DataMining_V2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [],
      "mount_file_id": "1-dD5QCU8X6IGaVEm2rEqPlZKFxIZ7MPS",
      "authorship_tag": "ABX9TyP8ofmeLgmzh875DMIm3PXE",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "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/mralamdari/Machine_Learning_Projects/blob/main/DataMining_V2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "I6T0TsXvhyBy"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import datetime, os, sys\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib\n",
        "from sklearn import preprocessing, linear_model, metrics, model_selection, decomposition, svm, datasets"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DweoaZSlLz0B"
      },
      "source": [
        "#### store data values in x,y test and train\n",
        "mnist = datasets.fetch_openml('mnist_784')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gp9xjDhPaasE"
      },
      "source": [
        "x_pc = EFA_PCA(mnist.data, 650)\n",
        "x_train_pca, x_test_pca, y_train, y_test = model_selection.train_test_split(x_pc, mnist.target, test_size=0.2)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "46TjFxLybX1L",
        "outputId": "ad430b17-2d50-4750-ae75-b1e6a496feb6"
      },
      "source": [
        "EFA_tester(x_train_pca, y_train, x_test_pca, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train Delay: 35, Test Delay: 0, Score: 0.917\n",
            "Train Delay: 746, Test Delay: 241, Score: 0.9645\n",
            "Train Delay: 410, Test Delay: 0, Score: 0.8994285714285715\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RWxyA3MxJZDD"
      },
      "source": [
        "def EFA_PCA(X, components=320):\n",
        "    x_std = preprocessing.StandardScaler().fit_transform(X)\n",
        "    if components != 0:\n",
        "      return decomposition.PCA(n_components=components).fit_transform(x_std)\n",
        "    return x_std\n",
        "\n",
        "\n",
        "def EFA_tester(X, Y, XP, YT):    \n",
        "    # first method : Logistic Regression\n",
        "    start_11 = datetime.datetime.now()\n",
        "    logisticRegression = linear_model.LogisticRegression(C=1.25, n_jobs=-1)\n",
        "    logisticRegression.fit(X, Y)\n",
        "    start_21 = datetime.datetime.now()\n",
        "    delay11 = start_21 - start_11\n",
        "    clf1 = logisticRegression.predict(XP)\n",
        "    start_31 = datetime.datetime.now()\n",
        "    delay12 = start_31 - start_21\n",
        "    score1 = np.sum(YT == clf1) / YT.shape[0]\n",
        "    print(f'Train Delay: {delay11.seconds}, Test Delay: {delay12.seconds}, Score: {score1}')\n",
        "\n",
        "    # Second method : SVM\n",
        "    start_21 = datetime.datetime.now()\n",
        "    svm_clf = svm.SVC()\n",
        "    svm_clf.fit(X, Y)\n",
        "    start_22 = datetime.datetime.now()\n",
        "    delay21 = start_22 - start_21\n",
        "    clf2 = svm_clf.predict(XP)\n",
        "    start_23 = datetime.datetime.now()\n",
        "    delay22 = start_23 - start_22\n",
        "    score2 = np.sum(YT == clf2) / YT.shape[0]\n",
        "    print(f'Train Delay: {delay21.seconds}, Test Delay: {delay22.seconds}, Score: {score2}')\n",
        "\n",
        "\n",
        "    #  third metohd : Stochastic Gradient Descent\n",
        "    start_31 = datetime.datetime.now()\n",
        "    sgd = linear_model.SGDClassifier()\n",
        "    sgd.fit(X, Y)\n",
        "    start_32 = datetime.datetime.now()\n",
        "    delay31 = start_32 - start_31\n",
        "    clf3 = sgd.predict(XP)\n",
        "    start_33 = datetime.datetime.now()\n",
        "    delay32 = start_33 - start_32\n",
        "    score3 = np.sum(YT == clf3) / YT.shape[0]\n",
        "    print(f'Train Delay: {delay31.seconds}, Test Delay: {delay32.seconds}, Score: {score3}')\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bw3Ytf6jj6xk"
      },
      "source": [
        "## first way: normal way\n",
        "def standardization(x):\n",
        "    return (x - np.mean(x, axis=0)) / np.std(x, axis=0)\n",
        "\n",
        "x_s = standardization(mnist.data)\n",
        "x_std = np.nan_to_num(x_s)\n",
        "\n",
        "## second way: using sklearn\n",
        "# x_std = preprocessing.StandardScaler().fit_transform(mnist.data)\n",
        "x_train_std, x_test_std, y_train, y_test = model_selection.train_test_split(x_std, mnist.target, test_size=0.2)\n",
        "# x_train_std, x_test_std, y_train, y_test = model_selection.train_test_split(mnist.data, mnist.target, test_size=0.2)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Krru0dD0ok9X"
      },
      "source": [
        "def pcaMaker(k, x_std):\n",
        "  x_meaned = x_std - np.mean(x_std, axis=0)\n",
        "  cov_mat = x_meaned.T.dot(x_meaned) / (x_std.shape[0] - 1)\n",
        "  # cov_mat = np.cov(x_meaned, rowvar = False)  \n",
        "  eigen_values, eigen_vectors = np.linalg.eig(cov_mat.T)\n",
        "  sorted_index = np.argsort(eigen_values)[::-1]\n",
        "  sorted_eigenvalue = eigen_values[sorted_index]\n",
        "  sorted_eigenvectors = eigen_vectors[:,sorted_index]\n",
        "  eigenvector_subset = sorted_eigenvectors[:,0:k]\n",
        "  X_reduced = eigenvector_subset.T.dot( x_meaned.T).T   \n",
        "  return X_reduced"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jboGyMsUpv--"
      },
      "source": [
        "## first way: without library\n",
        "x_train_pca = pcaMaker(320, x_train_std)\n",
        "x_test_pca = np.float64(pcaMaker(320, x_test_std))\n",
        "\n",
        "\n",
        "## second way: with sklearn.decomposiotion.PCA()\n",
        "# {'auto', 'full', 'arpack', 'randomized'}\n",
        "pca = decomposition.PCA(n_components=320)\n",
        "pca.fit(x_train_std)\n",
        "# x_train_pca = pca.transform(x_train_std)\n",
        "# x_test_pca = pca.transform(x_test_std)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tL6NkAAIx0mf",
        "outputId": "de85c313-6f70-4ac1-9fd5-b9193f5d4de3"
      },
      "source": [],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<module 'scipy.misc' from '/usr/local/lib/python3.7/dist-packages/scipy/misc/__init__.py'>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 216
        },
        "id": "B8JGI6qzukyu",
        "outputId": "61259672-0464-41d8-b106-4829e341b93f"
      },
      "source": [
        "num, n= 2000, 10\n",
        "x_std = preprocessing.StandardScaler().fit_transform(mnist.data)\n",
        "pca = decomposition.PCA(n_components=n)\n",
        "x = pca.fit_transform(x_std)\n",
        "sample2 = pca.inverse_transform(x[num])\n",
        "sample2.shape = (28, 28)\n",
        "b = plt.subplot(1, 2, 1)\n",
        "b.set_title(f'Reduced after PCA_{n}')\n",
        "plt.imshow(sample2, cmap=plt.cm.gray_r)\n",
        "matplotlib.image.imsave(f'PCA_{n}.jpeg', sample2, cmap=plt.cm.gray_r)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        },
        "id": "5LOb3dNvp-_R",
        "outputId": "f20f230a-101b-490f-f8b1-cb747cc389b3"
      },
      "source": [
        "#### This code shows difference between data's images, before and after pca\n",
        "num = 2000\n",
        "\n",
        "sample1 = x_train_std[num]\n",
        "sample1.shape = (28, 28)\n",
        "a = plt.subplot(1, 2, 1)\n",
        "a.set_title('Original Image')\n",
        "plt.imshow(sample1, cmap=plt.cm.gray_r)\n",
        "# This is the result of what I was trying to get in above codes\n",
        "\n",
        "sample2 = pca.inverse_transform(x_train_pca[num])\n",
        "# sample2 = x_train_pca[num]\n",
        "sample2.shape=(28, 28)\n",
        "b = plt.subplot(1, 2, 1)\n",
        "b.set_title('Reduced after PCA')\n",
        "plt.imshow(sample2, cmap=plt.cm.gray_r)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:14: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
            "  \n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f9c6a65a3d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CgQQ00DZ39AE",
        "outputId": "469a99c0-776c-4b64-f460-030dbd8a5880"
      },
      "source": [
        "# this part is only made to estimate the difference of training delay on OriginalData vs PCAData\n",
        "logistic = linear_model.LogisticRegression(C=1.1, n_jobs=-1)\n",
        "\n",
        "start = datetime.datetime.now()\n",
        "logistic.fit(x_train_pca, y_train)\n",
        "delay1=datetime.datetime.now() - start\n",
        "\n",
        "start2 = datetime.datetime.now()\n",
        "logistic.fit(x_train_std, y_train)\n",
        "delay2=datetime.datetime.now() - start2\n",
        "\n",
        "print('Original Data Classification Delay: ', delay2.seconds)\n",
        "print('PCA Data Classification Delay: ', delay1.seconds)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Original Data Classification Delay:  38\n",
            "PCA Data Classification Delay:  18\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EuFDea6F4CpM"
      },
      "source": [
        "# first method : Logistic Regression\n",
        "logisticRegression = linear_model.LogisticRegression(C=1.25, n_jobs=-1)\n",
        "logisticRegression.fit(x_train_pca, y_train)\n",
        "clf1 = logisticRegression.predict(x_test_pca)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eoZRXo73lw-i"
      },
      "source": [
        "# Second method : SVM\n",
        "svm_clf = svm.SVC()\n",
        "svm_clf.fit(x_train_pca, y_train)\n",
        "clf2 = svm_clf.predict(x_test_pca)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EibGNW3AoBB7"
      },
      "source": [
        "#  third metohd : Stochastic Gradient Descent\n",
        "sgd = linear_model.SGDClassifier()\n",
        "sgd.fit(x_train_pca, y_train)\n",
        "clf3 = sgd.predict(x_test_pca)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T-C6vwi7o_ae",
        "outputId": "8d090827-b010-4233-9621-695c6f3eef2b"
      },
      "source": [
        "# In here we estimate how much our algorithm is correct\n",
        "score = np.sum(y_test == clf3) / y_test.shape[0]\n",
        "score"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.23364285714285715"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 188
        },
        "id": "Inp7XOJd7Kja",
        "outputId": "d2b47ca3-62d9-43e0-d9c7-7e15e7da7784"
      },
      "source": [
        "# This method gives details about the classification\n",
        "metrics.classification_report(y_test, clf3)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'              precision    recall  f1-score   support\\n\\n           0       0.00      0.00      0.00      1366\\n           1       0.00      0.00      0.00      1572\\n           2       0.30      0.41      0.35      1377\\n           3       0.36      0.32      0.34      1471\\n           4       0.31      0.10      0.15      1390\\n           5       0.30      0.25      0.27      1224\\n           6       0.85      0.65      0.74      1419\\n           7       0.08      0.05      0.06      1444\\n           8       0.14      0.52      0.21      1358\\n           9       0.09      0.06      0.07      1379\\n\\n    accuracy                           0.23     14000\\n   macro avg       0.24      0.24      0.22     14000\\nweighted avg       0.24      0.23      0.22     14000\\n'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ici4DrYRVXg8"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    }
  ]
}