solveig_master/plot.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Angle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "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>category</th>\n",
       "      <th>angle</th>\n",
       "      <th>count</th>\n",
       "      <th>pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>high angle</td>\n",
       "      <td>2</td>\n",
       "      <td>9.090909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>eye-level</td>\n",
       "      <td>13</td>\n",
       "      <td>59.090909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>low angle</td>\n",
       "      <td>7</td>\n",
       "      <td>31.818182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>high angle</td>\n",
       "      <td>5</td>\n",
       "      <td>27.777778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>eye-level</td>\n",
       "      <td>8</td>\n",
       "      <td>44.444444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>low angle</td>\n",
       "      <td>5</td>\n",
       "      <td>27.777778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>casual</td>\n",
       "      <td>high angle</td>\n",
       "      <td>6</td>\n",
       "      <td>60.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>casual</td>\n",
       "      <td>eye-level</td>\n",
       "      <td>3</td>\n",
       "      <td>30.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>casual</td>\n",
       "      <td>low angle</td>\n",
       "      <td>1</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       category       angle  count        pct\n",
       "0  white-collar  high angle      2   9.090909\n",
       "1  white-collar   eye-level     13  59.090909\n",
       "2  white-collar   low angle      7  31.818182\n",
       "3   blue-collar  high angle      5  27.777778\n",
       "4   blue-collar   eye-level      8  44.444444\n",
       "5   blue-collar   low angle      5  27.777778\n",
       "6        casual  high angle      6  60.000000\n",
       "7        casual   eye-level      3  30.000000\n",
       "8        casual   low angle      1  10.000000"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "df = pd.DataFrame({\n",
    "    'category': [\n",
    "        'white-collar',\n",
    "        'white-collar',\n",
    "        'white-collar',\n",
    "        'blue-collar',\n",
    "        'blue-collar',\n",
    "        'blue-collar',\n",
    "        'casual',\n",
    "        'casual',\n",
    "        'casual',\n",
    "    ],\n",
    "    'angle': [\n",
    "        'high angle',\n",
    "        'eye-level',\n",
    "        'low angle',\n",
    "        'high angle',\n",
    "        'eye-level',\n",
    "        'low angle',\n",
    "        'high angle',\n",
    "        'eye-level',\n",
    "        'low angle',\n",
    "    ],\n",
    "    'count': [\n",
    "        2,\n",
    "        13,\n",
    "        7,\n",
    "        5,\n",
    "        8,\n",
    "        5,\n",
    "        6,\n",
    "        3,\n",
    "        1,\n",
    "    ],\n",
    "    'pct': [\n",
    "            2/22*100,\n",
    "            13/22*100,\n",
    "            7/22*100,\n",
    "            5/18*100,\n",
    "            8/18*100,\n",
    "            5/18*100,\n",
    "            6/10*100,\n",
    "            3/10*100,\n",
    "            1/10*100,\n",
    "        ]\n",
    "})\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 725.625x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "g = sns.catplot(\n",
    "    data=df, kind=\"bar\",\n",
    "    x=\"category\", y=\"pct\", hue=\"angle\",\n",
    "    palette=\"Blues\", alpha=.6, height=6\n",
    ")\n",
    "g.despine(left=True)\n",
    "sns.set_style(\"ticks\",{'axes.grid' : True})\n",
    "g.set_axis_labels(\"\", \"percent of total images in category (%)\")\n",
    "g.legend.set_title(\"\")\n",
    "g.savefig('plots/angle.png')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Contact"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "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>category</th>\n",
       "      <th>contact</th>\n",
       "      <th>count</th>\n",
       "      <th>pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>demand structure</td>\n",
       "      <td>14</td>\n",
       "      <td>63.636364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>offer structure</td>\n",
       "      <td>8</td>\n",
       "      <td>36.363636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>demand structure</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>offer structure</td>\n",
       "      <td>18</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>casual</td>\n",
       "      <td>demand structure</td>\n",
       "      <td>1</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>casual</td>\n",
       "      <td>offer structure</td>\n",
       "      <td>9</td>\n",
       "      <td>90.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       category           contact  count         pct\n",
       "0  white-collar  demand structure     14   63.636364\n",
       "1  white-collar   offer structure      8   36.363636\n",
       "2   blue-collar  demand structure      0    0.000000\n",
       "3   blue-collar   offer structure     18  100.000000\n",
       "4        casual  demand structure      1   10.000000\n",
       "5        casual   offer structure      9   90.000000"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.DataFrame({\n",
    "    'category': [\n",
    "        'white-collar',\n",
    "        'white-collar',\n",
    "        'blue-collar',\n",
    "        'blue-collar',\n",
    "        'casual',\n",
    "        'casual',\n",
    "    ],\n",
    "    'contact': [\n",
    "        'demand structure',\n",
    "        'offer structure',\n",
    "        'demand structure',\n",
    "        'offer structure',\n",
    "        'demand structure',\n",
    "        'offer structure',\n",
    "    ],\n",
    "    'count': [\n",
    "        14,\n",
    "        8,\n",
    "        0,\n",
    "        18,\n",
    "        1,\n",
    "        9,\n",
    "    ],\n",
    "    'pct': [\n",
    "            14/22*100,\n",
    "            8/22*100,\n",
    "            0/18*100,\n",
    "            18/18*100,\n",
    "            1/10*100,\n",
    "            9/10*100,\n",
    "        ]\n",
    "})\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 773.75x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.catplot(\n",
    "    data=df, kind=\"bar\",\n",
    "    x=\"category\", y=\"pct\", hue=\"contact\",\n",
    "    palette=\"Blues\", alpha=.6, height=6\n",
    ")\n",
    "g.despine(left=True)\n",
    "sns.set_style(\"ticks\",{'axes.grid' : True})\n",
    "g.set_axis_labels(\"\", \"percent of total images in category (%)\")\n",
    "g.legend.set_title(\"\")\n",
    "g.savefig('plots/contact.png')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Point-of-view"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>category</th>\n",
       "      <th>poit_of_view</th>\n",
       "      <th>count</th>\n",
       "      <th>pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>frontal</td>\n",
       "      <td>16</td>\n",
       "      <td>72.727273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>oblique</td>\n",
       "      <td>2</td>\n",
       "      <td>9.090909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>frontal</td>\n",
       "      <td>4</td>\n",
       "      <td>22.222222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>oblique</td>\n",
       "      <td>14</td>\n",
       "      <td>77.777778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>casual</td>\n",
       "      <td>frontal</td>\n",
       "      <td>3</td>\n",
       "      <td>30.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>casual</td>\n",
       "      <td>oblique</td>\n",
       "      <td>7</td>\n",
       "      <td>70.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       category poit_of_view  count        pct\n",
       "0  white-collar      frontal     16  72.727273\n",
       "1  white-collar      oblique      2   9.090909\n",
       "2   blue-collar      frontal      4  22.222222\n",
       "3   blue-collar      oblique     14  77.777778\n",
       "4        casual      frontal      3  30.000000\n",
       "5        casual      oblique      7  70.000000"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame({\n",
    "    'category': [\n",
    "        'white-collar',\n",
    "        'white-collar',\n",
    "        'blue-collar',\n",
    "        'blue-collar',\n",
    "        'casual',\n",
    "        'casual',\n",
    "    ],\n",
    "    'poit_of_view': [\n",
    "        'frontal',\n",
    "        'oblique',\n",
    "        'frontal',\n",
    "        'oblique',\n",
    "        'frontal',\n",
    "        'oblique',\n",
    "    ],\n",
    "    'count': [\n",
    "        16,\n",
    "        2,\n",
    "        4,\n",
    "        14,\n",
    "        3,\n",
    "        7,\n",
    "    ],\n",
    "    'pct': [\n",
    "            16/22*100,\n",
    "            2/22*100,\n",
    "            4/18*100,\n",
    "            14/18*100,\n",
    "            3/10*100,\n",
    "            7/10*100,\n",
    "        ]\n",
    "})\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 704.847x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.catplot(\n",
    "    data=df, kind=\"bar\",\n",
    "    x=\"category\", y=\"pct\", hue=\"poit_of_view\",\n",
    "    palette=\"Blues\", alpha=.6, height=6\n",
    ")\n",
    "g.despine(left=True)\n",
    "sns.set_style(\"ticks\",{'axes.grid' : True})\n",
    "g.set_axis_labels(\"\", \"percent of total images in category (%)\")\n",
    "g.legend.set_title(\"\")\n",
    "g.savefig('plots/point-of-view.png')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "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>category</th>\n",
       "      <th>distance</th>\n",
       "      <th>count</th>\n",
       "      <th>pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>long shot</td>\n",
       "      <td>6</td>\n",
       "      <td>27.272727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>medium shot</td>\n",
       "      <td>16</td>\n",
       "      <td>72.727273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>white-collar</td>\n",
       "      <td>close-up</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>long shot</td>\n",
       "      <td>15</td>\n",
       "      <td>83.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>medium shot</td>\n",
       "      <td>3</td>\n",
       "      <td>16.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>close-up</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>casual</td>\n",
       "      <td>long shot</td>\n",
       "      <td>3</td>\n",
       "      <td>30.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>casual</td>\n",
       "      <td>medium shot</td>\n",
       "      <td>5</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>casual</td>\n",
       "      <td>close-up</td>\n",
       "      <td>2</td>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       category     distance  count        pct\n",
       "0  white-collar    long shot      6  27.272727\n",
       "1  white-collar  medium shot     16  72.727273\n",
       "2  white-collar     close-up      0   0.000000\n",
       "3   blue-collar    long shot     15  83.333333\n",
       "4   blue-collar  medium shot      3  16.666667\n",
       "5   blue-collar     close-up      0   0.000000\n",
       "6        casual    long shot      3  30.000000\n",
       "7        casual  medium shot      5  50.000000\n",
       "8        casual     close-up      2  20.000000"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.DataFrame({\n",
    "    'category': [\n",
    "        'white-collar',\n",
    "        'white-collar',\n",
    "        'white-collar',\n",
    "        'blue-collar',\n",
    "        'blue-collar',\n",
    "        'blue-collar',\n",
    "        'casual',\n",
    "        'casual',\n",
    "        'casual',\n",
    "    ],\n",
    "    'distance': [\n",
    "        'long shot',\n",
    "        'medium shot',\n",
    "        'close-up',\n",
    "        'long shot',\n",
    "        'medium shot',\n",
    "        'close-up',\n",
    "        'long shot',\n",
    "        'medium shot',\n",
    "        'close-up',\n",
    "    ],\n",
    "    'count': [\n",
    "        6,\n",
    "        16,\n",
    "        0,\n",
    "        15,\n",
    "        3,\n",
    "        0,\n",
    "        3,\n",
    "        5,\n",
    "        2,\n",
    "    ],\n",
    "    'pct': [\n",
    "            6/22*100,\n",
    "            16/22*100,\n",
    "            0/22*100,\n",
    "            15/18*100,\n",
    "            3/18*100,\n",
    "            0/18*100,\n",
    "            3/10*100,\n",
    "            5/10*100,\n",
    "            2/10*100,\n",
    "        ]\n",
    "})\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 742.25x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.catplot(\n",
    "    data=df, kind=\"bar\",\n",
    "    x=\"category\", y=\"pct\", hue=\"distance\",\n",
    "    palette=\"Blues\", alpha=.6, height=6\n",
    ")\n",
    "g.despine(left=True)\n",
    "sns.set_style(\"ticks\",{'axes.grid' : True})\n",
    "g.set_axis_labels(\"\", \"percent of total images in category (%)\")\n",
    "g.legend.set_title(\"\")\n",
    "g.savefig('plots/distance.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.8.16"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}