svm_Tests.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "\n",
    "import utils.ImagesProcessor as ip\n",
    "from sklearn import svm\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from matplotlib import cm\n",
    "from matplotlib.ticker import LinearLocator, FormatStrFormatter\n",
    "import itertools\n",
    "import time\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def loadImages(directory_name):\n",
    "    directory = os.fsencode(directory_name)\n",
    "    imgs = []\n",
    "    filenames = []\n",
    "    for file in os.listdir(directory):\n",
    "        filename = os.fsdecode(file)\n",
    "        if filename.lower().endswith(\".jpg\"): \n",
    "            img = IP.readImage(directory_name + \"/\" + filename)\n",
    "            img = IP.resizeImage(img, (1000, 1000))\n",
    "            #chromaticity = IP.extractChromaticity(img)\n",
    "            #colors_hist = np.array(IP.extractRGBHistogram(chromaticity))\n",
    "            texture_features = np.array(IP.extractTexturefeatures(img))\n",
    "            #img = img.reshape(-1)\n",
    "            imgs.append(texture_features)\n",
    "            filenames.append(filename)\n",
    "    return np.array(imgs), filenames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/skimage/feature/texture.py:109: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  if np.issubdtype(image.dtype, np.float):\n"
     ]
    }
   ],
   "source": [
    "TRAINING_PATH = \"datasets/jardin_landscape/validation/dataset_train\"\n",
    "NORMAL_PATH = \"datasets/jardin_landscape/validation/dataset_normal\"\n",
    "ANOMALY_PATH = \"datasets/jardin_landscape/validation/dataset_anomaly\"\n",
    "IP = ip.ImagesProcessor()\n",
    "\n",
    "# Load the images\n",
    "X_train, trainFilenames = loadImages(TRAINING_PATH)\n",
    "X_normal, normalFilenames= loadImages(NORMAL_PATH)\n",
    "X_anomaly, anomalyFilenames = loadImages(ANOMALY_PATH)\n",
    "\n",
    "# Normalize them\n",
    "X = np.append(np.append(X_train, X_normal, axis=0), X_anomaly, axis=0)\n",
    "X_train = X_train.astype('float32')/X.max(axis=0)\n",
    "X_normal = X_normal.astype('float32')/X.max(axis=0)\n",
    "X_anomaly = X_anomaly.astype('float32')/X.max(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7, 6)\n",
      "(12, 6)\n",
      "(18, 6)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "#print(X_train)\n",
    "print(X_normal.shape)\n",
    "#print(X_normal)\n",
    "print(X_anomaly.shape)\n",
    "#print(X_anomaly)\n",
    "#print(X)\n",
    "#print(np.random.choice(X_train, int(len(X_train)*1)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Explore the Nu and Gamma parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(0.5666666666666667, 1.1920928955078125)\n"
     ]
    }
   ],
   "source": [
    "def testSVM(nu=0.8, gamma=0.2, kernel=\"rbf\", coef0=0, degree=3, sizeDataset=7):\n",
    "    time1 = time.time()\n",
    "    clf = svm.OneClassSVM(nu=nu, kernel=kernel, gamma=gamma, degree=degree, coef0=coef0)\n",
    "    np.random.shuffle(X_train)\n",
    "    clf.fit(X_train[0:sizeDataset])\n",
    "    pred_ref = clf.predict(X_train)\n",
    "    #print(\"ref:\", pred_ref)\n",
    "    #print(trainFilenames)\n",
    "    pred_nor = clf.predict(X_normal)\n",
    "    #print(\"normal:\", pred_nor)\n",
    "    #print(normalFilenames)\n",
    "    pred_ano = clf.predict(X_anomaly)\n",
    "    #print(\"anomalies:\", pred_ano)\n",
    "    #print(anomalyFilenames)\n",
    "    time2 = time.time()\n",
    "    return accuracy_score(np.append(pred_nor, pred_ano), [1]*len(pred_nor) + [-1]*len(pred_ano)), (time2-time1)*1000\n",
    "print(testSVM(0.74, 0.2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\nnu_params = np.arange(0.01,1.0,0.1)\\ngamma_params = np.arange(0.01,14.0,0.1)\\n#coef0_params = np.arange(0,50,1)\\n#degree_params = np.arange(1,7,1)\\n\\nfig = plt.figure(num=None, figsize=(15, 10), dpi=80, facecolor='w', edgecolor='k')\\nax = fig.gca(projection='3d')\\n# Make data.\\nnu_params, gamma_params = np.meshgrid(nu_params, gamma_params)\\n#coef0_params, degree_params = np.meshgrid(coef0_params, degree_params)\\n\\naccurencies = np.array([testSVM(nu=x,gamma=y, coef0=20, degree=4) for x,y in zip(np.ravel(nu_params), np.ravel(gamma_params))])\\naccurencies = accurencies.reshape(gamma_params.shape)\\n\\n#accurencies = np.array([testSVM(coef0=x,degree=y, nu=0.5, gamma=0.166) for x,y in zip(np.ravel(coef0_params), np.ravel(degree_params))])\\n#accurencies = accurencies.reshape(degree_params.shape)\\n\\ngraph = ax.plot_surface(nu_params, gamma_params, accurencies, cmap=cm.coolwarm,\\n                       linewidth=0, antialiased=False)\\n\\n#ax.set_xlabel('Coef0 Parameter')\\n#ax.set_ylabel('Degree Parameter')\\n\\nax.set_xlabel('Nu Parameter')\\nax.set_ylabel('Gamma Parameter')\\n\\nax.set_zlabel('Accuracy')\\n\\n# Customize the z axis.\\nax.set_zlim(0, 1.00)\\nax.zaxis.set_major_locator(LinearLocator(10))\\nax.zaxis.set_major_formatter(FormatStrFormatter('%.02f'))\\n\\n# Add a color bar which maps values to colors.\\nfig.colorbar(graph, shrink=0.5, aspect=5)\\n\\nplt.show()\\n\\n\""
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "nu_params = np.arange(0.01,1.0,0.1)\n",
    "gamma_params = np.arange(0.01,14.0,0.1)\n",
    "#coef0_params = np.arange(0,50,1)\n",
    "#degree_params = np.arange(1,7,1)\n",
    "\n",
    "fig = plt.figure(num=None, figsize=(15, 10), dpi=80, facecolor='w', edgecolor='k')\n",
    "ax = fig.gca(projection='3d')\n",
    "# Make data.\n",
    "nu_params, gamma_params = np.meshgrid(nu_params, gamma_params)\n",
    "#coef0_params, degree_params = np.meshgrid(coef0_params, degree_params)\n",
    "\n",
    "accurencies = np.array([testSVM(nu=x,gamma=y, coef0=20, degree=4) for x,y in zip(np.ravel(nu_params), np.ravel(gamma_params))])\n",
    "accurencies = accurencies.reshape(gamma_params.shape)\n",
    "\n",
    "#accurencies = np.array([testSVM(coef0=x,degree=y, nu=0.5, gamma=0.166) for x,y in zip(np.ravel(coef0_params), np.ravel(degree_params))])\n",
    "#accurencies = accurencies.reshape(degree_params.shape)\n",
    "\n",
    "graph = ax.plot_surface(nu_params, gamma_params, accurencies, cmap=cm.coolwarm,\n",
    "                       linewidth=0, antialiased=False)\n",
    "\n",
    "#ax.set_xlabel('Coef0 Parameter')\n",
    "#ax.set_ylabel('Degree Parameter')\n",
    "\n",
    "ax.set_xlabel('Nu Parameter')\n",
    "ax.set_ylabel('Gamma Parameter')\n",
    "\n",
    "ax.set_zlabel('Accuracy')\n",
    "\n",
    "# Customize the z axis.\n",
    "ax.set_zlim(0, 1.00)\n",
    "ax.zaxis.set_major_locator(LinearLocator(10))\n",
    "ax.zaxis.set_major_formatter(FormatStrFormatter('%.02f'))\n",
    "\n",
    "# Add a color bar which maps values to colors.\n",
    "fig.colorbar(graph, shrink=0.5, aspect=5)\n",
    "\n",
    "plt.show()\n",
    "\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\ncoef0_params = np.arange(0,50,1)\\nfig, axes = plt.subplots()\\naccurencies = np.array([testSVM(nu=0.75, gamma=4, coef0=x) for x in coef0_params])\\ngraph = axes.plot(coef0_params, accurencies, 'r')\\naxes.set_xlabel('Coef0 Parameter')\\naxes.set_ylabel('Accuracy')\\nplt.show()\\n\""
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "coef0_params = np.arange(0,50,1)\n",
    "fig, axes = plt.subplots()\n",
    "accurencies = np.array([testSVM(nu=0.75, gamma=4, coef0=x) for x in coef0_params])\n",
    "graph = axes.plot(coef0_params, accurencies, 'r')\n",
    "axes.set_xlabel('Coef0 Parameter')\n",
    "axes.set_ylabel('Accuracy')\n",
    "plt.show()\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\ndatasetSize = np.arange(1,8,1)\\nfig, axe1 = plt.subplots()\\naxes = [testSVM(nu=0.75, gamma=0.2, sizeDataset=x) for x in datasetSize]\\naccurencies = [i[0] for i in axes]\\ntimes = [i[1] for i in axes]\\nprint(accurencies)\\ngraph = axe1.plot(datasetSize, accurencies, 'r')\\naxe1.set_xlabel('number of train samples')\\naxe1.set_ylabel('Accuracy', color='r')\\n\\naxe2 = axe1.twinx()\\naxe2.plot(datasetSize, times, 'b')\\naxe2.set_ylabel('Time(s)', color='b')\\nplt.show()\\n\""
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "datasetSize = np.arange(1,8,1)\n",
    "fig, axe1 = plt.subplots()\n",
    "axes = [testSVM(nu=0.75, gamma=0.2, sizeDataset=x) for x in datasetSize]\n",
    "accurencies = [i[0] for i in axes]\n",
    "times = [i[1] for i in axes]\n",
    "print(accurencies)\n",
    "graph = axe1.plot(datasetSize, accurencies, 'r')\n",
    "axe1.set_xlabel('number of train samples')\n",
    "axe1.set_ylabel('Accuracy', color='r')\n",
    "\n",
    "axe2 = axe1.twinx()\n",
    "axe2.plot(datasetSize, times, 'b')\n",
    "axe2.set_ylabel('Time(s)', color='b')\n",
    "plt.show()\n",
    "'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Final Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = svm.OneClassSVM(nu=0.75, kernel=\"rbf\", gamma=0.2)\n",
    "clf.fit(X_train)\n",
    "pred_ref = clf.predict(X_train)\n",
    "pred_nor = clf.predict(X_normal)\n",
    "pred_ano = clf.predict(X_anomaly)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c240ca6d8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "conf_mat = confusion_matrix(np.append(pred_nor, pred_ano), [1]*len(pred_nor) + [-1]*len(pred_ano))\n",
    "\n",
    "def plot_confusion_matrix(cm, title='Confusion matrix', cmap=plt.cm.Blues):     \n",
    "    labels_names_ref = [\"Anomalie\", \"Normal\"]\n",
    "        \n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(labels_names_ref))\n",
    "    plt.xticks(tick_marks, labels_names_ref, rotation=45)\n",
    "    plt.yticks(tick_marks, labels_names_ref)\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('Predicted label')\n",
    "    plt.xlabel('True label')\n",
    "\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j]),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "        \n",
    "plot_confusion_matrix(conf_mat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.5666666666666667\n",
      "Precision:  0.6111111111111112\n",
      "Recall:  0.626984126984127\n",
      "F1-Score:  0.618945868945869\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy:\", accuracy_score(np.append(pred_nor, pred_ano), [1]*len(pred_nor) + [-1]*len(pred_ano)))\n",
    "precision = (conf_mat[0][0]/(conf_mat[0][0] + conf_mat[1][0]) + conf_mat[1][1]/(conf_mat[1][1] + conf_mat[0][1]))/2\n",
    "print(\"Precision: \", precision)\n",
    "recall = (conf_mat[0][0]/(conf_mat[0][0] + conf_mat[0][1]) + conf_mat[1][1]/(conf_mat[1][1] + conf_mat[1][0]))/2\n",
    "print(\"Recall: \", recall)\n",
    "f1score = 2*(precision*recall)/(precision+recall)\n",
    "print(\"F1-Score: \", f1score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NA_IMG_0360.jpg  -  1\n",
      "NA_IMG_0362.jpg  -  1\n",
      "NA_IMG_0376.jpg  -  1\n",
      "NA_0244.jpg  -  -1\n",
      "NA_IMG_0411.jpg  -  1\n",
      "NA_0240.jpg  -  1\n",
      "NA_IMG_0366.jpg  -  1\n",
      "NA_IMG_0428.jpg  -  -1\n",
      "NA_IMG_0370.jpg  -  1\n",
      "NA_IMG_0364.jpg  -  1\n",
      "NA_IMG_0385.jpg  -  1\n",
      "NA_IMG_0278.jpg  -  1\n",
      "A_IMG_0321.jpg  -  -1\n",
      "A_IMG_0322.jpg  -  -1\n",
      "HA_IMG_0425.jpg  -  1\n",
      "HA_IMG_0424.jpg  -  1\n",
      "HA_IMG_0418.jpg  -  1\n",
      "A_IMG_0423.jpg  -  1\n",
      "A_IMG_0345.jpg  -  -1\n",
      "A_IMG_0390.jpg  -  1\n",
      "A_IMG_0409.jpg  -  1\n",
      "A_IMG_0353.jpg  -  -1\n",
      "A_IMG_0395.jpg  -  1\n",
      "HA_IMG_0420.jpg  -  1\n",
      "A_IMG_0417.jpg  -  1\n",
      "A_IMG_0402.jpg  -  1\n",
      "A_IMG_0413.jpg  -  1\n",
      "A_IMG_0272.jpg  -  -1\n",
      "A_IMG_0271.jpg  -  -1\n",
      "A_IMG_0329.jpg  -  -1\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(normalFilenames)):\n",
    "    print(normalFilenames[i], \" - \", pred_nor[i])\n",
    "for i in range(len(anomalyFilenames)):\n",
    "    print(anomalyFilenames[i], \" - \", pred_ano[i])"
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}