0_Exploratory_Data_Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CelebA - Exploratory Data Analysis for GANs\n",
    "\n",
    "In the following, we perform an EDA on the CelebA dataset.\n",
    "\n",
    "Since we are not interested in performing classification, we don't necessarily care about looking for discriminative features, but rather we simply take a look at the number of occurrences of each annotation."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setting up the environment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importing modules"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading the dataset (annotations only)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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>5_o_Clock_Shadow</th>\n",
       "      <th>Arched_Eyebrows</th>\n",
       "      <th>Attractive</th>\n",
       "      <th>Bags_Under_Eyes</th>\n",
       "      <th>Bald</th>\n",
       "      <th>Bangs</th>\n",
       "      <th>Big_Lips</th>\n",
       "      <th>Big_Nose</th>\n",
       "      <th>Black_Hair</th>\n",
       "      <th>Blond_Hair</th>\n",
       "      <th>...</th>\n",
       "      <th>Sideburns</th>\n",
       "      <th>Smiling</th>\n",
       "      <th>Straight_Hair</th>\n",
       "      <th>Wavy_Hair</th>\n",
       "      <th>Wearing_Earrings</th>\n",
       "      <th>Wearing_Hat</th>\n",
       "      <th>Wearing_Lipstick</th>\n",
       "      <th>Wearing_Necklace</th>\n",
       "      <th>Wearing_Necktie</th>\n",
       "      <th>Young</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 40 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   5_o_Clock_Shadow  Arched_Eyebrows  Attractive  Bags_Under_Eyes  Bald  \\\n",
       "0                 0                1           1                0     0   \n",
       "1                 0                0           0                1     0   \n",
       "2                 0                0           0                0     0   \n",
       "3                 0                0           1                0     0   \n",
       "4                 0                1           1                0     0   \n",
       "\n",
       "   Bangs  Big_Lips  Big_Nose  Black_Hair  Blond_Hair  ...  Sideburns  Smiling  \\\n",
       "0      0         0         0           0           0  ...          0        1   \n",
       "1      0         0         1           0           0  ...          0        1   \n",
       "2      0         1         0           0           0  ...          0        0   \n",
       "3      0         0         0           0           0  ...          0        0   \n",
       "4      0         1         0           0           0  ...          0        0   \n",
       "\n",
       "   Straight_Hair  Wavy_Hair  Wearing_Earrings  Wearing_Hat  Wearing_Lipstick  \\\n",
       "0              1          0                 1            0                 1   \n",
       "1              0          0                 0            0                 0   \n",
       "2              0          1                 0            0                 0   \n",
       "3              1          0                 1            0                 1   \n",
       "4              0          0                 0            0                 1   \n",
       "\n",
       "   Wearing_Necklace  Wearing_Necktie  Young  \n",
       "0                 0                0      1  \n",
       "1                 0                0      1  \n",
       "2                 0                0      1  \n",
       "3                 1                0      1  \n",
       "4                 0                0      1  \n",
       "\n",
       "[5 rows x 40 columns]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"data/list_attr_celeba.csv\")\n",
    "data = data.drop(\"Image_Name\", axis=1)  # drop the first column\n",
    "data.head() # display the first n=5 rows\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of images: 202599\n"
     ]
    }
   ],
   "source": [
    "tot_images = len(data.index)\n",
    "print(\"Total number of images:\", tot_images)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Counting occurrences of each annotation"
   ]
  },
  {
   "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>5_o_Clock_Shadow</th>\n",
       "      <th>Arched_Eyebrows</th>\n",
       "      <th>Attractive</th>\n",
       "      <th>Bags_Under_Eyes</th>\n",
       "      <th>Bald</th>\n",
       "      <th>Bangs</th>\n",
       "      <th>Big_Lips</th>\n",
       "      <th>Big_Nose</th>\n",
       "      <th>Black_Hair</th>\n",
       "      <th>Blond_Hair</th>\n",
       "      <th>...</th>\n",
       "      <th>Sideburns</th>\n",
       "      <th>Smiling</th>\n",
       "      <th>Straight_Hair</th>\n",
       "      <th>Wavy_Hair</th>\n",
       "      <th>Wearing_Earrings</th>\n",
       "      <th>Wearing_Hat</th>\n",
       "      <th>Wearing_Lipstick</th>\n",
       "      <th>Wearing_Necklace</th>\n",
       "      <th>Wearing_Necktie</th>\n",
       "      <th>Young</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>180083</td>\n",
       "      <td>148509</td>\n",
       "      <td>98766</td>\n",
       "      <td>161153</td>\n",
       "      <td>198052</td>\n",
       "      <td>171890</td>\n",
       "      <td>153814</td>\n",
       "      <td>155083</td>\n",
       "      <td>154127</td>\n",
       "      <td>172616</td>\n",
       "      <td>...</td>\n",
       "      <td>191150</td>\n",
       "      <td>104930</td>\n",
       "      <td>160377</td>\n",
       "      <td>137855</td>\n",
       "      <td>164323</td>\n",
       "      <td>192781</td>\n",
       "      <td>106884</td>\n",
       "      <td>177686</td>\n",
       "      <td>187867</td>\n",
       "      <td>45865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>22516</td>\n",
       "      <td>54090</td>\n",
       "      <td>103833</td>\n",
       "      <td>41446</td>\n",
       "      <td>4547</td>\n",
       "      <td>30709</td>\n",
       "      <td>48785</td>\n",
       "      <td>47516</td>\n",
       "      <td>48472</td>\n",
       "      <td>29983</td>\n",
       "      <td>...</td>\n",
       "      <td>11449</td>\n",
       "      <td>97669</td>\n",
       "      <td>42222</td>\n",
       "      <td>64744</td>\n",
       "      <td>38276</td>\n",
       "      <td>9818</td>\n",
       "      <td>95715</td>\n",
       "      <td>24913</td>\n",
       "      <td>14732</td>\n",
       "      <td>156734</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 40 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   5_o_Clock_Shadow  Arched_Eyebrows  Attractive  Bags_Under_Eyes    Bald  \\\n",
       "0            180083           148509       98766           161153  198052   \n",
       "1             22516            54090      103833            41446    4547   \n",
       "\n",
       "    Bangs  Big_Lips  Big_Nose  Black_Hair  Blond_Hair  ...  Sideburns  \\\n",
       "0  171890    153814    155083      154127      172616  ...     191150   \n",
       "1   30709     48785     47516       48472       29983  ...      11449   \n",
       "\n",
       "   Smiling  Straight_Hair  Wavy_Hair  Wearing_Earrings  Wearing_Hat  \\\n",
       "0   104930         160377     137855            164323       192781   \n",
       "1    97669          42222      64744             38276         9818   \n",
       "\n",
       "   Wearing_Lipstick  Wearing_Necklace  Wearing_Necktie   Young  \n",
       "0            106884            177686           187867   45865  \n",
       "1             95715             24913            14732  156734  \n",
       "\n",
       "[2 rows x 40 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "counts = data.apply(pd.value_counts)\n",
    "display(counts)\n",
    "# 1 means the image is labeled as having that attribute\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now take the second row only, since it gives us the number of times an image has been classified as having that attribute, and we plot the occurrences in descending order"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "labeled_as_1 = counts.iloc[lambda x: x.index == 1]  # take only the row labeled as 1\n",
    "descending_counts_pairs = sorted(\n",
    "    zip(labeled_as_1.values.flatten(), labeled_as_1.columns), reverse=True\n",
    ")   # sort annotations by descending frequency\n",
    "x = [x for _, x in descending_counts_pairs] # unzip annotations\n",
    "y = [y for y, _ in descending_counts_pairs] # unzip counts\n",
    "\n",
    "plt.figure(figsize=(20, 8))\n",
    "plt.xticks(rotation=80)\n",
    "sns.set_style(\"whitegrid\")\n",
    "sns.barplot(x=x, y=y, palette=\"rocket\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It could also be useful to take a look at the occurrences in terms of percentage over the total number of images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Value for `orient` not understood: auto",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32mg:\\Git Repos\\mlinapp\\0_Exploratory_Data_Analysis.ipynb Cell 13'\u001b[0m in \u001b[0;36m<cell line: 4>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      <a href='vscode-notebook-cell:/g%3A/Git%20Repos/mlinapp/0_Exploratory_Data_Analysis.ipynb#ch0000012?line=1'>2</a>\u001b[0m plt\u001b[39m.\u001b[39mfigure(figsize\u001b[39m=\u001b[39m(\u001b[39m20\u001b[39m, \u001b[39m8\u001b[39m))\n\u001b[0;32m      <a href='vscode-notebook-cell:/g%3A/Git%20Repos/mlinapp/0_Exploratory_Data_Analysis.ipynb#ch0000012?line=2'>3</a>\u001b[0m plt\u001b[39m.\u001b[39mxticks(rotation\u001b[39m=\u001b[39m\u001b[39m80\u001b[39m)\n\u001b[1;32m----> <a href='vscode-notebook-cell:/g%3A/Git%20Repos/mlinapp/0_Exploratory_Data_Analysis.ipynb#ch0000012?line=3'>4</a>\u001b[0m sns\u001b[39m.\u001b[39;49mbarplot(x\u001b[39m=\u001b[39;49mx, y\u001b[39m=\u001b[39;49my_percentage, palette\u001b[39m=\u001b[39;49m\u001b[39m\"\u001b[39;49m\u001b[39mviridis\u001b[39;49m\u001b[39m\"\u001b[39;49m, orient\u001b[39m=\u001b[39;49m\u001b[39m\"\u001b[39;49m\u001b[39mauto\u001b[39;49m\u001b[39m\"\u001b[39;49m)\u001b[39m.\u001b[39mget_figure()\u001b[39m.\u001b[39msavefig(\u001b[39m\"\u001b[39m\u001b[39mresults/dataset/att_distribution.pdf\u001b[39m\u001b[39m\"\u001b[39m)\n",
      "File \u001b[1;32mg:\\PythonVenvs\\pytorch\\lib\\site-packages\\seaborn\\_decorators.py:46\u001b[0m, in \u001b[0;36m_deprecate_positional_args.<locals>.inner_f\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     36\u001b[0m     warnings\u001b[39m.\u001b[39mwarn(\n\u001b[0;32m     37\u001b[0m         \u001b[39m\"\u001b[39m\u001b[39mPass the following variable\u001b[39m\u001b[39m{}\u001b[39;00m\u001b[39m as \u001b[39m\u001b[39m{}\u001b[39;00m\u001b[39mkeyword arg\u001b[39m\u001b[39m{}\u001b[39;00m\u001b[39m: \u001b[39m\u001b[39m{}\u001b[39;00m\u001b[39m. \u001b[39m\u001b[39m\"\u001b[39m\n\u001b[0;32m     38\u001b[0m         \u001b[39m\"\u001b[39m\u001b[39mFrom version 0.12, the only valid positional argument \u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m     43\u001b[0m         \u001b[39mFutureWarning\u001b[39;00m\n\u001b[0;32m     44\u001b[0m     )\n\u001b[0;32m     45\u001b[0m kwargs\u001b[39m.\u001b[39mupdate({k: arg \u001b[39mfor\u001b[39;00m k, arg \u001b[39min\u001b[39;00m \u001b[39mzip\u001b[39m(sig\u001b[39m.\u001b[39mparameters, args)})\n\u001b[1;32m---> 46\u001b[0m \u001b[39mreturn\u001b[39;00m f(\u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n",
      "File \u001b[1;32mg:\\PythonVenvs\\pytorch\\lib\\site-packages\\seaborn\\categorical.py:3182\u001b[0m, in \u001b[0;36mbarplot\u001b[1;34m(x, y, hue, data, order, hue_order, estimator, ci, n_boot, units, seed, orient, color, palette, saturation, errcolor, errwidth, capsize, dodge, ax, **kwargs)\u001b[0m\n\u001b[0;32m   3169\u001b[0m \u001b[39m@_deprecate_positional_args\u001b[39m\n\u001b[0;32m   3170\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mbarplot\u001b[39m(\n\u001b[0;32m   3171\u001b[0m     \u001b[39m*\u001b[39m,\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m   3179\u001b[0m     \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs,\n\u001b[0;32m   3180\u001b[0m ):\n\u001b[1;32m-> 3182\u001b[0m     plotter \u001b[39m=\u001b[39m _BarPlotter(x, y, hue, data, order, hue_order,\n\u001b[0;32m   3183\u001b[0m                           estimator, ci, n_boot, units, seed,\n\u001b[0;32m   3184\u001b[0m                           orient, color, palette, saturation,\n\u001b[0;32m   3185\u001b[0m                           errcolor, errwidth, capsize, dodge)\n\u001b[0;32m   3187\u001b[0m     \u001b[39mif\u001b[39;00m ax \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[0;32m   3188\u001b[0m         ax \u001b[39m=\u001b[39m plt\u001b[39m.\u001b[39mgca()\n",
      "File \u001b[1;32mg:\\PythonVenvs\\pytorch\\lib\\site-packages\\seaborn\\categorical.py:1584\u001b[0m, in \u001b[0;36m_BarPlotter.__init__\u001b[1;34m(self, x, y, hue, data, order, hue_order, estimator, ci, n_boot, units, seed, orient, color, palette, saturation, errcolor, errwidth, capsize, dodge)\u001b[0m\n\u001b[0;32m   1579\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__init__\u001b[39m(\u001b[39mself\u001b[39m, x, y, hue, data, order, hue_order,\n\u001b[0;32m   1580\u001b[0m              estimator, ci, n_boot, units, seed,\n\u001b[0;32m   1581\u001b[0m              orient, color, palette, saturation, errcolor,\n\u001b[0;32m   1582\u001b[0m              errwidth, capsize, dodge):\n\u001b[0;32m   1583\u001b[0m     \u001b[39m\"\"\"Initialize the plotter.\"\"\"\u001b[39;00m\n\u001b[1;32m-> 1584\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mestablish_variables(x, y, hue, data, orient,\n\u001b[0;32m   1585\u001b[0m                              order, hue_order, units)\n\u001b[0;32m   1586\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mestablish_colors(color, palette, saturation)\n\u001b[0;32m   1587\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mestimate_statistic(estimator, ci, n_boot, seed)\n",
      "File \u001b[1;32mg:\\PythonVenvs\\pytorch\\lib\\site-packages\\seaborn\\categorical.py:156\u001b[0m, in \u001b[0;36m_CategoricalPlotter.establish_variables\u001b[1;34m(self, x, y, hue, data, orient, order, hue_order, units)\u001b[0m\n\u001b[0;32m    153\u001b[0m         \u001b[39mraise\u001b[39;00m \u001b[39mValueError\u001b[39;00m(err)\n\u001b[0;32m    155\u001b[0m \u001b[39m# Figure out the plotting orientation\u001b[39;00m\n\u001b[1;32m--> 156\u001b[0m orient \u001b[39m=\u001b[39m infer_orient(\n\u001b[0;32m    157\u001b[0m     x, y, orient, require_numeric\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mrequire_numeric\n\u001b[0;32m    158\u001b[0m )\n\u001b[0;32m    160\u001b[0m \u001b[39m# Option 2a:\u001b[39;00m\n\u001b[0;32m    161\u001b[0m \u001b[39m# We are plotting a single set of data\u001b[39;00m\n\u001b[0;32m    162\u001b[0m \u001b[39m# ------------------------------------\u001b[39;00m\n\u001b[0;32m    163\u001b[0m \u001b[39mif\u001b[39;00m x \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mor\u001b[39;00m y \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[0;32m    164\u001b[0m \n\u001b[0;32m    165\u001b[0m     \u001b[39m# Determine where the data are\u001b[39;00m\n",
      "File \u001b[1;32mg:\\PythonVenvs\\pytorch\\lib\\site-packages\\seaborn\\_core.py:1342\u001b[0m, in \u001b[0;36minfer_orient\u001b[1;34m(x, y, orient, require_numeric)\u001b[0m\n\u001b[0;32m   1339\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39m\"\u001b[39m\u001b[39mh\u001b[39m\u001b[39m\"\u001b[39m\n\u001b[0;32m   1341\u001b[0m \u001b[39melif\u001b[39;00m orient \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m-> 1342\u001b[0m     \u001b[39mraise\u001b[39;00m \u001b[39mValueError\u001b[39;00m(\u001b[39mf\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mValue for `orient` not understood: \u001b[39m\u001b[39m{\u001b[39;00morient\u001b[39m}\u001b[39;00m\u001b[39m\"\u001b[39m)\n\u001b[0;32m   1344\u001b[0m \u001b[39melif\u001b[39;00m x_type \u001b[39m!=\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mnumeric\u001b[39m\u001b[39m\"\u001b[39m \u001b[39mand\u001b[39;00m y_type \u001b[39m==\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mnumeric\u001b[39m\u001b[39m\"\u001b[39m:\n\u001b[0;32m   1345\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39m\"\u001b[39m\u001b[39mv\u001b[39m\u001b[39m\"\u001b[39m\n",
      "\u001b[1;31mValueError\u001b[0m: Value for `orient` not understood: auto"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_percentage = [(count / tot_images) * 100 for count in y]\n",
    "plt.figure(figsize=(20, 8))\n",
    "plt.xticks(rotation=80)\n",
    "sns.barplot(y=x, x=y_percentage, palette=\"viridis\").get_figure().savefig(\"results/dataset/att_distribution.pdf\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can immediately see that the dataset tends to have mostly young people without beard.\n",
    "\n",
    "Only a few attributes appear to be balanced (around 50%), like being 'Attractive' or 'Smiling'.\n",
    "\n",
    "We also notice that the dataset is a bit skewed towards females' images, as the 'Male' attribute only reaches around 40%.\n",
    "\n",
    "Finally, both people wearing 'Eyeglasses' and having 'Mustache' appear in less than 10% of the images.\n",
    "\n",
    "As such, either of those could very well be chosen as our target domain, while the rest of the dataset (sub-sampled, if need be) could be considered as our source domain."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualizing correlation between attributes\n",
    "\n",
    "Finally, even though this isn't really necessary for the sake of our application, we take a look at the correlation matrix between the 40 annotations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#_, ax = plt.subplots(figsize=(20, 20))\n",
    "#sns.set_context(\"paper\")\n",
    "sns.heatmap(data.corr(), linewidths=0.5, square=True, xticklabels=False, yticklabels=False,  cmap=\"viridis\", vmin=-1).get_figure().savefig(\"results/dataset/correlation.pdf\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Only a few pairs are positively correlated, like ('Heavy_Makeup', 'Wearing_Lipstick') and ('Smiling','High_Cheekbones'), while most annotations aren't.\n",
    "\n",
    "Also, it's important to notice that 'No_Beard' and 'Mustache' (or similar) are negatively correlated. As such, if our intention is to set 'Mustaches' to 1, we should also take care to change 'No_Beard' to 0."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.10 ('pytorch')",
   "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": "3.9.10"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "06eabec719de4ba5303677b3566faec9bd202a3f9529d5de5971ba8dbc8516c5"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}