notebooks/task3_point_detect_eval_demo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Demonstration of evaluation tools for Task 3\n",
    "Evaluate and analyze results for Task 3: \"Detection of graticule lines intersections\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Those are the imports you may need.\n",
    "from icdar21_mapseg_eval.point_detection import (\n",
    "    eval_pt_detect, show_predictions, show_predictions_classified, plot_f_vs_dist_curve)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extra imports for the demo\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# This is arbitrary, for the competition `radius_limit` is set to 50 pixels.\n",
    "radius_limit = 118\n",
    "# Beta for the F_beta score\n",
    "beta = 0.5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sample data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "ground_truth = np.float32([\n",
    "    [5710, 1170],  # A\n",
    "    [8080, 1170],  # B\n",
    "    [3330, 3530],  # C\n",
    "    [5710, 3550],  # D\n",
    "    [8085, 3540],  # E\n",
    "    [3327, 5922],  # F\n",
    "    [5715, 5940],  # G\n",
    "    [8085, 5942]]) # H"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted = np.float32([\n",
    "    # A\n",
    "    [5710, 1170], # exact match\n",
    "    # B\n",
    "    [8080 + 2*radius_limit, 1170+2*radius_limit], # Match outside acceptable area\n",
    "    # C\n",
    "    [3330+10, 3530+10], # multiple acceptable matches\n",
    "    [3330-10, 3530-10],\n",
    "    [3330+10, 3530+0],\n",
    "    [3330+10, 3530+30],\n",
    "    # D\n",
    "    [5710+10, 3550-10], # 1 good match\n",
    "    # E\n",
    "    [8085+radius_limit, 3540], # far match, on the edge (will be accepted)\n",
    "    # F\n",
    "    # Nothing, no match\n",
    "    # G and H\n",
    "    [(5715+8085)/2, (5940+5942)/2] # point on the perpendicular bisector of the two points\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Visualize raw predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_predictions(ground_truth, predicted, radius_limit)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Evaluation metrics for a single image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Considering predicted point 00: ( 5710.0, 1170.0)\n",
      "\tMatches gt point 00: ( 5710.0, 1170.0) @ 0.0\n",
      "\tFirst match (good)\n",
      "\ttp:  1, fn:  7, fp:  8, |expt|:8, |pred|:9, x:0.00, y:0.11, area:0.000\n",
      "Considering predicted point 04: ( 3340.0, 3530.0)\n",
      "\tMatches gt point 02: ( 3330.0, 3530.0) @ 10.0\n",
      "\tFirst match (good)\n",
      "\ttp:  2, fn:  6, fp:  7, |expt|:8, |pred|:9, x:0.08, y:0.23, area:0.014\n",
      "Considering predicted point 02: ( 3340.0, 3540.0)\n",
      "\tMatches gt point 02: ( 3330.0, 3530.0) @ 14.1\n",
      "\tEXTRA MATCH (noise)\n",
      "Considering predicted point 03: ( 3320.0, 3520.0)\n",
      "\tMatches gt point 02: ( 3330.0, 3530.0) @ 14.1\n",
      "\tEXTRA MATCH (noise)\n",
      "Considering predicted point 06: ( 5720.0, 3540.0)\n",
      "\tMatches gt point 03: ( 5710.0, 3550.0) @ 14.1\n",
      "\tFirst match (good)\n",
      "\ttp:  3, fn:  5, fp:  6, |expt|:8, |pred|:9, x:0.12, y:0.34, area:0.024\n",
      "Considering predicted point 05: ( 3340.0, 3560.0)\n",
      "\tMatches gt point 02: ( 3330.0, 3530.0) @ 31.6\n",
      "\tEXTRA MATCH (noise)\n",
      "Considering predicted point 07: ( 8203.0, 3540.0)\n",
      "\tMatches gt point 04: ( 8085.0, 3540.0) @ 118.0\n",
      "\tFirst match (good)\n",
      "\ttp:  4, fn:  4, fp:  5, |expt|:8, |pred|:9, x:1.00, y:0.45, area:0.374\n",
      "Considering predicted point 01: ( 8316.0, 1406.0)\n",
      "Reached radius limit (118)\n",
      "Considering predicted point 08: ( 6900.0, 5941.0)\n",
      "Reached radius limit (118)\n",
      "Propagating to right limit\n",
      "x:1.00, y:0.45, area:0.37\n",
      "Score: 0.374\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# debug=True displays matches\n",
    "# score is the AUC of f_beta vs distance\n",
    "# details is a dataframe containing match and score information for each predicted point\n",
    "# plt_data is the curve to plot\n",
    "score, details, plt_data = eval_pt_detect(ground_truth, predicted, radius_limit, \n",
    "                                          beta=beta, debug=True, show_plot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.2915897501635441"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# A different radius_limit gives a different result\n",
    "score, details, plt_data = eval_pt_detect(ground_truth, predicted, radius_limit/3, \n",
    "                                          beta=beta, debug=False, show_plot=True)\n",
    "score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Precision</th>\n",
       "      <th>Recall</th>\n",
       "      <th>F-beta</th>\n",
       "      <th>True Positives</th>\n",
       "      <th>False Positives</th>\n",
       "      <th>False Negatives</th>\n",
       "      <th>Error type</th>\n",
       "      <th>Prediction x</th>\n",
       "      <th>Prediction y</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Distance</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.000000</th>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.125</td>\n",
       "      <td>0.113636</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>5710.0</td>\n",
       "      <td>1170.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10.000000</th>\n",
       "      <td>0.222222</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0.227273</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3340.0</td>\n",
       "      <td>3530.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14.142136</th>\n",
       "      <td>0.222222</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0.227273</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>3340.0</td>\n",
       "      <td>3540.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14.142136</th>\n",
       "      <td>0.222222</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0.227273</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>3320.0</td>\n",
       "      <td>3520.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14.142136</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.340909</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>5720.0</td>\n",
       "      <td>3540.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31.622776</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.340909</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>3340.0</td>\n",
       "      <td>3560.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118.000000</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.340909</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>8203.0</td>\n",
       "      <td>3540.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>333.754395</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.340909</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>8316.0</td>\n",
       "      <td>1406.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1185.000366</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.340909</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>6900.0</td>\n",
       "      <td>5941.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Precision  Recall    F-beta  True Positives  False Positives  \\\n",
       "Distance                                                                    \n",
       "0.000000      0.111111   0.125  0.113636               1                8   \n",
       "10.000000     0.222222   0.250  0.227273               2                7   \n",
       "14.142136     0.222222   0.250  0.227273               2                7   \n",
       "14.142136     0.222222   0.250  0.227273               2                7   \n",
       "14.142136     0.333333   0.375  0.340909               3                6   \n",
       "31.622776     0.333333   0.375  0.340909               3                6   \n",
       "118.000000    0.333333   0.375  0.340909               3                6   \n",
       "333.754395    0.333333   0.375  0.340909               3                6   \n",
       "1185.000366   0.333333   0.375  0.340909               3                6   \n",
       "\n",
       "             False Negatives  Error type  Prediction x  Prediction y  \n",
       "Distance                                                              \n",
       "0.000000                   7           0        5710.0        1170.0  \n",
       "10.000000                  6           0        3340.0        3530.0  \n",
       "14.142136                  6           1        3340.0        3540.0  \n",
       "14.142136                  6           1        3320.0        3520.0  \n",
       "14.142136                  5           0        5720.0        3540.0  \n",
       "31.622776                  5           1        3340.0        3560.0  \n",
       "118.000000                 5           2        8203.0        3540.0  \n",
       "333.754395                 5           2        8316.0        1406.0  \n",
       "1185.000366                5           2        6900.0        5941.0  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Matching details provide extra help for debugging:\n",
    "details"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Code for error types:\n",
    "\n",
    "- `0`: correct match (true positive)\n",
    "- `1`: extra match withing acceptable range (error)\n",
    "- `2`: too far -- out of acceptable range (error)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Curve used to compute the AUC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_f_vs_dist_curve(plt_data, radius_limit, beta=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# We can check we obtain the same results as the auc function from scikit-learn\n",
    "from sklearn.metrics import auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2915897501635441"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc(plt_data[\"Normalized distance\"], plt_data[\"F-beta\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2915897501635441"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Normalized distance</th>\n",
       "      <th>F-beta</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.113636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.254237</td>\n",
       "      <td>0.227273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.359546</td>\n",
       "      <td>0.340909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.340909</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Normalized distance    F-beta\n",
       "0             0.000000  0.000000\n",
       "1             0.000000  0.113636\n",
       "2             0.254237  0.227273\n",
       "3             0.359546  0.340909\n",
       "4             1.000000  0.340909"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Show predictions according to error type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We can show predictions with a different color according to whether they are correct or not\n",
    "show_predictions_classified(ground_truth, details, radius_limit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# And you can save the figure for further study\n",
    "show_predictions_classified(ground_truth, details, radius_limit, filename=\"out.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 97. More tests\n",
    "We keep here some tests to show the implementation is correct."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Propagating to right limit\n",
      "x:1.00, y:0.00, area:0.00\n",
      "Score: 0.000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.0,\n",
       " Empty DataFrame\n",
       " Columns: [Precision, Recall, F-beta, True Positives, False Positives, False Negatives, Error type, Prediction x, Prediction y]\n",
       " Index: [])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Empty prediction: zero auc\n",
    "area, df, _ = eval_pt_detect(np.array([(0,0), (1,1)]), np.empty((0,2)), 10, beta=0.5, debug=True, show_plot=True)\n",
    "area, df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Propagating to right limit\n",
      "x:1.00, y:0.00, area:0.00\n",
      "Score: 0.000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.0,\n",
       " Empty DataFrame\n",
       " Columns: [Precision, Recall, F-beta, True Positives, False Positives, False Negatives, Error type, Prediction x, Prediction y]\n",
       " Index: [])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Empty prediction and ground truth\n",
    "area, df, _ = eval_pt_detect(np.empty((0,2)), np.empty((0,2)), 10, beta=0.5, debug=True, show_plot=True)\n",
    "area, df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Considering predicted point 00: (    0.0,    0.0)\n",
      "Reached radius limit (10)\n",
      "Considering predicted point 01: (    1.0,    1.0)\n",
      "Reached radius limit (10)\n",
      "Propagating to right limit\n",
      "x:1.00, y:0.00, area:0.00\n",
      "Score: 0.000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.0,\n",
       "           Precision  Recall  F-beta  True Positives  False Positives  \\\n",
       " Distance                                                               \n",
       " inf             0.0       0     0.0               0                2   \n",
       " inf             0.0       0     0.0               0                2   \n",
       " \n",
       "           False Negatives  Error type  Prediction x  Prediction y  \n",
       " Distance                                                           \n",
       " inf                     0           2             0             0  \n",
       " inf                     0           2             1             1  )"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Empty ground truth\n",
    "area, df, df_dbg = eval_pt_detect(np.empty((0,2)), np.array([(0,0), (1,1)]), 10, beta=0.5, debug=True, show_plot=True)\n",
    "area, df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Considering predicted point 00: (    1.0,    1.0)\n",
      "\tMatches gt point 01: (    1.0,    1.0) @ 0.0\n",
      "\tFirst match (good)\n",
      "\ttp:  1, fn:  1, fp:  0, |expt|:2, |pred|:1, x:0.00, y:0.83, area:0.000\n",
      "Propagating to right limit\n",
      "x:1.00, y:0.83, area:0.83\n",
      "Score: 0.833\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.8333333333333334,\n",
       "           Precision  Recall    F-beta  True Positives  False Positives  \\\n",
       " Distance                                                                 \n",
       " 0.0             1.0     0.5  0.833333               1                0   \n",
       " \n",
       "           False Negatives  Error type  Prediction x  Prediction y  \n",
       " Distance                                                           \n",
       " 0.0                     1           0             1             1  ,\n",
       "    Normalized distance    F-beta\n",
       " 0                  0.0  0.000000\n",
       " 1                  0.0  0.833333\n",
       " 2                  1.0  0.833333)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1 perfect match over 2 targets\n",
    "eval_pt_detect(np.array([(0,0), (1,1)]), np.array([(1,1)]), 10, beta=0.5, debug=True, show_plot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Considering predicted point 00: (    0.0,    0.0)\n",
      "\tMatches gt point 00: (    0.0,    0.0) @ 0.0\n",
      "\tFirst match (good)\n",
      "\ttp:  1, fn:  1, fp:  0, |expt|:2, |pred|:1, x:0.00, y:0.83, area:0.000\n",
      "Propagating to right limit\n",
      "x:1.00, y:0.83, area:0.83\n",
      "Score: 0.833\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.8333333333333334,\n",
       "           Precision  Recall    F-beta  True Positives  False Positives  \\\n",
       " Distance                                                                 \n",
       " 0.0             1.0     0.5  0.833333               1                0   \n",
       " \n",
       "           False Negatives  Error type  Prediction x  Prediction y  \n",
       " Distance                                                           \n",
       " 0.0                     1           0             0             0  ,\n",
       "    Normalized distance    F-beta\n",
       " 0                  0.0  0.000000\n",
       " 1                  0.0  0.833333\n",
       " 2                  1.0  0.833333)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# same\n",
    "eval_pt_detect(np.array([(0,0), (1,1)]), np.array([(0,0)]), 10, beta=0.5, debug=True, show_plot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Considering predicted point 00: (    0.0,    0.0)\n",
      "\tMatches gt point 00: (    0.0,    0.0) @ 0.0\n",
      "\tFirst match (good)\n",
      "\ttp:  1, fn:  0, fp:  1, |expt|:1, |pred|:2, x:0.00, y:0.56, area:0.000\n",
      "Considering predicted point 01: (    1.0,    1.0)\n",
      "\tMatches gt point 00: (    0.0,    0.0) @ 1.4\n",
      "\tEXTRA MATCH (noise)\n",
      "Propagating to right limit\n",
      "x:1.00, y:0.56, area:0.56\n",
      "Score: 0.556\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.5555555555555556,\n",
       "           Precision  Recall    F-beta  True Positives  False Positives  \\\n",
       " Distance                                                                 \n",
       " 0.000000        0.5     1.0  0.555556               1                1   \n",
       " 1.414214        0.5     1.0  0.555556               1                1   \n",
       " \n",
       "           False Negatives  Error type  Prediction x  Prediction y  \n",
       " Distance                                                           \n",
       " 0.000000                0           0             0             0  \n",
       " 1.414214                0           1             1             1  ,\n",
       "    Normalized distance    F-beta\n",
       " 0                  0.0  0.000000\n",
       " 1                  0.0  0.555556\n",
       " 2                  1.0  0.555556)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1 perfect match and 1 extra match\n",
    "eval_pt_detect(np.array([(0,0)]), np.array([(0,0), (1,1)]), 10, beta=0.5, debug=True, show_plot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Considering predicted point 00: (    2.0,    0.0)\n",
      "\tMatches gt point 00: (    0.0,    0.0) @ 2.0\n",
      "\tFirst match (good)\n",
      "\ttp:  1, fn:  1, fp:  0, |expt|:2, |pred|:1, x:0.50, y:0.83, area:0.208\n",
      "Propagating to right limit\n",
      "x:1.00, y:0.83, area:0.62\n",
      "Score: 0.625\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.625,\n",
       "           Precision  Recall    F-beta  True Positives  False Positives  \\\n",
       " Distance                                                                 \n",
       " 2.0             1.0     0.5  0.833333               1                0   \n",
       " \n",
       "           False Negatives  Error type  Prediction x  Prediction y  \n",
       " Distance                                                           \n",
       " 2.0                     1           0             2             0  ,\n",
       "    Normalized distance    F-beta\n",
       " 0                  0.0  0.000000\n",
       " 1                  0.5  0.833333\n",
       " 2                  1.0  0.833333)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1 match, 1 miss\n",
    "eval_pt_detect(np.array([(0,0), (10,10)]), np.array([(2,0)]), 4, beta=0.5, debug=True, show_plot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Considering predicted point 00: (   10.0,    0.0)\n",
      "\tMatches gt point 00: (    0.0,    0.0) @ 10.0\n",
      "\tFirst match (good)\n",
      "\ttp:  1, fn:  0, fp:  0, |expt|:1, |pred|:1, x:1.00, y:1.00, area:0.500\n",
      "Propagating to right limit\n",
      "x:1.00, y:1.00, area:0.50\n",
      "Score: 0.500\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.5,\n",
       "           Precision  Recall  F-beta  True Positives  False Positives  \\\n",
       " Distance                                                               \n",
       " 10.0            1.0     1.0     1.0               1                0   \n",
       " \n",
       "           False Negatives  Error type  Prediction x  Prediction y  \n",
       " Distance                                                           \n",
       " 10.0                    0           0            10             0  ,\n",
       "    Normalized distance  F-beta\n",
       " 0                  0.0     0.0\n",
       " 1                  1.0     1.0\n",
       " 2                  1.0     1.0)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1 match on edge\n",
    "eval_pt_detect(np.array([(0,0)]), np.array([(10,0)]), 10, beta=0.5, debug=True, show_plot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Considering predicted point 00: (    0.0,    0.0)\n",
      "\tMatches gt point 00: (    0.0,    0.0) @ 0.0\n",
      "\tFirst match (good)\n",
      "\ttp:  1, fn:  0, fp:  1, |expt|:1, |pred|:2, x:0.00, y:0.56, area:0.000\n",
      "Considering predicted point 01: (    0.0,    0.0)\n",
      "\tMatches gt point 00: (    0.0,    0.0) @ 0.0\n",
      "\tEXTRA MATCH (noise)\n",
      "Propagating to right limit\n",
      "x:1.00, y:0.56, area:0.56\n",
      "Score: 0.556\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAm0AAAJcCAYAAACrLZO2AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAlDklEQVR4nO3de7ztdV3v+/dHbt5BBTNZKB7FlIQtuFLTSre6EyzB00UFPaa5Zbd3mJX22LYzc7vrPI6607ZKGpW3DBWzbBkUu1Oo56i4XYiigHYWpLHABBTJvHL5nD/Gb+lkMudac03mHHN+8fl8POaDcfmN3/iMOViLF7/fb4xfdXcAANjcbrfRAwAAsGeiDQBgAKINAGAAog0AYACiDQBgAKINAGAAog0YRlW9v6r+/Tqt+79U1R+tx7oB1oJoA9ZcVX2uqr5RVf+64Of1Gz3XLlX12KraufC27v4/u3tdghBgLey70QMAt1lP7u7/e6OHuK2oqn27+4aNngPYOLa0AXNTVQdU1Veq6iELbjtk2ip3z6q6W1X9VVVdXVXXTpe3LLOul1XV2xdcP7yquqr2na4/p6ouqaqvVtVlVfUfptvvlOSvk9x7wVbAey+xvhOq6qJp3vdX1YMX3Pe5qnpRVV1YVddV1buq6va7ed3PWzDLxVV17HR7V9UDFiz3lqr67enyY6tqZ1X956r65yRvntbxkwuW33f6Xe1a3yOr6sPTzJ+sqseu8K0BBiDagLnp7m8l+fMkJy24+alJPtDdV2X2d9Kbk9w3yX2SfCPJanerXpXkJ5PcNclzkrymqo7t7q8lOT7Jld195+nnyoUPrKoHJnlHkl9OckiSs5O8r6r2XzT3cUnul+ToJM9eaoiq+tkkL0vyrGmWE5J8aYWv4V5J7p7Z7+OUaaaFv7snJrmmuz9eVYcmOSvJb0+PeVGS91TVISt8LmCTE23AennvtMVn18/zptvPSPL0BcudPN2W7v5Sd7+nu7/e3V9N8jtJHrOaJ+/us7r70p75QJL/meRHV/jwpyU5q7v/truvT/Lfk9whyaMWLPPa7r6yu7+c5H1JHrrMuv59kld298emWXZ09+dXOMdNSX6ru7/V3d/I7Pd0QlXdcbr/5MxCLkmemeTs7j67u2/q7r9Nsj3Jk1b4XMAm55g2YL08ZZlj2s5NcseqekSSL2YWO3+RJFOMvCazLVh3m5a/S1Xt09037s2TV9XxSX4ryQMz+x/UOyb51Aoffu8k3wmr7r6pqi5PcuiCZf55weWvT49ZymFJLl3h8y52dXd/c8EcO6rqkiRPrqr3ZbbV7pjp7vsm+dmqevKCx++X2e8buA0QbcBcdfeNVXVmZrv5vpjkr6ataknywiQ/kOQR3f3PVfXQJBckqSVW9bXMQmyXe+26UFUHJHlPZrsk/7K7r6+q9y5YT+9hzCuTHLVgfZVZfF2xkte4yOVJ7r/MfV/PLV/Dwk+1LjXnrl2kt0tycXfvWPA8f9Ldz1viMcBtgN2jwEY4I7NdkM+YLu9yl8yOY/tKVd09sy1ly/lEkh+rqvtU1YFJfn3BffsnOSDJ1UlumLa6/fiC+7+Y5B7T45ZyZpKfqKrHV9V+mcXkt5J8eIWvb6E/SvKiqnpYzTygqu674DWcXFX7VNVxWdmu4HdOr+U/5ua/u7dntgXuidP6bj99mGHJD3IA4xFtwHp536LvafuLXXd090cz21J278w+ybnL72V27Ng1Sc5L8jfLrXw6ZutdSS5Mcn6Sv1pw31eT/FJm8XVtZsd+bVtw/2cy22J12XS83c12bXb3ZzM7Rux10yxPzuwrTL69l7+DdPe7Mzs274wkX03y3sw+KJAkL5jW/ZXMAva9K1jfF5J8JLPj69614PbLk5yY5L9kFquXJ/m1+HsebjOqe097CQAA2Gj+DwwAYABzjbaqelNVXVVVn17m/qqq11bVjulLK4+d53wAAJvVvLe0vSWzj/Iv5/gkR0w/pyR5wxxmAgDY9OYabd39wSRf3s0iJyZ52/QFlOclOaiqvn8+0wEAbF6b7XvaDs3sE0+77Jxu+8LCharqlMy2xOVOd7rTwx70oAfNbUAAgNU6//zzr+nuVZ1ebrNF24p09+lJTk+SrVu39vbt2zd4IgCAPauqlZ7G7hY226dHr8jsW8d32ZLVfQM5AMBtymaLtm1JnjV9ivSRSa6bvkgSAOB72lx3j1bVO5I8NsnBVbUzs1PU7Jck3f3GJGcneVKSHZmdk+8585wPAGCzmmu0dfdJe7i/k/zinMYBABjGZts9CgDAEkQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAOYabVV1XFV9tqp2VNWLl7j/PlV1blVdUFUXVtWT5jkfAMBmNbdoq6p9kpyW5PgkRyY5qaqOXLTYS5Kc2d3HJHl6kt+f13wAAJvZPLe0PTzJju6+rLu/neSdSU5ctEwnuet0+cAkV85xPgCATWue0XZokssXXN853bbQy5I8s6p2Jjk7yfOXWlFVnVJV26tq+9VXX70eswIAbCqb7YMIJyV5S3dvSfKkJH9SVbeYsbtP7+6t3b31kEMOmfuQAADzNs9ouyLJYQuub5luW+i5Sc5Mku7+SJLbJzl4LtMBAGxi84y2jyU5oqruV1X7Z/ZBg22LlvmnJI9Pkqp6cGbRZv8nAPA9b27R1t03JDk1yTlJLsnsU6IXVdXLq+qEabEXJnleVX0yyTuSPLu7e14zAgBsVvvO88m6++zMPmCw8LaXLrh8cZJHz3MmAIARbLYPIgAAsATRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMIC5RltVHVdVn62qHVX14mWWeWpVXVxVF1XVGfOcDwBgs9p3Xk9UVfskOS3Jv0uyM8nHqmpbd1+8YJkjkvx6kkd397VVdc95zQcAsJnNc0vbw5Ps6O7LuvvbSd6Z5MRFyzwvyWndfW2SdPdVc5wPAGDTmme0HZrk8gXXd063LfTAJA+sqg9V1XlVddxSK6qqU6pqe1Vtv/rqq9dpXACAzWOzfRBh3yRHJHlskpOS/GFVHbR4oe4+vbu3dvfWQw45ZL4TAgBsgHlG2xVJDltwfct020I7k2zr7uu7+x+T/ENmEQcA8D1tntH2sSRHVNX9qmr/JE9Psm3RMu/NbCtbqurgzHaXXjbHGQEANqW5RVt335Dk1CTnJLkkyZndfVFVvbyqTpgWOyfJl6rq4iTnJvm17v7SvGYEANisqrs3eoZbZevWrb19+/aNHgMAYI+q6vzu3rqax262DyIAALAE0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwANEGADAA0QYAMADRBgAwgNtMtH340mvyxg9cutFjAADczBs/cGk+fOk1t3o9t4lo+/Cl1+TUMy7I0VsO3OhRAABu5ugtB+bUMy641eFW3b1GI22MO2/5gb7Xs16TB9zzzjnwDvtt9DgAALdw3Teuz46r/jWfe/3PfeGGf7n63qtZx4q3tNXMM6vqpdP1+1TVw1fzpGvp+htvyvfd9QDBBgBsWgfeYb98310PyD53Ouj7V7uOffdi2d9PclOSxyV5eZKvJnlPkh9a7ZOvhf32uV2+cf1NecETjsij7n/wRo4CALCkXYdy3fi1r3xhtevYm2PaHtHdv5jkm0nS3dcm2X+1T7xW9t/ndnn9ycesyb5iAIC1tivYXn/yMbnxq9dcudr17E20XV9V+yTpJKmqQzLb8rbhHnX/g/P6k4/JhTuv2+hRAABu5sKd1+X1Jx9zq/cIrviDCFX1jCRPS3Jskrcm+ZkkL+nud9+qCW6lu9/3wf3lz1+ykSMAAKxIVZ3f3VtX89gVHdNWVZXkg0nOT/L4JJXkKd2tlgAA5mBF0dbdXVVnd/dRST6zzjMBALDI3hzT9vGq2tBPigIAfK/am6/8eESSZ1TV55N8LbNdpN3dR6/LZAAAfMfeRNsT120KAAB2a8W7R7v780kOSvLk6eeg6TYAANbZ3pzG6gVJ/jTJPaeft1fV89drMAAAvmtvdo8+N7OzInwtSarqFUk+kuR16zEYAADftTefHq0kNy64fuN0GwAA62xvtrS9OclHq+ovputPSfKmNZ8IAIBbWHG0dferq+r9SX5kuuk53X3BukwFAMDNrDjaquqtSV7Q3R+frt+tqt7U3T+/btMBAJBk745pO7q7v7LrSndfm+SYNZ8IAIBb2Jtou11V3W3Xlaq6e/bumDgAAFZpb6Lrd5N8pKrePV3/2SS/s/YjAQCw2N58EOFtVbU9yeOmm36quy9en7EAAFhor3ZvTpEm1AAA5mxvTmP1s1V1l+nyS6rqz6vq2PUbDQCAXfbmgwi/2d1fraofSfKEJH+c5A3rMxYAAAvtTbTtOoXVTyQ5vbvPSrL/2o8EAMBiexNtV1TVHyR5WpKzq+qAvXw8AACrtDfR9dQk5yR54vQlu3dP8mvrMRQAADe3N1/58fUkf77g+heSfGE9hgIA4Obs3gQAGIBoAwAYgGgDABjAXp/wvapOTnJCZl8BUkne193vWOvBAAD4rr2OtiSP6e6n77pSVaclEW0AAOtoNdF2QFX9RJLLkxyW5A5rOxIAAIut5pi2/5TkbkmOT3JQklPXciAAAG5pNdG2NclHkhyV5MTpOgAA62g1u0dPSnJAkl9Jcl2Styb54FoOBQDAza0m2n4wyVe7++okqarr1nYkAAAWW020/WaSXnD9nDWaBQCAZex1tHX3Bxbd9H1rNAsAAMtYizMifGIN1gEAwG7c6mjr7vPWYhAAAJbnNFYAAANwGisAgAHc2tNYbYnTWAEArLtbcxqrJ03/dBorAIB1tpqv/Ph6krevwywAACxj1Z8eraqDquqgNZwFAIBlrOaYtl1+K8k+SX5pjWYBAGAZa/HlugAArDPRBgAwANEGADCAW3NM2+szOyMCAADrbNXR1t2XruUgAAAsz+5RAIABrOaE8Wcl+VSShyT5YpI/6O7/tdaDAQDwXavZPfoPSX6gu3+yqirJ65KINgCAdbSa3aOXJ6mq+oUkxyS559qOBADAYnuMtqp628Lr3f3qJE/LbBfp45LcWFV/uj7jAQCQrGz36FG7LlTV/+zuH+/ubyX50PQDAMA6W8nu0V5w+ZD1GgQAgOWtZEvbvarq2Uk+GV+mCwCwIVYSbS9L8rAkz0mypao+leSi6efi7n7P+o0HAECygmjr7tMXXq+qLZkd53Z0kqckEW0AAOtsr7+nrbt3JtmZ5K/XfhwAAJbiNFYAAAMQbQAAAxBtAAADEG0AAAMQbQAAAxBtAAADEG0AAAMQbQAAAxBtAAADEG0AAAMQbQAAAxBtAAADEG0AAAMQbQAAAxBtAAADEG0AAAOYa7RV1XFV9dmq2lFVL97Ncj9dVV1VW+c5HwDAZjW3aKuqfZKcluT4JEcmOamqjlxiubskeUGSj85rNgCAzW6eW9oenmRHd1/W3d9O8s4kJy6x3H9L8ook35zjbAAAm9o8o+3QJJcvuL5zuu07qurYJId191m7W1FVnVJV26tq+/XXX7/2kwIAbDKb5oMIVXW7JK9O8sI9Ldvdp3f31u7eut9++63/cAAAG2ye0XZFksMWXN8y3bbLXZI8JMn7q+pzSR6ZZJsPIwAAzDfaPpbkiKq6X1Xtn+TpSbbturO7r+vug7v78O4+PMl5SU7o7u1znBEAYFOaW7R19w1JTk1yTpJLkpzZ3RdV1cur6oR5zQEAMKLq7o2e4Va5+30f3F/+/CUbPQYAwB5V1fndvapDvzbNBxEAAFieaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABiAaAMAGIBoAwAYgGgDABjAXKOtqo6rqs9W1Y6qevES9/9qVV1cVRdW1d9V1X3nOR8AwGY1t2irqn2SnJbk+CRHJjmpqo5ctNgFSbZ299FJ/izJK+c1HwDAZjbPLW0PT7Kjuy/r7m8neWeSExcu0N3ndvfXp6vnJdkyx/kAADateUbboUkuX3B953Tbcp6b5K+XuqOqTqmq7VW1/frrr1/DEQEANqdN+UGEqnpmkq1JXrXU/d19endv7e6t++2333yHAwDYAPvO8bmuSHLYgutbpttupqqekOQ3kjymu781p9kAADa1eW5p+1iSI6rqflW1f5KnJ9m2cIGqOibJHyQ5obuvmuNsAACb2tyirbtvSHJqknOSXJLkzO6+qKpeXlUnTIu9Ksmdk7y7qj5RVduWWR0AwPeU6u6NnuFWuft9H9xf/vwlGz0GAMAeVdX53b11NY/dlB9EAADg5kQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAEQbAMAARBsAwABEGwDAAOYabVV1XFV9tqp2VNWLl7j/gKp613T/R6vq8JWu+8OXXpM3fuDSNZ0XAGCzmFu0VdU+SU5LcnySI5OcVFVHLlrsuUmu7e4HJHlNklesZN0fvvSanHrGBTl6y4FrOTIAwKYxzy1tD0+yo7sv6+5vJ3lnkhMXLXNikrdOl/8syeOrqna30m/feFNOPeOCvP7kY/Ko+x+85kMDAGwG+87xuQ5NcvmC6zuTPGK5Zbr7hqq6Lsk9klyzcKGqOiXJKUlyuzvcNRe+6uQvPPql11y5XoOzrg7OoveXoXj/xuW9G5v3b1w/sNoHzjPa1kx3n57k9CSpqu3dvXWDR2KVvH9j8/6Ny3s3Nu/fuKpq+2ofO8/do1ckOWzB9S3TbUsuU1X7JjkwyZfmMh0AwCY2z2j7WJIjqup+VbV/kqcn2bZomW1Jfm66/DNJ/r67e44zAgBsSnPbPTodo3ZqknOS7JPkTd19UVW9PMn27t6W5I+T/ElV7Ujy5czCbk9OX7ehmQfv39i8f+Py3o3N+zeuVb93ZUMWAMDm54wIAAADEG0AAAMYJtrW8xRYrL8VvH+/WlUXV9WFVfV3VXXfjZiTW9rTe7dguZ+uqq4qX0Owiazk/auqp05//i6qqjPmPSPLW8HfnfepqnOr6oLp788nbcSc3FJVvamqrqqqTy9zf1XVa6f39sKqOnZP6xwi2tbzFFisvxW+fxck2drdR2d2NoxXzndKlrLC9y5VdZckL0jy0flOyO6s5P2rqiOS/HqSR3f3Dyb55XnPydJW+OfvJUnO7O5jMvvw3u/Pd0p24y1JjtvN/ccnOWL6OSXJG/a0wiGiLet0CizmZo/vX3ef291fn66el9n3+LHxVvJnL0n+W2b/o/TNeQ7HHq3k/XtektO6+9ok6e6r5jwjy1vJ+9dJ7jpdPjCJswNtEt39wcy+CWM5JyZ5W8+cl+Sgqvr+3a1zlGhb6hRYhy63THffkGTXKbDYeCt5/xZ6bpK/XteJWKk9vnfTJv3DuvuseQ7Giqzkz94Dkzywqj5UVedV1e62DDBfK3n/XpbkmVW1M8nZSZ4/n9FYA3v738YxT2PFbVdVPTPJ1iSP2ehZ2LOqul2SVyd59gaPwurtm9numcdmtoX7g1V1VHd/ZSOHYsVOSvKW7v7dqvrhzL7r9CHdfdNGD8baG2VLm1NgjW0l71+q6glJfiPJCd39rTnNxu7t6b27S5KHJHl/VX0uySOTbPNhhE1jJX/2dibZ1t3Xd/c/JvmHzCKOjbeS9++5Sc5Mku7+SJLbZ3YyeTa/Ff23caFRos0psMa2x/evqo5J8geZBZtjajaP3b533X1ddx/c3Yd39+GZHY94Qnev+oTIrKmV/N353sy2sqWqDs5sd+llc5yR5a3k/funJI9Pkqp6cGbRdvVcp2S1tiV51vQp0kcmua67v7C7Bwyxe3QdT4HFHKzw/XtVkjsneff0+ZF/6u4TNmxokqz4vWOTWuH7d06SH6+qi5PcmOTXutteik1ghe/fC5P8YVX9SmYfSni2DRabQ1W9I7P/ITp4Oubwt5LslyTd/cbMjkF8UpIdSb6e5Dl7XKf3FgBg8xtl9ygAwPc00QYAMADRBgAwANEGADAA0QYAMADRBgOrqq6q311w/UVV9bI5z/D+XV+mW1VnV9VBt3J9j62qv1rL562qX66qO96auW6Nqvq9qvqx6fL9quqjVbWjqt41ff/WUo/5naq6vKr+ddHtB0yP2zGt5/BlHn9cVX12Wu7FeznvCXv7mAWP/c77ssz9/72qHreadcP3OtEGY/tWkp+avhR1r01nD1kz3f2kjTj90Qqe95eTbEi0VdU9kjxyOnl0krwiyWu6+wFJrs3sG+2X8r7MThi+2HOTXDs9/jXT+hY/5z5JTktyfJIjk5xUVUeudObu3tbd/9dKl99Lr0uyqiCE73WiDcZ2Q5LTk/zK4juq6vCq+vuqurCq/q6q7jPd/paqemNVfTTJK6frb5hOFn7ZtKXrTVV1SVW9ZcH63lBV26vqoqr6r0sNU1Wfq6qDq+oXquoT088/VtW50/0/XlUfqaqPV9W7q+rO0+3HVdVnqurjSX5qmXXfoareOc31F0nusMTz3qmqzqqqT1bVp6vqaVX1S0nuneTcBXMs+Vqm9fzXab5PVdWDptvvXFVvnm67sKp+enevZ5GfTvI30/KV5HFJ/my6761JnrLU6+3u85b5dvQTp8dlWs/jp/Uu9PAkO7r7su7+dpJ3To9b/Dt9f1X9j+l9+nRVPXy6/dlV9frp8l9W1bOmy/+hqv50Ja+9qvaZ/t369PR7+5XpdX0+yT2q6l5LvW5geaINxndakmdU1YGLbn9dkrd299FJ/jTJaxfctyXJo7r7V6frd0vyw5nF37bMtuD8YJKjquqh0zK/0d1bkxyd5DFVdfRyA3X3G7v7oUl+KLNzW7562hr4kiRP6O5jk2xP8qtVdfskf5jkyUkelmS5/5j/xyRf7+4HZ/bN4g9bYpnjklzZ3f+mux+S5G+6+7VJrkzyb7v7367gtVwzzfeGJC+abvvNzE4xc9T0+/z75V7PEjM9Osn50+V7JPlKd98wXd+Z5NBlXu9yDk1yeTL7xvwk103rXXKZFTzPHaf36j8ledMS95+S5KVV9aOZffv+81f42h+a5NDufkh3H5XkzQvu+3hmvxdgL4g2GFx3/0uStyX5pUV3/XCSM6bLf5LkRxbc9+7uvnHB9fdNp775VJIvdvenuvumJBclOXxa5qnTlrALMgu6lexu+x+ZnQf4fZmdTP7IJB+qqk9kdq7g+yZ5UJJ/7O7/b5rh7cus68d23dfdFya5cIllPpXk31XVK6rqR7v7umXWtbvX8ufTP8/Pd1/7EzKL40zPf+1uXs9i35/NfS7IdyTJtPv2rrXo2MDu/mKSlyY5N8kLu/vLWdlrvyzJ/1ZVr6uq45L8y4L7rsps6yewF4Y49yiwR7+X2daLN+9huV2+tuj6t6Z/3rTg8q7r+1bV/TLb6vRD3X3ttNv09rt7gqp6dmb/IT91101J/ra7T1q03ENXOPMedfc/VNWxmZ3P77er6u+6++WLnm9Pr2XX678xu/87csnXs4RvLFj/l5IcVFX7TlvJtiS5YjoGbdfWuG3d/dLdrO+KJIcl2VmzYxIPnNa71DK7bJluW8ricxkudW7Do6bn2BVae3zt0+/23yR5YpJfSPLUJD8/3X37zH4vwF6wpQ1uA6atH2fm5ge1fzjJ06fLz0jy/9yKp7hrZqF3XVV9X2YHuC+rqh6WWRg9c9pilyTnJXl0VT1gWuZOVfXAJJ9JcnhV3X9abrkQ+GCSk6fHPiSzXZuLn/feme1CfXuSVyU5drrrq0nusprXMvnbJL+44HnutpvXs9glSR6QJNOWxHOT/Mx0388l+cvuvrG7Hzr97C7Yktnu65+bLv9MZlsyF4fWx5IcUbNPqu6f2b8H25ZZ39Om+X8ks13AN9s6OR3ndnySY5K8aIrePb72aRfq7br7PZntSj12wd0PTPLpPbxOYBHRBrcdv5tk4adIn5/kOVV1YZL/I8kLVrvi7v5kZrsSP5PZLtcP7eEhpya5e2YH/3+iqv6ou69O8uwk75hm+kiSB3X3NzM7buqsaZflVcus8w1J7lxVlyR5eb67ZWqho5L8r2mX3W8l+e3p9tOT/E1VnbuK15JpPXebDqr/ZGbHxy35epZ47FlJHrvg+n/O7Fi+HZkdi/bHSz1hVb2yqnYmuWNV7azvfpXLH2d2IP+OzI4je/G0/L2r6uzkO8e6nZrknMyi8czuvmiZ1/bNqrogyRuz6JOsVXVAZscb/nx3X5nZMW1vSnLNCl77oUneP70Xb0/y69M698ssYrcvMw+wjLrl/6ABsJaq6v9N8pMb8XUou1NV70/you6eW0BV1f+e5Nju/s15PSfcVtjSBrD+XpjkPhs9xCaxb2ZbhYG9ZEsbAMAAbGkDABiAaAMAGIBoAwAYgGgDABiAaAMAGMD/DxZq067B+oBQAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.5555555555555556"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# duplicate matches\n",
    "area, _, _ = eval_pt_detect(np.array([(0,0)]), np.array([(0,0), (0,0)]), 10, beta=0.5, debug=True, show_plot=True)\n",
    "area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 98. Influence of the `radius_limit` parameter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.37447791253840795"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "area, _, _ = eval_pt_detect(ground_truth, predicted, radius_limit, beta=0.5, debug=False, show_plot=True)\n",
    "area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.30211120952259407"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "area, _, _ = eval_pt_detect(ground_truth, predicted, 50, beta=0.5, debug=False, show_plot=True)\n",
    "area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.11363636363636363"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "area, _, _ = eval_pt_detect(ground_truth, predicted, 1, beta=0.5, debug=False, show_plot=True)\n",
    "area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.17045454545454544"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "area, _, _ = eval_pt_detect(ground_truth, predicted, 10, beta=0.5, debug=False, show_plot=True)\n",
    "area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.32151015021584245"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "area, _, _ = eval_pt_detect(ground_truth, predicted, 100, beta=0.5, debug=False, show_plot=True)\n",
    "area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.5330659848993475"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "area, _, _ = eval_pt_detect(ground_truth, predicted, 1000, beta=0.5, debug=False, show_plot=True)\n",
    "area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 99. Fuzzy testing\n",
    "We also generated some random tests for extra safety."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    " np.random.seed(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "targets = np.float32(np.random.randint(0,10000,(5,2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.spatial import distance_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "375.9154751329672"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dmat = distance_matrix(targets, targets)\n",
    "dmat[np.diag_indices_from(dmat)] = np.inf\n",
    "radius_limit = np.min(dmat) / 3\n",
    "radius_limit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = np.float32(np.random.randint(0,10000,(100,2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.014580940066544888"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score, details, _ = eval_pt_detect(targets, predictions, radius_limit, beta=0.5, debug=False, show_plot=True)\n",
    "score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_predictions_classified(targets, details, radius_limit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = eval_pt_detect(predictions, targets, radius_limit, beta=0.5, debug=False, show_plot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluation against self should give maximal score\n",
    "area, _, _ = eval_pt_detect(predictions, predictions, radius_limit, beta=0.5, debug=False, show_plot=True)\n",
    "area"
   ]
  }
 ],
 "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}