Q3_PCA_LDA.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Q3 PCA-LDA.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPsbM9eB/nbNkHL3zfGw/SY",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/lhmin0614/CS492_Coursework1/blob/main/Q3_PCA_LDA.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gct6cyeBIYR8",
        "outputId": "28b17deb-0db7-4356-e5f8-5787a4547b7b"
      },
      "source": [
        "%tensorflow_version 2.0x\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from google.colab import drive\n",
        "import numpy as np\n",
        "from scipy.io import loadmat\n",
        "import matplotlib.pyplot as plt\n",
        "import scipy.ndimage as ndimage\n",
        "import time\n",
        "from sklearn.metrics import confusion_matrix\n",
        "import seaborn as sn\n",
        "\n",
        "drive.mount('/content/drive')\n",
        "mat = loadmat('/content/drive/MyDrive/CS492_teamProject/face.mat')\n",
        "\n",
        "X = mat['X']\n",
        "y = mat['l'][0]"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "`%tensorflow_version` only switches the major version: 1.x or 2.x.\n",
            "You set: `2.0x`. This will be interpreted as: `2.x`.\n",
            "\n",
            "\n",
            "TensorFlow 2.x selected.\n",
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PR0DgBJDOvfP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4be0f037-30b8-45bf-cd7c-b8be940a3ee8"
      },
      "source": [
        "#8:2 train:test split\n",
        "train_X = [i[[True, True, True, True, False]*104] for i in X]         \n",
        "test_X  = [i[[False, False, False, False, True]*104] for i in X]      \n",
        "train_y = y[[True, True, True, True, False]*104]\n",
        "test_y  = y[[False, False, False, False, True]*104]\n",
        "\n",
        "print(\"<<Data dimension>>\")\n",
        "print(\"train_X : \", len(train_X),\" * \", len(train_X[0]))\n",
        "print(\"test_X  : \", len(test_X),\" * \",  len(test_X[0]))\n",
        "print(\"train_y : \", len(train_y))\n",
        "print(\"test_y  : \", len(test_y))"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<<Data dimension>>\n",
            "train_X :  2576  *  416\n",
            "test_X  :  2576  *  104\n",
            "train_y :  416\n",
            "test_y  :  104\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Po1ZtBi7M986"
      },
      "source": [
        "avg_X = [ sum(i)/len(i) for i in train_X ]   \n",
        "\n",
        "def PCA(Mpca):\n",
        "  sub_X = [ list(np.asarray(x) - x_bar) for x, x_bar in zip(train_X, avg_X) ]\n",
        "  A = np.matrix(sub_X)\n",
        "  S = np.dot(A, A.T)\n",
        "  eigvals, eigvecs_column = np.linalg.eigh(S) \n",
        "  eigvecs = eigvecs_column.T\n",
        "  best_eigvals = eigvals[-Mpca:]\n",
        "  best_eigvecs = eigvecs[-Mpca:]\n",
        "  W = np.dot(A.T, np.matrix(best_eigvecs).T).T \n",
        "  return W, best_eigvecs"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MY1xWXCYOx1c"
      },
      "source": [
        "def LDA(Mlda, best_eigvecs):\n",
        "  class_mean  = []\n",
        "  global_mean = avg_X\n",
        "  for i in range (52):\n",
        "    class_mean.append([ sum(j)/len(j) for j in np.asarray(np.matrix(train_X)[:,i*8:(i+1)*8]) ])   \n",
        "\n",
        "  between_diff = []\n",
        "  within_diff = []\n",
        "  for i in range(52):\n",
        "    between_diff.append([a - b for a, b in zip(class_mean[i], global_mean)])\n",
        "  for i in range(416):\n",
        "    within_diff.append(np.subtract(np.asarray(np.matrix(train_X).T)[i],class_mean[i//8]))\n",
        "\n",
        "  W =  np.matrix(best_eigvecs).T\n",
        "  Sb = np.dot( np.matrix(between_diff).T, np.matrix(between_diff) )     #52 is the number of samples in class ci\n",
        "  Sw = np.dot( np.matrix(within_diff).T , np.matrix(within_diff) )\n",
        "  reduced_Sb = np.dot(np.dot(W.T, Sb), W)\n",
        "  reduced_Sw = np.dot(np.dot(W.T, Sw), W)\n",
        "  matrix_LDA = np.dot( np.linalg.inv(reduced_Sw), reduced_Sb )\n",
        "\n",
        "  fisher_eigvals, fisher_eigvecs = np.linalg.eigh(matrix_LDA) \n",
        "  fisherface = fisher_eigvecs.T\n",
        "  fisherface  = np.asarray(fisherface)[-Mlda:] \n",
        "  return fisherface"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HLW7Y5TXJl4b"
      },
      "source": [
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "\n",
        "def projection(X,  P) :\n",
        "  W = np.dot(np.matrix(X).T, P)\n",
        "  return W\n",
        "\n",
        "\n",
        "def NN_classification(Mpca, Mlda):\n",
        "  #PCA\n",
        "  W, best_eigvecs = PCA(Mpca)\n",
        "  A = np.matrix([ list(np.asarray(x) - x_bar) for x, x_bar in zip(train_X, avg_X) ])\n",
        "  prj_X  = projection(A, np.matrix(best_eigvecs).T)\n",
        "  B = np.matrix([ list(np.asarray(x) - x_bar) for x, x_bar in zip(test_X, avg_X) ])\n",
        "  validation_X = projection(B, np.matrix(best_eigvecs).T)\n",
        "\n",
        "  #LDA\n",
        "  fisherface = LDA(Mlda, best_eigvecs)\n",
        "  prj_X_LDA = projection(prj_X.T, np.matrix(fisherface).T)\n",
        "  validation_X_LDA = projection(validation_X.T, np.matrix(fisherface).T)\n",
        "\n",
        "  #kNN\n",
        "  classifier_LDA = KNeighborsClassifier(n_neighbors = 1)\n",
        "  classifier_LDA.fit(prj_X_LDA.tolist(), train_y)\n",
        "  train_accuracy = (classifier_LDA.score(prj_X_LDA.tolist(), train_y))\n",
        "  test_accuracy = (classifier_LDA.score(validation_X_LDA.tolist(), test_y))\n",
        "\n",
        "  return (train_accuracy, test_accuracy)"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RRul-LLOYyZK",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        },
        "outputId": "c323dd47-7e56-4fb6-e8be-e9bc8b72a8f6"
      },
      "source": [
        "pca_cand = [1, 2, 5, 10, 20, 50, 100, 150, 200]\n",
        "lda_cand = [1, 2, 5, 10, 20, 50, 100, 150, 200]\n",
        "\n",
        "# change Mpca value \n",
        "Mpca_change_train = []\n",
        "Mpca_change_test = []\n",
        "for i in pca_cand:\n",
        "  train, test = NN_classification(i, 50)\n",
        "  Mpca_change_train.append(train)\n",
        "  Mpca_change_test.append(test)\n",
        "\n",
        "plt.xlabel(\"Mpca (Mlda=50)\")\n",
        "plt.ylabel(\"kNN accuracy (k=1)\")\n",
        "plt.plot(lda_cand, Mpca_change_test)\n",
        "plt.show()\n",
        "\n",
        "# change Mlda value\n",
        "Mlda_change_train = []\n",
        "Mlda_change_test = []\n",
        "for j in lda_cand:\n",
        "  train, test = NN_classification(100, j)\n",
        "  Mlda_change_test.append(test)\n",
        "\n",
        "plt.xlabel(\"Mlda (Mpca=100)\")\n",
        "plt.ylabel(\"kNN accuracy (k=1)\")\n",
        "plt.plot(lda_cand, Mlda_change_test)\n",
        "plt.show()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "tJa2XUhADL0A",
        "outputId": "9c65452a-a865-4812-b717-65da1025ed19"
      },
      "source": [
        "#comparison of PCA and LDA\n",
        "Mpca = 100\n",
        "Mlda = 50\n",
        "\n",
        "#PCA\n",
        "start = time.time()\n",
        "W, best_eigvecs = PCA(Mpca)\n",
        "A = np.matrix([ list(np.asarray(x) - x_bar) for x, x_bar in zip(train_X, avg_X) ])\n",
        "prj_X  = projection(A, np.matrix(best_eigvecs).T)\n",
        "B = np.matrix([ list(np.asarray(x) - x_bar) for x, x_bar in zip(test_X, avg_X) ])\n",
        "validation_X = projection(B, np.matrix(best_eigvecs).T)\n",
        "stop = time.time()\n",
        "print(f\"PCA time: {stop - start}s\")\n",
        "\n",
        "#kNN for PCA\n",
        "classifier_PCA = KNeighborsClassifier(1)\n",
        "classifier_PCA.fit(prj_X.tolist(), train_y)\n",
        "preds_pca=classifier_PCA.predict(validation_X)\n",
        "print(\"<PCA confusion matrix>\")\n",
        "plt.figure(figsize = (15,10))\n",
        "plt.xlabel(\"predicted class\")\n",
        "plt.ylabel(\"actual class\")\n",
        "sn.heatmap(confusion_matrix(test_y, preds_pca), annot=True)\n",
        "\n",
        "#LDA\n",
        "start = time.time()\n",
        "fisherface = LDA(Mlda, best_eigvecs)\n",
        "train_accuracy = []\n",
        "test_accuracy = []\n",
        "k=[]\n",
        "prj_X_LDA = projection(prj_X.T, np.matrix(fisherface).T)\n",
        "validation_X_LDA = projection(validation_X.T, np.matrix(fisherface).T)\n",
        "stop = time.time()\n",
        "print(f\"LDA additional time: {stop - start}s\")\n",
        "\n",
        "#kNN for PCA-LDA\n",
        "classifier_LDA = KNeighborsClassifier(1)\n",
        "classifier_LDA.fit(prj_X_LDA.tolist(), train_y)\n",
        "preds_pcalda=classifier_LDA.predict(validation_X_LDA)\n",
        "print(\"<PCA-LDA confusion matrix>\")\n",
        "plt.figure(figsize = (15,10))\n",
        "plt.xlabel(\"predicted class\")\n",
        "plt.ylabel(\"actual class\")\n",
        "sn.heatmap(confusion_matrix(test_y, preds_pcalda), annot=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "PCA time: 4.61076021194458s\n",
            "LDA additional time: 1.8418149948120117s\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f983f95cf10>"
            ]
          },
          "metadata": {},
          "execution_count": 95
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}