Covisibility_evaluation.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Co-visibilidad Evaluation"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This code implements the evaluation of the proposed co-visibility detection system.\n",
    "The co-visibility estimate is compared with the ground-truth value."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Libreries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import cv2 as cv\n",
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import auc"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Estimated co-visibility matrix\n",
    "These functions calculate the co-visibility matrix with the proposed method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def occupancy(pts1,pts2,cell_x,cell_y,size):\n",
    "# This function calculates the percentage of co-visibility between two images based on the position of the inliers.\n",
    "# An occupancy grid is defined for each image, where cells containing one or more inliers are marked as occupied.\n",
    "# The occupancy percentage is calculated as the number of occupied cells divided by the total number of cells.\n",
    "# The highest occupancy value among the images is chosen to establish the co-visibility percentage in order to obtain a symmetric adjacency matrix.\n",
    "#   -Input:\n",
    "#       pts1, pts2      ---> Lists of inliers, positions of features on the image plane.\n",
    "#       cell_x, cell_y  ---> Occupancy cell size.\n",
    "#       size            ---> Size of the original images (x,y).\n",
    "#   -Return: \n",
    "#       w               ---> Occupancy percentage, co-visibility value of the edge connecting the nodes.\n",
    "\n",
    "    # Defining occupancy cells for each image:\n",
    "    occupancy1=np.array([[0 for i in range(cell_y)] for j in range(cell_x)])\n",
    "    occupancy2=np.array([[0 for i in range(cell_y)] for j in range(cell_x)])\n",
    "  \n",
    "    # We iterate through the lists of inliers and mark the occupied cells.\n",
    "    iter=len(pts1)\n",
    "    for i in range(iter):\n",
    "        #Occupancy 1:\n",
    "        x1=pts1[i][0]   # X position on the image plane.\n",
    "        y1=pts1[i][1]   # Y position on the image plane.\n",
    "        # Adjusting to the size of the occupancy cell:\n",
    "        x1_idx=np.int32((x1*cell_x)/size[1])    \n",
    "        y1_idx=np.int32((y1*cell_y)/size[0])\n",
    "        # Marking the cell as occupied:\n",
    "        occupancy1[x1_idx][y1_idx]=1\n",
    "       \n",
    "        #Occupancy 2:\n",
    "        x2=pts2[i][0]   # X position on the image plane.\n",
    "        y2=pts2[i][1]   # Y position on the image plane.\n",
    "        # Adjusting to the size of the occupancy cel\n",
    "        x2_idx=np.int32((x2*cell_x)/size[1])\n",
    "        y2_idx=np.int32((y2*cell_y)/size[0])\n",
    "        # Marking the cell as occupied:\n",
    "        occupancy2[x2_idx][y2_idx]=1\n",
    "    \n",
    "    # Summing the occupied cells:\n",
    "    sum1=0\n",
    "    sum2=0\n",
    "    for i in range(cell_y):\n",
    "        for j in range(cell_x):\n",
    "            if(occupancy1[j][i]==1):\n",
    "                sum1+=1\n",
    "            if(occupancy2[j][i]==1):\n",
    "                sum2+=1\n",
    "    \n",
    "    # Calculating the occupancy percentage as the number of occupied cells divided by the total number of cells:\n",
    "    occupancy_size=cell_x*cell_y\n",
    "    w1=np.float64(sum1/occupancy_size)\n",
    "    w2=np.float64(sum2/occupancy_size)\n",
    "\n",
    "    # Selecting the maximum occupancy value as the co-visibility percentage between the images.\n",
    "    w=np.max([w1,w2])   # This is done to achieve a symmetric adjacency matrix for the Appearance Graph.\n",
    "\n",
    "    return w"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def overlap_matrix(im_dir,im_path,umb_F):\n",
    "    # This function calculates the co-visibility matrix for a threshold value 'umb_F' from a list of image names.\n",
    "    # The co-visibility detection process is carried out with all the image pairs in the list\n",
    "    #   -Input:\n",
    "    #       im_dir          ---> List with image names\n",
    "    #       im_path         ---> Image directory\n",
    "    #       umb_F           ---> Threshold number of inliers to determine the existence of co-visibility\n",
    "    #   -Return: \n",
    "    #       overlap         ---> Binary co-visibility matrix, save True or False depending on whether co-visibility is detected\n",
    "    #       covisibilidad   ---> Real co-visibility matrix with the percentage of overlap between images\n",
    "\n",
    "    # Calculation of the number of iterations to show the evolution of the process\n",
    "    im_size=len(im_dir)\n",
    "    total_iter=(im_size**2 + im_size)/2 \n",
    "    cont_iter=0\n",
    "\n",
    "\n",
    "    # Return matrix:\n",
    "    overlap=np.array([[0 for i in range(im_size)] for j in range(im_size)]) # Matrix of binary co-visibility matrix, the result of the co-visibility detection is saved\n",
    "    covisibilidad=np.array([[0 for i in range(im_size)] for j in range(im_size)])       # Real co-visibility matrix, the percentage of co-visibility is saved\n",
    "\n",
    "    # SIFT descriptor class:\n",
    "    sift=cv.xfeatures2d.SIFT_create()\n",
    "\n",
    "    # Loop comparing pairs of images:\n",
    "    for i in range(im_size):\n",
    "        # Load image 1\n",
    "        suffix1=im_dir[i]\n",
    "        im1=cv.imread(im_path+'/'+suffix1)\n",
    "\n",
    "        # Calculate SIFT descriptor for image 1, return feature descriptors (d) and keypoints (kp)\n",
    "        kp1, des1 = sift.detectAndCompute(im1,None)\n",
    "        \n",
    "        for j in range(i,im_size):\n",
    "                # Load image 2\n",
    "                suffix2=im_dir[j]\n",
    "                im2=cv.imread(im_path+'/'+suffix2)\n",
    "                \n",
    "                # If they are the same image (i==j) assign co-visibility equal to 1\n",
    "                if i==j:\n",
    "                    overlap[i][j]=True\n",
    "                    covisibilidad[i][j]=1\n",
    "                    \n",
    "                else:\n",
    "                    # Calculate SIFT descriptor for image 2, return feature descriptors (d) and keypoints (kp)\n",
    "                    kp2, des2 = sift.detectAndCompute(im2,None)\n",
    "\n",
    "                    # Descriptor matching by the Brute-Force-Matcher (BFM) method\n",
    "                    if len(kp1) > 0 and len(kp2) > 0:\n",
    "                        # BFMatcher\n",
    "                        bf = cv.BFMatcher(cv.NORM_L2, crossCheck=True)      # BFM class with positive cross-correlation (crossCheck=True).\n",
    "                        matches = bf.match(des1,des2)                       # Matching of feature descriptors (d).\n",
    "                        matches=sorted(matches, key= lambda x:x.distance)   # The matches are ordered from least to greatest distance between pairs of descriptors.\n",
    "                        if len(matches)>100: matches=matches[:100]          # It operates with the 100 matches with the shortest distance, the best correspondences.\n",
    "                        \n",
    "                        # We obtain keypoints with correspondence:\n",
    "                        pts1=[]\n",
    "                        pts2=[]\n",
    "\n",
    "                        for mat in matches:\n",
    "                            pts1.append(kp1[mat.queryIdx].pt)\n",
    "                            pts2.append(kp2[mat.trainIdx].pt)\n",
    "                        \n",
    "                        pts1=np.int32(np.array(pts1))\n",
    "                        pts2=np.int32(np.array(pts2))\n",
    "                        \n",
    "\n",
    "                         # Calculation of the Fundamental Matrix (F) with RANSAC algorithm\n",
    "                        if len(pts1)>=50: \n",
    "                            F, mask = cv.findFundamentalMat(pts1,pts2,cv.FM_RANSAC,3,0.99,10) # Fundamental Matrix calculation, nº iterations in RANSAC algorithm = 10\n",
    "\n",
    "                            # We select inliers, pairs of keypoints represented by the Fundamental Matrix obtained\n",
    "                            pts1 = pts1[mask.ravel()==1]\n",
    "                            pts2 = pts2[mask.ravel()==1]\n",
    "            \n",
    "                            # Co-visibility detection based on the number of inliers threshold\n",
    "                            if len(pts1)>umb_F:\n",
    "                                OVERLAP_ENABLE=True\n",
    "\n",
    "                                # We store positive co-visibility\n",
    "                                overlap[i, j]=True\n",
    "                                overlap[j, i]=True\n",
    "\n",
    "                                # We calculate the percentage of co-visibility\n",
    "                                tam=im1.shape\n",
    "                                # Occupancy cell size\n",
    "                                cell_x=4\n",
    "                                cell_y=np.int32((tam[0]/tam[1])*cell_x)\n",
    "\n",
    "                                # Occupancy percentage calculation\n",
    "                                cov_ratio=occupancy(pts1,pts2,cell_x,cell_y,tam)\n",
    "                                # Save occupancy percentage in symmetric co-visibility matrix\n",
    "                                covisibilidad[i, j]=cov_ratio\n",
    "                                covisibilidad[j, i]=cov_ratio\n",
    "\n",
    "                                del cov_ratio\n",
    "\n",
    "                                del pts1,pts2\n",
    "\n",
    "                # Print percentage of execution:\n",
    "                cont_iter+=1\n",
    "                if j%10==0: print('Proceso: ',np.round((cont_iter/total_iter)*100,decimals=6),'%',end='\\r')\n",
    "    \n",
    "    return overlap, covisibilidad"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ground-truth Co-visibility Matrix\n",
    "This matrix stores the actual co-visibility values between pairs of images in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def gt_matrix(im_path):\n",
    "    # This function calculates the co-visibility matrix of true values from the name of the elements of the ordered dataset.\n",
    "    # In this dataset, co-visible images are saved with a characteristic name of the place they represent.\n",
    "\n",
    "    im_dir=sorted(os.listdir(im_path))  # Images names list\n",
    "    mat_size=len(im_dir)\n",
    "\n",
    "    # Initialization of the ground-truth array with the size of the list of images\n",
    "    ground_truth=[[0 for i in range(mat_size)] for j in range(mat_size)]\n",
    "\n",
    "    # The list of images is traversed in a nested loop to compare their names:\n",
    "    for i, suffix1 in enumerate(im_dir):    \n",
    "        for j, suffix2 in enumerate(im_dir):\n",
    "            # If the name of the images are the same, determine co-visibility 1\n",
    "            if suffix1 == suffix2: ground_truth[i][j]=1\n",
    "\n",
    "            else:\n",
    "                # Gets the place tag of the image names\n",
    "                label1=suffix1.partition(\"_\")[0]\n",
    "                label2=suffix2.partition(\"_\")[0]\n",
    "                # If the tags match they are marked as co-visible:\n",
    "                if label1 == label2: ground_truth[i][j]=1\n",
    "                \n",
    "    return ground_truth"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluation - Precision and Recall\n",
    "In this part, the function that calculates the precision and sensitivity of an estimated co-visibility matrix is implemented by comparing it with the ground-truth matrix."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def eval_PR(covisibilidad, ground_truth):\n",
    "# This function compares element by element the values of the estimated co-visibility matrix and the ground-truth matrix.\n",
    "# Depending on whether the prediction is correct or not, 4 situations are determined: True Positive, False Positive and False Negative.\n",
    "# Based on these values, the precision and sensitivity are calculated\n",
    "#   -Input:\n",
    "#       covisibilidad  ---> Estimated co-visibility matrix\n",
    "#       ground_truth   ---> Ground-truth co-visibility matrix\n",
    "#   -Return: \n",
    "#       Precision      ---> Precision obtained by the estimated co-visibility matrix\n",
    "#       Recall         ---> Sensitivity obtained by the estimated co-visibility matrix\n",
    "\n",
    "    # Initialization of variables\n",
    "    TP=0    # True Positive: The positive co-visibility estimate is correct\n",
    "    FP=0    # False Positive: The positive co-visibility estimate is incorrect\n",
    "    FN=0    # False Negative: The negative co-visibility estimate is incorrect\n",
    "\n",
    "    # The rows and columns of the co-visibility and ground truth matrices are traversed\n",
    "    for i in range(covisibilidad.shape[0]):\n",
    "        for j in range(covisibilidad.shape[1]):\n",
    "            # The cases of TP, FP and FN are determined:\n",
    "            if covisibilidad[i][j] == True and ground_truth[i][j]==True: TP+=1\n",
    "            if covisibilidad[i][j] == False and ground_truth[i][j]==True: FN+=1\n",
    "            if covisibilidad[i][j] == True and ground_truth[i][j]==False: FP+=1\n",
    "\n",
    "    # Precision and recall values are calculated\n",
    "    Precision = TP/(TP+FP)\n",
    "    Recall = TP/(TP+FN)\n",
    "    \n",
    "    return Precision, Recall"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Co-visibility evaluation\n",
    "In this part, a sweep of the threshold is performed, which determines the number of inliers of the co-visibility detection process.\n",
    "\n",
    "The Precision and Recall values for each matrix obtained are compared."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define range of values\n",
    "trh_F=np.arange(2,40,2) \n",
    "trh_F=[int(i) for i in trh_F]\n",
    "\n",
    "# Directory:\n",
    "im_path=\".../Test/Images\"             # Directory of images of the controlled dataset\n",
    "save_im_path=\".../Covisibilidad_trhVariable/\"  # Directory to save the co-visibility matrices obtained\n",
    "seq_arr=[\"cloudy\",\"night\",\"sunny\"]      # Different sequences of the controlled dataset for the evaluation\n",
    "\n",
    "# Iterates over the different datasets\n",
    "for seq in seq_arr:\n",
    "    # Initialize precision and sensitivity lists for different threshold values\n",
    "    P_hist=[]\n",
    "    R_hist=[]\n",
    "\n",
    "    # A co-visibility matrix is calculated for each value in the range of values of the parameter\n",
    "    for i in range(len(trh_F)):\n",
    "        # Images name list\n",
    "        im_dir=sorted(os.listdir(im_path+seq))\n",
    "\n",
    "        # Indicates execution process:\n",
    "        print('Epoch: ',i,' / ',len(trh_F))\n",
    "\n",
    "        # Calculation of estimated co-visibility matrices and ground-truth\n",
    "        overlap,w, match_arr=overlap_matrix(im_dir,im_path+seq,trh_F[i])    # The co-visibility matrix is calculated for each threshold\n",
    "        ground_truth=gt_matrix(im_path+seq)\n",
    "\n",
    "        # Saves the co-visibility matrix obtained\n",
    "        fig_path=save_im_path+\"Fig_COLD_\"+seq+\"_ovl_Thr\"+str(trh_F[i])+\".npy\"\n",
    "        np.save(fig_path,w)\n",
    "\n",
    "        # Calculation of precision and sensitivity for the co-visibility matrix obtained\n",
    "        P, R = eval_PR(overlap,ground_truth)\n",
    "\n",
    "        # Saves the evaluation result for each threshold\n",
    "        P_hist.append(P)\n",
    "        R_hist.append(R)\n",
    "\n",
    "    # Saves precision and sensitivity lists for each dataset used\n",
    "    save_dict={}\n",
    "    save_dict.update({\"Precision\":P_hist})\n",
    "    save_dict.update({\"Recall\":R_hist})\n",
    "    save_dict.update({\"Trheshold_cov\":trh_F})\n",
    "\n",
    "    save_json=\".../Utils/Cov_PR_COLD_\"+seq+\".json\"\n",
    "    with open(save_json, \"w\") as outfile:\n",
    "        json.dump(save_dict, outfile)\n",
    "    "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot results - PR curve\n",
    "The precision and recall curves are plotted for the different threshold values on the three datasets used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "cloudy_eval=json.load(open(\".../Utils/Cov_PR_COLD_cloudy.json\"))\n",
    "night_eval=json.load(open(\".../Utils/Cov_PR_COLD_night.json\"))\n",
    "sunny_eval=json.load(open(\".../Utils/Cov_PR_COLD_sunny.json\"))\n",
    "\n",
    "Trh=cloudy_eval[\"Trheshold_cov\"]\n",
    "n=len(Trh)\n",
    "cloudy_P=cloudy_eval[\"Precision\"]\n",
    "cloudy_R=cloudy_eval[\"Recall\"]\n",
    "night_P=night_eval[\"Precision\"]\n",
    "night_R=night_eval[\"Recall\"]\n",
    "sunny_P=sunny_eval[\"Precision\"]\n",
    "sunny_R=sunny_eval[\"Recall\"]\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Precision curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Curva de Precisión para la detección de Co-visibilidad')"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create figure\n",
    "plt.figure(figsize=[7,5])\n",
    "plt_curve=plt.axes()\n",
    "\n",
    "# Step through the precision lists for each threshold value:\n",
    "for k in range(len(Trh)):\n",
    "    # Plot points on precision curve:\n",
    "    plt_curve.plot(Trh[k],cloudy_P[k],marker='^',markersize=10,color='r') #,label=my_label[k])\n",
    "    plt_curve.plot(Trh[k],night_P[k],marker='^',markersize=10,color='b') #,label=my_label[k])\n",
    "    plt_curve.plot(Trh[k],sunny_P[k],marker='^',markersize=10,color='g') #,label=my_label[k])\n",
    "    \n",
    "# Plot precision curve:\n",
    "plt_curve.plot(Trh,cloudy_P,color='r', linewidth=2,linestyle='--',label='Seq2_cloudy1')\n",
    "plt_curve.plot(Trh,night_P,color='b', linewidth=2,linestyle='--',label='Seq2_night1')\n",
    "plt_curve.plot(Trh,sunny_P,color='g', linewidth=2,linestyle='--',label='Seq2_sunny1')\n",
    "\n",
    "# Show plot:\n",
    "plt.legend()\n",
    "plt.xlim(3,32)\n",
    "plt.grid()\n",
    "plt.xlabel('Threshold')\n",
    "plt.ylabel('Precision')\n",
    "plt.title(\"Precision curve for Co-visibility detection\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Recall')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create figure\n",
    "plt.figure(figsize=[7,5])\n",
    "plt_curve=plt.axes()\n",
    " \n",
    "# Step through the recall lists for each threshold value:\n",
    "for k in range(len(Trh)):\n",
    "    # Plot points on recall curve:\n",
    "    plt_curve.plot(Trh[k],cloudy_R[k],marker='^',markersize=10,color='r') \n",
    "    plt_curve.plot(Trh[k],night_R[k],marker='^',markersize=10,color='b') \n",
    "    plt_curve.plot(Trh[k],sunny_R[k],marker='^',markersize=10,color='g') \n",
    "   \n",
    "# Plot recall curve\n",
    "plt_curve.plot(Trh,cloudy_R,color='r',linewidth=2,linestyle='--',label='Seq2_cloudy1')\n",
    "plt_curve.plot(Trh,night_R,color='b',linewidth=2,linestyle='--',label='Seq2_night1')\n",
    "plt_curve.plot(Trh,sunny_R,color='g',linewidth=2,linestyle='--',label='Seq2_sunny1')\n",
    "\n",
    "# Show plot\n",
    "plt.legend()\n",
    "plt.title(\"Curva de Sensibilidad para la detección de Co-visibilidad\")\n",
    "plt.xticks(Trh, Trh)\n",
    "plt.grid()\n",
    "plt.xlabel('Umbral')\n",
    "plt.ylabel('Recall')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Precision-Recall curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AUC para Seq2_cloudy1:  0.9596975523326542\n",
      "AUC para Seq2_night1:  0.9836591501628735\n",
      "AUC para Seq2_sunny1:  0.9877299145450056\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# In this part, the precision-sensitivity curves are plotted for the 3 datasets separately\n",
    "\n",
    "# Precision and Sensitivity lists for the three datasets\n",
    "Precision=[cloudy_P,night_P,sunny_P]\n",
    "Recall=[cloudy_R,night_R,sunny_R]\n",
    "\n",
    "my_color=['r','b','g']  # Define colors of the plot:\n",
    "my_label=['Seq2_cloudy1','Seq2_night1','Seq2_sunny1']   # Define labels for the plot\n",
    "\n",
    "for i in range(3):\n",
    "    # Create the figure, one for each dataset\n",
    "    plt.figure(figsize=[7,5])\n",
    "    plt_curve=plt.axes()\n",
    "\n",
    "    # Copy Precision and Recall lists\n",
    "    P=np.copy(Precision[i]).tolist()\n",
    "    R=np.copy(Recall[i]).tolist()\n",
    "\n",
    "    # Index the value (P=1, R=0) to correctly calculate the area under the curve\n",
    "    P.extend([float(1)])\n",
    "    R.extend([float(0)])\n",
    "\n",
    "    # Plot points precision and recall\n",
    "    for k in range(len(Trh)):\n",
    "        # Plot points\n",
    "        plt_curve.plot(R[k],P[k],marker='^',markersize=10,color=my_color[i])\n",
    "        # Plot text with the threshold value at even indices\n",
    "        if (k%2)== 0: plt_curve.text(R[k]-0.01,P[k]+0.05,\"μ=\"+str(Trh[k]))\n",
    "  \n",
    "    # Plot precision-recall curve\n",
    "    plt_curve.plot(R,P,color=my_color[i],label=my_label[i])    \n",
    "    plt.fill_between(R,P,color=my_color[i],alpha=0.2)  # Area under the curve shaded\n",
    "\n",
    "    # Calculate the area under the curve\n",
    "    auc_score = auc(R, P)\n",
    "   \n",
    "    # Show plot:\n",
    "    plt.grid()\n",
    "    plt.ylim(0,1.1)\n",
    "    plt.xlim(R[-2]-0.05,1.015)\n",
    "    plt.xlabel('Recall')\n",
    "    plt.ylabel('Precission')\n",
    "    plt.title(\"Curva de Precisión-Sensibilidad - \"+my_label[i])\n",
    "\n",
    "    # Print area under the curve:\n",
    "    print('AUC para '+my_label[i]+': ',auc_score)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.10 64-bit",
   "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": "3.7.15"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "767d51c1340bd893661ea55ea3124f6de3c7a262a8b4abca0554b478b1e2ff90"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}