassified images in testing set: 0.0282\n",
      "elapsed: 5.06944 seconds\n",
      "Random projection: Proportion of misclassified images in testing set: 0.0462\n",
      "elapsed: 5.21839 seconds\n"
     ]
    }
   ],
   "source": [
    "## We will use the full training set n=6.5e4 in the reduced space of dimension roughly 1e4.\n",
    "n_train = 65000\n",
    "I = random.choice(range(n_tot), n_train, replace=False)\n",
    "trainingX = X[I, :] \n",
    "trainingTarget = y[I]\n",
    "\n",
    "# testing set: we may not use them, except at the end to assert the quality of the classifier\n",
    "mask = np.ones(n_tot, dtype=bool) #np.ones_like(a,dtype=bool)\n",
    "mask[I] = False\n",
    "testingX = X[mask, :]\n",
    "testingTarget = y[mask]\n",
    "\n",
    "# Number of clusters as before\n",
    "k = 7\n",
    "J = random.choice(range(len(testingTarget)), 5000, replace=False)\n",
    "\n",
    "## PCA:  To ensure the time constraint and due to matrix inversion cost we can consider PCA on r=120 eigenvectors\n",
    "# but in fact, 20 or 25 are enough\n",
    "r = 25\n",
    "\n",
    "pca = PCA(n_components=r)\n",
    "trainingXred = pca.fit_transform(trainingX)\n",
    "nearestNeighborsClassifierRed = KNeighborsClassifier(n_neighbors=k) # L2 distance is the default\n",
    "nearestNeighborsClassifierRed.fit(trainingXred, trainingTarget)\n",
    "testingXred = pca.transform(testingX)\n",
    "start_time = time.time()\n",
    "predRed = nearestNeighborsClassifierRed.predict(testingXred[J])\n",
    "print(\"PCA projection: Proportion of misclassified images in testing set: %g\" %(mean(predRed != testingTarget[J])))\n",
    "print(\"elapsed: %g seconds\"%(time.time()-start_time))\n",
    "\n",
    "## Random Proj: in dimension d = 70 here to fit the time constraint\n",
    "d = 70\n",
    "\n",
    "W = random.normal(0, 1/sqrt(d), dimX*dimY*d).reshape((d, dimX*dimY))\n",
    "trainingXrproj = [dot(W, x) for x in trainingX]\n",
    "nearestNeighborsClassifierRproj = KNeighborsClassifier(n_neighbors=k) # L2 distance is the default\n",
    "nearestNeighborsClassifierRproj.fit(trainingXrproj, trainingTarget)\n",
    "testingXrproj = np.array([dot(W, x) for x in testingX])\n",
    "start_time = time.time()\n",
    "predRproj = nearestNeighborsClassifierRproj.predict(testingXrproj[J])\n",
    "print(\"Random projection: Proportion of misclassified images in testing set: %g\" %(mean(predRproj != testingTarget[J])))\n",
    "print(\"elapsed: %g seconds\"%(time.time()-start_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "CfYHph4ThlW7"
   },
   "source": [
    "### Q4. Try to use some non-Gaussian law for the random projection. When do you observe similar results, when not?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 435
    },
    "colab_type": "code",
    "id": "uYn1d9MkhlW8",
    "outputId": "a46c3f53-7176-49e4-c814-e726e06a6d7a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Last Classification Error Random_Red is: 0.0698\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# It is almost the same with subgaussian laws (uniform on compacts)\n",
    "# Let us try with Rademacher variables: P(-1)=P(1)=1/2\n",
    "# btw: the normalization does not matter since we rely on the relative distances\n",
    "\n",
    "n_train = 10000\n",
    "I = random.choice(range(n_tot), n_train, replace=False)\n",
    "trainingX = X[I, :] \n",
    "trainingTarget = y[I]\n",
    "\n",
    "# testing set: we may not use them, except at the end to assert the quality of the classifier\n",
    "mask = np.ones(n_tot, dtype=bool) #np.ones_like(a,dtype=bool)\n",
    "mask[I] = False\n",
    "testingX = X[mask, :]\n",
    "testingTarget = y[mask]\n",
    "\n",
    "ClassificationErrorRedRandomRademacher=np.zeros(N)\n",
    "\n",
    "cut = max(Ds)\n",
    "Wn = random.choice([-1,1], dimX*dimY*cut).reshape((cut, dimX*dimY))\n",
    "\n",
    "for j in range(N):\n",
    "  d = Ds[j]\n",
    "  W = Wn[:(d+1),:]\n",
    "  trainingXrproj = [dot(W, x) for x in trainingX]\n",
    "  nearestNeighborsClassifierRproj = KNeighborsClassifier(n_neighbors=k) # L2 distance is the default\n",
    "  nearestNeighborsClassifierRproj.fit(trainingXrproj, trainingTarget)\n",
    "  testingXrproj = np.array([dot(W, x) for x in testingX])\n",
    "  predRproj = nearestNeighborsClassifierRproj.predict(testingXrproj[J])\n",
    "  ClassificationErrorRedRandomRademacher[j] = mean(predRproj != testingTarget[J])\n",
    "\n",
    "print('Last Classification Error Random_Red is:', ClassificationErrorRedRandom[N-1])\n",
    "\n",
    "# plot\n",
    "fig = plt.figure()\n",
    "ylim((0,0.8))\n",
    "grid(linestyle='--', linewidth=0.3)\n",
    "plt.plot(Ds, ClassificationErrorRedRandomRademacher)\n",
    "plt.plot(Ds, ClassificationErrorRedRandom)\n",
    "plt.legend(['Error Random with Rademacher'])\n",
    "fig.suptitle('Classification Error with Random Projections')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "correction_mnist_dimred.ipynb",
   "provenance": []
  },
  "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}
