{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Automatically created module for IPython interactive environment\n"
     ]
    }
   ],
   "source": [
    "print(__doc__)\n",
    "\n",
    "from time import time\n",
    "import logging\n",
    "import pylab as pl\n",
    "import numpy as np\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.datasets import fetch_lfw_people\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "#from sklearn.decomposition import RandomizedPCA\n",
    "from sklearn.decomposition import PCA as RandomizedPCA\n",
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total dataset size:\n",
      "n_samples: 1288\n",
      "n_features: 1850\n",
      "n_classes: 7\n",
      "Extracting the top 300 eigenfaces from 966 faces\n",
      "done in 0.486s\n",
      "Projecting the input data on the eigenfaces orthonormal basis\n",
      "done in 0.021s\n",
      "Fitting the classifier to the training set\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/lzl/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done in 50.085s\n",
      "Best estimator found by grid search:\n",
      "SVC(C=1000.0, cache_size=200, class_weight='balanced', coef0=0.0,\n",
      "    decision_function_shape='ovr', degree=3, gamma=0.0005, kernel='rbf',\n",
      "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "    tol=0.001, verbose=False)\n",
      "Predicting the people names on the testing set\n",
      "done in 0.149s\n",
      "                   precision    recall  f1-score   support\n",
      "\n",
      "     Ariel Sharon       0.53      0.69      0.60        13\n",
      "     Colin Powell       0.74      0.88      0.80        60\n",
      "  Donald Rumsfeld       0.76      0.70      0.73        27\n",
      "    George W Bush       0.89      0.86      0.87       146\n",
      "Gerhard Schroeder       0.85      0.68      0.76        25\n",
      "      Hugo Chavez       0.73      0.53      0.62        15\n",
      "       Tony Blair       0.78      0.78      0.78        36\n",
      "\n",
      "         accuracy                           0.80       322\n",
      "        macro avg       0.75      0.73      0.74       322\n",
      "     weighted avg       0.81      0.80      0.80       322\n",
      "\n",
      "[[  9   2   1   1   0   0   0]\n",
      " [  2  53   1   3   0   1   0]\n",
      " [  2   2  19   2   0   0   2]\n",
      " [  3   9   4 125   0   1   4]\n",
      " [  0   0   0   6  17   1   1]\n",
      " [  0   4   0   0   2   8   1]\n",
      " [  1   2   0   4   1   0  28]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 518.4x518.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 518.4x518.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Display progress logs on stdout\n",
    "logging.basicConfig(level=logging.INFO, format='%(asctime)s %(message)s')\n",
    "\n",
    "\n",
    "###############################################################################\n",
    "# Download the data, if not already on disk and load it as numpy arrays\n",
    "lfw_people = fetch_lfw_people(min_faces_per_person=70, resize=0.4)\n",
    "\n",
    "# introspect the images arrays to find the shapes (for plotting)\n",
    "n_samples, h, w = lfw_people.images.shape\n",
    "np.random.seed(42)\n",
    "\n",
    "# for machine learning we use the data directly (as relative pixel\n",
    "# position info is ignored by this model)\n",
    "X = lfw_people.data\n",
    "n_features = X.shape[1]\n",
    "\n",
    "# the label to predict is the id of the person\n",
    "y = lfw_people.target\n",
    "target_names = lfw_people.target_names\n",
    "n_classes = target_names.shape[0]\n",
    "\n",
    "print(\"Total dataset size:\")\n",
    "print(\"n_samples: %d\" % n_samples)\n",
    "print(\"n_features: %d\" % n_features)\n",
    "print(\"n_classes: %d\" % n_classes)\n",
    "\n",
    "\n",
    "###############################################################################\n",
    "# Split into a training and testing set\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)\n",
    "\n",
    "###############################################################################\n",
    "# Compute a PCA (eigenfaces) on the face dataset (treated as unlabeled\n",
    "# dataset): unsupervised feature extraction / dimensionality reduction\n",
    "n_components = 150\n",
    "\n",
    "print(\"Extracting the top %d eigenfaces from %d faces\" % (n_components, X_train.shape[0]))\n",
    "t0 = time()\n",
    "pca = RandomizedPCA(n_components=n_components, whiten=True).fit(X_train)\n",
    "print(\"done in %0.3fs\" % (time() - t0))\n",
    "\n",
    "eigenfaces = pca.components_.reshape((n_components, h, w))\n",
    "\n",
    "print(\"Projecting the input data on the eigenfaces orthonormal basis\")\n",
    "t0 = time()\n",
    "X_train_pca = pca.transform(X_train)\n",
    "X_test_pca = pca.transform(X_test)\n",
    "print(\"done in %0.3fs\" % (time() - t0))\n",
    "\n",
    "\n",
    "###############################################################################\n",
    "# Train a SVM classification model\n",
    "\n",
    "print(\"Fitting the classifier to the training set\")\n",
    "t0 = time()\n",
    "param_grid = {\n",
    "         'C': [1e3, 5e3, 1e4, 5e4, 1e5],\n",
    "          'gamma': [0.0001, 0.0005, 0.001, 0.005, 0.01, 0.1],\n",
    "          }\n",
    "# for sklearn version 0.16 or prior, the class_weight parameter value is 'auto'\n",
    "clf = GridSearchCV(SVC(kernel='rbf', class_weight='balanced'), param_grid)\n",
    "clf = clf.fit(X_train_pca, y_train)\n",
    "print(\"done in %0.3fs\" % (time() - t0))\n",
    "print(\"Best estimator found by grid search:\")\n",
    "print(clf.best_estimator_)\n",
    "\n",
    "\n",
    "###############################################################################\n",
    "# Quantitative evaluation of the model quality on the test set\n",
    "\n",
    "print(\"Predicting the people names on the testing set\")\n",
    "t0 = time()\n",
    "y_pred = clf.predict(X_test_pca)\n",
    "print(\"done in %0.3fs\" % (time() - t0))\n",
    "\n",
    "print(classification_report(y_test, y_pred, target_names=target_names))\n",
    "print(confusion_matrix(y_test, y_pred, labels=list(range(n_classes))))\n",
    "\n",
    "\n",
    "###############################################################################\n",
    "# Qualitative evaluation of the predictions using matplotlib\n",
    "\n",
    "def plot_gallery(images, titles, h, w, n_row=3, n_col=4):\n",
    "    \"\"\"Helper function to plot a gallery of portraits\"\"\"\n",
    "    pl.figure(figsize=(1.8 * n_col, 2.4 * n_row))\n",
    "    pl.subplots_adjust(bottom=0, left=.01, right=.99, top=.90, hspace=.35)\n",
    "    for i in range(n_row * n_col):\n",
    "        pl.subplot(n_row, n_col, i + 1)\n",
    "        pl.imshow(images[i].reshape((h, w)), cmap=pl.cm.gray)\n",
    "        pl.title(titles[i], size=12)\n",
    "        pl.xticks(())\n",
    "        pl.yticks(())\n",
    "\n",
    "\n",
    "# plot the result of the prediction on a portion of the test set\n",
    "\n",
    "def title(y_pred, y_test, target_names, i):\n",
    "    pred_name = target_names[y_pred[i]].rsplit(' ', 1)[-1]\n",
    "    true_name = target_names[y_test[i]].rsplit(' ', 1)[-1]\n",
    "    return 'predicted: %s\\ntrue:      %s' % (pred_name, true_name)\n",
    "\n",
    "prediction_titles = [title(y_pred, y_test, target_names, i)\n",
    "                         for i in range(y_pred.shape[0])]\n",
    "\n",
    "plot_gallery(X_test, prediction_titles, h, w)\n",
    "\n",
    "# plot the gallery of the most significative eigenfaces\n",
    "\n",
    "eigenface_titles = [\"eigenface %d\" % i for i in range(eigenfaces.shape[0])]\n",
    "plot_gallery(eigenfaces, eigenface_titles, h, w)\n",
    "\n",
    "pl.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**下面的代码输出了每个主成分的可释方差**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the raio is  [0.19346525 0.15116832 0.0708368  0.05951807 0.05157493 0.02887155\n",
      " 0.02514483 0.02176463 0.0201938  0.01902122 0.0168221  0.01580598\n",
      " 0.01223363 0.01087936 0.01064451 0.00979653 0.00892398 0.00854845\n",
      " 0.00835711 0.00722635 0.00696569 0.00653856 0.00639558 0.00561317\n",
      " 0.00531107 0.00520152 0.00507466 0.00484209 0.00443588 0.0041783\n",
      " 0.00393705 0.00381728 0.00356061 0.00351201 0.00334556 0.00329931\n",
      " 0.00314626 0.00296217 0.00290136 0.00284723 0.00280004 0.00267555\n",
      " 0.00259901 0.00258401 0.00240919 0.00238994 0.00235403 0.00222587\n",
      " 0.00217508 0.00216566 0.00209064 0.00205426 0.00200421 0.00197396\n",
      " 0.0019383  0.00188764 0.00180173 0.00178897 0.00174819 0.00173054\n",
      " 0.00165647 0.00162947 0.00157409 0.00153429 0.00149965 0.00147262\n",
      " 0.00143931 0.00141882 0.00139698 0.00138147 0.00134005 0.00133169\n",
      " 0.00128811 0.00125595 0.00124247 0.00121864 0.00120948 0.00118305\n",
      " 0.00115092 0.00113676 0.00112622 0.00111621 0.00109399 0.00107175\n",
      " 0.00105669 0.0010436  0.00102396 0.00101696 0.00099776 0.00096374\n",
      " 0.00094236 0.00091984 0.00091323 0.00089209 0.00087176 0.00086249\n",
      " 0.00084313 0.00083927 0.00082853 0.00080313 0.00078747 0.00078228\n",
      " 0.00075691 0.00075292 0.00074768 0.0007347  0.00073158 0.00071689\n",
      " 0.00070577 0.00069715 0.00066866 0.00066503 0.00065623 0.00063944\n",
      " 0.00063688 0.00062721 0.00061795 0.00061033 0.00060175 0.00059334\n",
      " 0.00058297 0.00057581 0.00056649 0.00056236 0.00055077 0.00054499\n",
      " 0.00053279 0.00052337 0.00051409 0.00051363 0.00051042 0.00049526\n",
      " 0.00048876 0.00047817 0.00047104 0.00046536 0.00046507 0.00045444\n",
      " 0.00044802 0.00044516 0.00043847 0.00043143 0.00042947 0.00042014\n",
      " 0.00041234 0.0004096  0.0004038  0.00040279 0.00039076 0.00038592]\n"
     ]
    }
   ],
   "source": [
    "print (\"the raio is \", pca.explained_variance_ratio_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-30T03:03:49.715652Z",
     "start_time": "2019-08-30T03:01:27.405043Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting the top 10 eigenfaces from 966 faces\n",
      "done in 0.052s\n",
      "Projecting the input data on the eigenfaces orthonormal basis\n",
      "done in 0.007s\n",
      "Fitting the classifier to the training set\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/lzl/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done in 49.981s\n",
      "Best estimator found by grid search:\n",
      "Predicting the people names on the testing set\n",
      "done in 0.014s\n",
      "Extracting the top 15 eigenfaces from 966 faces\n",
      "done in 0.048s\n",
      "Projecting the input data on the eigenfaces orthonormal basis\n",
      "done in 0.007s\n",
      "Fitting the classifier to the training set\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/lzl/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done in 16.526s\n",
      "Best estimator found by grid search:\n",
      "Predicting the people names on the testing set\n",
      "done in 0.013s\n",
      "Extracting the top 25 eigenfaces from 966 faces\n",
      "done in 0.059s\n",
      "Projecting the input data on the eigenfaces orthonormal basis\n",
      "done in 0.007s\n",
      "Fitting the classifier to the training set\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/lzl/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done in 8.275s\n",
      "Best estimator found by grid search:\n",
      "Predicting the people names on the testing set\n",
      "done in 0.027s\n",
      "Extracting the top 50 eigenfaces from 966 faces\n",
      "done in 0.223s\n",
      "Projecting the input data on the eigenfaces orthonormal basis\n",
      "done in 0.017s\n",
      "Fitting the classifier to the training set\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/lzl/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done in 7.747s\n",
      "Best estimator found by grid search:\n",
      "Predicting the people names on the testing set\n",
      "done in 0.026s\n",
      "Extracting the top 100 eigenfaces from 966 faces\n",
      "done in 0.094s\n",
      "Projecting the input data on the eigenfaces orthonormal basis\n",
      "done in 0.011s\n",
      "Fitting the classifier to the training set\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/lzl/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done in 15.724s\n",
      "Best estimator found by grid search:\n",
      "Predicting the people names on the testing set\n",
      "done in 0.050s\n",
      "Extracting the top 250 eigenfaces from 966 faces\n",
      "done in 0.201s\n",
      "Projecting the input data on the eigenfaces orthonormal basis\n",
      "done in 0.019s\n",
      "Fitting the classifier to the training set\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/lzl/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done in 43.009s\n",
      "Best estimator found by grid search:\n",
      "Predicting the people names on the testing set\n",
      "done in 0.135s\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score as f1\n",
    "f1_score = []\n",
    "n_com = [10, 15, 25, 50, 100, 250]\n",
    "for x in n_com:\n",
    "    # Compute a PCA (eigenfaces) on the face dataset (treated as unlabeled\n",
    "    # dataset): unsupervised feature extraction / dimensionality reduction\n",
    "    n_components = x\n",
    "\n",
    "    print(\"Extracting the top %d eigenfaces from %d faces\" % (n_components, X_train.shape[0]))\n",
    "    t0 = time()\n",
    "    pca = RandomizedPCA(n_components=n_components, whiten=True).fit(X_train)\n",
    "    print(\"done in %0.3fs\" % (time() - t0))\n",
    "\n",
    "    eigenfaces = pca.components_.reshape((n_components, h, w))\n",
    "\n",
    "    print(\"Projecting the input data on the eigenfaces orthonormal basis\")\n",
    "    t0 = time()\n",
    "    X_train_pca = pca.transform(X_train)\n",
    "    X_test_pca = pca.transform(X_test)\n",
    "    print(\"done in %0.3fs\" % (time() - t0))\n",
    "\n",
    "\n",
    "    ###############################################################################\n",
    "    # Train a SVM classification model\n",
    "\n",
    "    print(\"Fitting the classifier to the training set\")\n",
    "    t0 = time()\n",
    "    param_grid = {\n",
    "             'C': [1e3, 5e3, 1e4, 5e4, 1e5],\n",
    "              'gamma': [0.0001, 0.0005, 0.001, 0.005, 0.01, 0.1],\n",
    "              }\n",
    "    # for sklearn version 0.16 or prior, the class_weight parameter value is 'auto'\n",
    "    clf = GridSearchCV(SVC(kernel='rbf', class_weight='balanced'), param_grid)\n",
    "    clf = clf.fit(X_train_pca, y_train)\n",
    "    print(\"done in %0.3fs\" % (time() - t0))\n",
    "    print(\"Best estimator found by grid search:\")\n",
    "    #print(clf.best_estimator_)\n",
    "\n",
    "\n",
    "    ###############################################################################\n",
    "    # Quantitative evaluation of the model quality on the test set\n",
    "\n",
    "    print(\"Predicting the people names on the testing set\")\n",
    "    t0 = time()\n",
    "    y_pred = clf.predict(X_test_pca)\n",
    "    print(\"done in %0.3fs\" % (time() - t0))\n",
    "    f1_score.append(f1(y_test,y_pred,average=None))\n",
    "    \n",
    "    #print(classification_report(y_test, y_pred, target_names=target_names))\n",
    "    #print(confusion_matrix(y_test, y_pred, labels=list(range(n_classes))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-30T03:05:02.522015Z",
     "start_time": "2019-08-30T03:05:02.515006Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The n_com:10\n",
      "The f1 score:0.117647\n",
      "\n",
      "\n",
      "The n_com:15\n",
      "The f1 score:0.324324\n",
      "\n",
      "\n",
      "The n_com:25\n",
      "The f1 score:0.642857\n",
      "\n",
      "\n",
      "The n_com:50\n",
      "The f1 score:0.689655\n",
      "\n",
      "\n",
      "The n_com:100\n",
      "The f1 score:0.714286\n",
      "\n",
      "\n",
      "The n_com:250\n",
      "The f1 score:0.666667\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for x in range(0,len(n_com)):\n",
    "    print(\"The n_com:%d\\nThe f1 score:%f\\n\\n\" % (n_com[x],f1_score[x][0]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}