notebooks/0.1_EDA.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "21c6797f-dd1c-4bda-90d4-659ab88264e3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "eb8795e8-d000-4ac6-ab7a-3bddb68055d6",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import sys\n",
    "\n",
    "from pathlib import Path\n",
    "import pandas as pd\n",
    "from IPython.display import display\n",
    "import matplotlib.pyplot as plt\n",
    "from functional import seq\n",
    "import librosa\n",
    "import seaborn as sns\n",
    "\n",
    "sns.set_context(\"talk\")\n",
    "\n",
    "sys.path.append(\"..\")\n",
    "from GSSP_utils.path_conf import speech_data_session_dir, loc_data_dir\n",
    "from GSSP_utils.conf import color_palette\n",
    "\n",
    "pd.options.display.max_columns = None\n",
    "pd.options.display.max_rows = 50\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a79fe8cf",
   "metadata": {},
   "source": [
    "## Downloading all the data"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "9ccb963b",
   "metadata": {},
   "source": [
    "- project: https://osf.io/8wrmn\n",
    "- download and unzip the zips in the  `speech_web_app/backup` data directory\n",
    "- make sure that you've set the the correct path to the data directory in the `data_dir` variable of [path_conf.py](../GSSP_utils/path_conf.py)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c317f314",
   "metadata": {},
   "source": [
    "The folder structure:"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "74364319",
   "metadata": {},
   "source": [
    "```txt\n",
    "web_app_data\n",
    "|-- backup\n",
    "|   |-- 2020-11-26__12:40:00+01:00__4949e55b-daff-4255-9741-3206033a736b\n",
    "|   |   |-- PiSCES\n",
    "|   |   |-- Radboud\n",
    "|   |   |-- marloes\n",
    "|   |   `-- metadata.json\n",
    "|   |-- 2020-11-26__14:29:00+01:00__8fb4d923-48d0-4704-9c32-235c61227f76\n",
    "|   |   |-- PiSCES\n",
    "|   |   |-- Radboud\n",
    "|   |   |-- marloes\n",
    "|   |   `-- metadata.json\n",
    "|   |-- 2020-11-26__16:20:00+01:00__b8b6a8ee-2b45-41e1-9d93-3eacbc25fc34\n",
    "|   |   |-- PiSCES\n",
    "|   |   |-- Radboud\n",
    "|   |   |-- marloes\n",
    "|   |   `-- metadata.json\n",
    "|   |-- ...\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c834491c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total 0\n",
      "drwxr-xr-x 5 jonas jonas 71 Aug  3  2022 2020-11-26__12_40_00+01_00__4949e55b-daff-4255-9741-3206033a736b\n",
      "drwxr-xr-x 5 jonas jonas 71 Jun 27  2022 2020-11-26__14_29_00+01_00__8fb4d923-48d0-4704-9c32-235c61227f76\n",
      "drwxr-xr-x 5 jonas jonas 71 Jun 27  2022 2020-11-26__16_20_00+01_00__b8b6a8ee-2b45-41e1-9d93-3eacbc25fc34\n",
      "drwxr-xr-x 5 jonas jonas 71 Jun 27  2022 2020-11-26__18_50_00+01_00__33583a00-841f-4bef-b711-a8fe58c14476\n",
      "drwxr-xr-x 5 jonas jonas 71 Jun 27  2022 2020-11-27__08_06_00+01_00__f46c79e9-7c35-4bd4-b384-078aeae84af7\n",
      "drwxr-xr-x 5 jonas jonas 71 Jun 27  2022 2020-11-27__08_26_00+01_00__9223720f-1a2a-4811-84a1-a421646f63a0\n",
      "drwxr-xr-x 5 jonas jonas 71 Jun 27  2022 2020-11-27__12_06_00+01_00__716dc108-5588-49ed-bf3c-04fb87212e2b\n",
      "drwxr-xr-x 5 jonas jonas 71 Jun 27  2022 2020-11-27__12_24_00+01_00__99d276a0-83fe-4d47-949e-481dc67b2688\n",
      "drwxr-xr-x 5 jonas jonas 71 Jun 27  2022 2020-11-27__16_37_00+01_00__b55d08fb-30f3-499f-8cf4-f0f44321f1a9\n",
      "ls: write error: Broken pipe\n"
     ]
    }
   ],
   "source": [
    "# look at the data\n",
    "!ls -lh {speech_data_session_dir} | head -n 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "ed9ebcc6",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Radboud    89\n",
      "marloes    89\n",
      "PiSCES     89\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# This snippet summarizes the dir-names of the various folders\n",
    "# within `speech_data_session_dir`\n",
    "dir_list = []\n",
    "for folder in speech_data_session_dir.iterdir():\n",
    "    if not folder.is_dir():\n",
    "        continue\n",
    "    dir_list.extend(\n",
    "        seq(folder.iterdir())\n",
    "        .filter(lambda x: x.is_dir())\n",
    "        .map(lambda x: x.name)\n",
    "        .to_list()\n",
    "    )\n",
    "print(pd.Series(dir_list).value_counts())\n",
    "del dir_list, folder\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7998e716",
   "metadata": {},
   "source": [
    "## Parsing the `metadata.json` file"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bcf11db9",
   "metadata": {},
   "source": [
    "The `metadata.json` file resides within each user ID folder and withholds the users's \n",
    "metadata such as:\n",
    "* gender\n",
    "* age\n",
    "* education\n",
    "* used mic\n",
    "* etc ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "9cdb2143",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 89 entries, 0 to 88\n",
      "Data columns (total 10 columns):\n",
      " #   Column          Non-Null Count  Dtype                          \n",
      "---  ------          --------------  -----                          \n",
      " 0   age             89 non-null     int64                          \n",
      " 1   device          75 non-null     category                       \n",
      " 2   education       89 non-null     category                       \n",
      " 3   prolific_token  54 non-null     object                         \n",
      " 4   sex             89 non-null     category                       \n",
      " 5   timestamp       89 non-null     datetime64[ns, Europe/Brussels]\n",
      " 6   ID              89 non-null     object                         \n",
      " 7   Marloes         89 non-null     int64                          \n",
      " 8   PiSCES          89 non-null     int64                          \n",
      " 9   Radboud         89 non-null     int64                          \n",
      "dtypes: category(3), datetime64[ns, Europe/Brussels](1), int64(4), object(2)\n",
      "memory usage: 5.7+ KB\n"
     ]
    }
   ],
   "source": [
    "metadata_dict_list = []\n",
    "\n",
    "# creating alldata.csv\n",
    "for folder in speech_data_session_dir.iterdir():\n",
    "    if not folder.is_dir():\n",
    "        continue\n",
    "\n",
    "    metadata = folder.joinpath(\"metadata.json\")\n",
    "    if not metadata.is_file():\n",
    "        continue\n",
    "\n",
    "    with open(metadata, \"r\") as f:\n",
    "        metadata_dict = json.load(f)\n",
    "    metadata_dict[\"ID\"] = folder.name.split(\"__\")[-1]\n",
    "\n",
    "    metadata_dict[\"Marloes\"] = len(list(folder.joinpath(\"marloes\").glob(\"*.wav\")))\n",
    "    metadata_dict[\"PiSCES\"] = len(list(folder.joinpath(\"PiSCES\").glob(\"*.wav\")))\n",
    "    metadata_dict[\"Radboud\"] = len(list(folder.joinpath(\"Radboud\").glob(\"*.wav\")))\n",
    "\n",
    "    metadata_dict_list.append(metadata_dict)\n",
    "\n",
    "df_metadata = pd.DataFrame(metadata_dict_list)\n",
    "\n",
    "# convert the columns to the correct data format\n",
    "df_metadata[\"age\"] = df_metadata[\"age\"].astype(int)\n",
    "df_metadata[[\"sex\", \"device\", \"education\"]] = df_metadata[\n",
    "    [\"sex\", \"device\", \"education\"]\n",
    "].astype(\"category\")\n",
    "df_metadata[\"timestamp\"] = pd.to_datetime(\n",
    "    df_metadata.timestamp, utc=True\n",
    ").dt.tz_convert(\"Europe/Brussels\")\n",
    "\n",
    "del folder, metadata_dict, metadata_dict_list, f, metadata\n",
    "df_metadata.info()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "2bf24a63",
   "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>age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>89.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>27.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>19.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>23.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>26.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>30.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>53.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         age\n",
       "count  89.00\n",
       "mean   27.53\n",
       "std     6.63\n",
       "min    19.00\n",
       "25%    23.00\n",
       "50%    26.00\n",
       "75%    30.00\n",
       "max    53.00"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_metadata[\"age\"].describe().to_frame().round(2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "488642a5",
   "metadata": {},
   "source": [
    "## Participant metadata analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "40b5b682",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([<AxesSubplot:ylabel='Frequency'>, <AxesSubplot:ylabel='Frequency'>,\n",
       "       <AxesSubplot:ylabel='Frequency'>], dtype=object)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_metadata[[\"Marloes\", \"PiSCES\", \"Radboud\"]].plot(\n",
    "    subplots=True,\n",
    "    kind=\"hist\",\n",
    "    figsize=(8, 5),\n",
    "    sharex=False,\n",
    "    color=color_palette,\n",
    "    title=\"Number of recordings per dataset\",\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "7fc49e61",
   "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>age</th>\n",
       "      <th>Marloes</th>\n",
       "      <th>PiSCES</th>\n",
       "      <th>Radboud</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>89.000000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>89.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>27.528090</td>\n",
       "      <td>6.932584</td>\n",
       "      <td>14.831461</td>\n",
       "      <td>14.865169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6.629976</td>\n",
       "      <td>0.330212</td>\n",
       "      <td>1.298915</td>\n",
       "      <td>0.906931</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>19.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>23.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>15.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>26.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>15.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>30.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>15.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>53.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>15.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             age    Marloes     PiSCES    Radboud\n",
       "count  89.000000  89.000000  89.000000  89.000000\n",
       "mean   27.528090   6.932584  14.831461  14.865169\n",
       "std     6.629976   0.330212   1.298915   0.906931\n",
       "min    19.000000   5.000000   3.000000   7.000000\n",
       "25%    23.000000   7.000000  15.000000  15.000000\n",
       "50%    26.000000   7.000000  15.000000  15.000000\n",
       "75%    30.000000   7.000000  15.000000  15.000000\n",
       "max    53.000000   7.000000  16.000000  15.000000"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# NOTE: as for now we do not filter on the number of recordings\n",
    "df_metadata_relevant = (\n",
    "    # df_metadata[\n",
    "    # ~(\n",
    "    #     (df_metadata.Marloes >= 6)\n",
    "    #     & (df_metadata.Radboud >= 14)\n",
    "    #     & (df_metadata.PiSCES >= 14)\n",
    "    # )\n",
    "    # ]\n",
    "    df_metadata.reset_index(drop=True).copy()\n",
    ")\n",
    "\n",
    "df_metadata_relevant[\n",
    "    [\"age\", \"sex\", \"device\", \"Marloes\", \"PiSCES\", \"Radboud\"]\n",
    "].describe()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "95bcc4cd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "28\t7\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    f\"{round(df_metadata_relevant.age.mean())}\\t{round(df_metadata_relevant.age.std())}\"\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "68d92d7b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "female    45\n",
       "male      43\n",
       "other      1\n",
       "Name: sex, dtype: int64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "pc-mic       36\n",
       "headset      17\n",
       "earphones    16\n",
       "other         6\n",
       "Name: device, dtype: int64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([<AxesSubplot:ylabel='Frequency'>], dtype=object)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(df_metadata_relevant.sex.value_counts())\n",
    "print(\"-\" * 40)\n",
    "display(df_metadata_relevant.device.value_counts())\n",
    "print(\"-\" * 40)\n",
    "df_metadata_relevant.education.value_counts()\n",
    "\n",
    "# Also plot the distributions\n",
    "df_metadata_relevant[[\"age\"]].plot(\n",
    "    kind=\"hist\",\n",
    "    subplots=True,\n",
    "    sharex=False,\n",
    "    figsize=(10, 5),\n",
    "    backend=\"matplotlib\",\n",
    "    color=color_palette,\n",
    "    bins=25,\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e2ebb1e9",
   "metadata": {},
   "source": [
    "## Extract recording info for each session"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "6816aa09",
   "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>pic_name</th>\n",
       "      <th>time_str</th>\n",
       "      <th>wav_size</th>\n",
       "      <th>wav_sample_rate</th>\n",
       "      <th>wav_duration_s</th>\n",
       "      <th>wav_timestamp</th>\n",
       "      <th>ID</th>\n",
       "      <th>DB</th>\n",
       "      <th>arousal</th>\n",
       "      <th>valence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>687</th>\n",
       "      <td>Picture 80</td>\n",
       "      <td>14_51_30</td>\n",
       "      <td>6709292</td>\n",
       "      <td>48000</td>\n",
       "      <td>69.89</td>\n",
       "      <td>2022-06-23 14:51:36+02:00</td>\n",
       "      <td>31e224eb-52af-4edf-b42b-a22960958dcf</td>\n",
       "      <td>PiSCES</td>\n",
       "      <td>30</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1596</th>\n",
       "      <td>Rafd090_58_Caucasian_female_neutral_frontal</td>\n",
       "      <td>12_45_25</td>\n",
       "      <td>2727980</td>\n",
       "      <td>48000</td>\n",
       "      <td>28.42</td>\n",
       "      <td>2022-06-23 12:45:26+02:00</td>\n",
       "      <td>992fb536-7504-4c58-adaf-bd6498d96dc7</td>\n",
       "      <td>Radboud</td>\n",
       "      <td>22</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1315</th>\n",
       "      <td>Picture 65</td>\n",
       "      <td>13_46_13</td>\n",
       "      <td>6660140</td>\n",
       "      <td>44100</td>\n",
       "      <td>75.51</td>\n",
       "      <td>2020-11-29 13:46:18+01:00</td>\n",
       "      <td>46914c9d-f87b-4ee5-aaf4-fe30d0ec272a</td>\n",
       "      <td>PiSCES</td>\n",
       "      <td>72</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         pic_name  time_str  wav_size  \\\n",
       "687                                    Picture 80  14_51_30   6709292   \n",
       "1596  Rafd090_58_Caucasian_female_neutral_frontal  12_45_25   2727980   \n",
       "1315                                   Picture 65  13_46_13   6660140   \n",
       "\n",
       "      wav_sample_rate  wav_duration_s             wav_timestamp  \\\n",
       "687             48000           69.89 2022-06-23 14:51:36+02:00   \n",
       "1596            48000           28.42 2022-06-23 12:45:26+02:00   \n",
       "1315            44100           75.51 2020-11-29 13:46:18+01:00   \n",
       "\n",
       "                                        ID       DB  arousal  valence  \n",
       "687   31e224eb-52af-4edf-b42b-a22960958dcf   PiSCES       30       55  \n",
       "1596  992fb536-7504-4c58-adaf-bd6498d96dc7  Radboud       22       75  \n",
       "1315  46914c9d-f87b-4ee5-aaf4-fe30d0ec272a   PiSCES       72       19  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------------------------------------------------\n"
     ]
    },
    {
     "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>age</th>\n",
       "      <th>device</th>\n",
       "      <th>education</th>\n",
       "      <th>prolific_token</th>\n",
       "      <th>sex</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>ID</th>\n",
       "      <th>Marloes</th>\n",
       "      <th>PiSCES</th>\n",
       "      <th>Radboud</th>\n",
       "      <th>pic_name</th>\n",
       "      <th>time_str</th>\n",
       "      <th>wav_size</th>\n",
       "      <th>wav_sample_rate</th>\n",
       "      <th>wav_duration_s</th>\n",
       "      <th>wav_timestamp</th>\n",
       "      <th>DB</th>\n",
       "      <th>arousal</th>\n",
       "      <th>valence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3084</th>\n",
       "      <td>23</td>\n",
       "      <td>earphones</td>\n",
       "      <td>unif</td>\n",
       "      <td>5c5c3370dcea9700016e0bc4</td>\n",
       "      <td>male</td>\n",
       "      <td>2022-07-16 12:02:13.631034+02:00</td>\n",
       "      <td>271093f8-4d3f-4ee0-8a6e-faf9a61c83a4</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>Rafd090_36_Caucasian_male_neutral_frontal</td>\n",
       "      <td>12_24_58</td>\n",
       "      <td>4243500</td>\n",
       "      <td>48000</td>\n",
       "      <td>44.20</td>\n",
       "      <td>2022-07-16 12:24:58+02:00</td>\n",
       "      <td>Radboud</td>\n",
       "      <td>41</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1821</th>\n",
       "      <td>23</td>\n",
       "      <td>pc-mic</td>\n",
       "      <td>ho</td>\n",
       "      <td>5eb9008b8224d385674c39b0</td>\n",
       "      <td>male</td>\n",
       "      <td>2022-06-29 18:55:57.528852+02:00</td>\n",
       "      <td>0e919efd-954e-4c12-98fb-dcf2d0a2b4e0</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>Rafd090_02_Caucasian_female_neutral_frontal</td>\n",
       "      <td>19_04_56</td>\n",
       "      <td>4063276</td>\n",
       "      <td>48000</td>\n",
       "      <td>42.33</td>\n",
       "      <td>2022-06-29 19:05:00+02:00</td>\n",
       "      <td>Radboud</td>\n",
       "      <td>68</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>905</th>\n",
       "      <td>21</td>\n",
       "      <td>headset</td>\n",
       "      <td>ho</td>\n",
       "      <td>n.a.</td>\n",
       "      <td>other</td>\n",
       "      <td>2022-07-28 09:56:59.413711+02:00</td>\n",
       "      <td>69a7b54c-9db3-4b35-9e85-dd2592c10a00</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>Picture 132</td>\n",
       "      <td>10_44_41</td>\n",
       "      <td>10125356</td>\n",
       "      <td>48000</td>\n",
       "      <td>105.47</td>\n",
       "      <td>2022-07-28 10:44:48+02:00</td>\n",
       "      <td>PiSCES</td>\n",
       "      <td>39</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      age     device education            prolific_token    sex  \\\n",
       "3084   23  earphones      unif  5c5c3370dcea9700016e0bc4   male   \n",
       "1821   23     pc-mic        ho  5eb9008b8224d385674c39b0   male   \n",
       "905    21    headset        ho                      n.a.  other   \n",
       "\n",
       "                            timestamp                                    ID  \\\n",
       "3084 2022-07-16 12:02:13.631034+02:00  271093f8-4d3f-4ee0-8a6e-faf9a61c83a4   \n",
       "1821 2022-06-29 18:55:57.528852+02:00  0e919efd-954e-4c12-98fb-dcf2d0a2b4e0   \n",
       "905  2022-07-28 09:56:59.413711+02:00  69a7b54c-9db3-4b35-9e85-dd2592c10a00   \n",
       "\n",
       "      Marloes  PiSCES  Radboud                                     pic_name  \\\n",
       "3084        7      15       15    Rafd090_36_Caucasian_male_neutral_frontal   \n",
       "1821        7      15       15  Rafd090_02_Caucasian_female_neutral_frontal   \n",
       "905         7      15       15                                  Picture 132   \n",
       "\n",
       "      time_str  wav_size  wav_sample_rate  wav_duration_s  \\\n",
       "3084  12_24_58   4243500            48000           44.20   \n",
       "1821  19_04_56   4063276            48000           42.33   \n",
       "905   10_44_41  10125356            48000          105.47   \n",
       "\n",
       "                 wav_timestamp       DB  arousal  valence  \n",
       "3084 2022-07-16 12:24:58+02:00  Radboud       41       38  \n",
       "1821 2022-06-29 19:05:00+02:00  Radboud       68       81  \n",
       "905  2022-07-28 10:44:48+02:00   PiSCES       39       66  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "session_dict_list = []\n",
    "\n",
    "\n",
    "# Iterate over each session\n",
    "for folder in speech_data_session_dir.iterdir():\n",
    "    if not folder.is_dir():\n",
    "        continue\n",
    "\n",
    "    def parse_pic_name_ts(db_folder, pic_name, time_str, sep=\"__\") -> dict:\n",
    "        \"\"\"Get the arousal valence json for the corresponding description\"\"\"\n",
    "        # what can we save from the wav file?\n",
    "        json_f = sep.join([pic_name, time_str]) + \".json\"\n",
    "        json_f = db_folder.joinpath(json_f)\n",
    "        if not json_f.is_file():\n",
    "            # print('json_f does not exist')\n",
    "            return {}\n",
    "\n",
    "        with open(json_f, \"r\") as f:\n",
    "            return json.load(f)\n",
    "\n",
    "    def get_wav_json_mapping(db_folder: Path) -> dict:\n",
    "        \"\"\"Get the mapping from wav file to corresponding json (arousal/valence) file\"\"\"\n",
    "        # parse the wav files to construct a dataframe\n",
    "        wav_list = list(db_folder.glob(\"*.wav\"))\n",
    "        df_db_wav = pd.Series(\n",
    "            seq(wav_list).map(lambda x: x.name), name=\"wav_files\", dtype=\"str\"\n",
    "        ).to_frame()\n",
    "        df_db_wav[\"wav_timestamp\"] = (\n",
    "            seq(wav_list)\n",
    "            .map(\n",
    "                lambda wav_file: pd.Timestamp(\n",
    "                    wav_file.stat().st_mtime, unit=\"s\", tz=\"Europe/Brussels\"\n",
    "                )\n",
    "            )\n",
    "            .to_list()\n",
    "        )\n",
    "        df_db_wav = df_db_wav.sort_values(by=\"wav_timestamp\")\n",
    "\n",
    "        # parse the json files to construct a dataframe\n",
    "        json_list = list(db_folder.glob(\"*.json\"))\n",
    "        df_db_json = pd.Series(\n",
    "            seq(json_list).map(lambda x: x.name), name=\"json_files\", dtype=\"str\"\n",
    "        ).to_frame()\n",
    "        df_db_json[\"json_timestamp\"] = (\n",
    "            seq(json_list)\n",
    "            .map(\n",
    "                lambda json_file: pd.Timestamp(\n",
    "                    json_file.stat().st_mtime, unit=\"s\", tz=\"Europe/Brussels\"\n",
    "                )\n",
    "            )\n",
    "            .to_list()\n",
    "        )\n",
    "        df_db_json = df_db_json.sort_values(by=\"json_timestamp\")\n",
    "\n",
    "        json_wav_mapping = pd.merge_asof(\n",
    "            left=df_db_wav,\n",
    "            right=df_db_json,\n",
    "            left_on=\"wav_timestamp\",\n",
    "            right_on=\"json_timestamp\",\n",
    "            tolerance=pd.Timedelta(minutes=5),\n",
    "            # A \"forward\" search selects the first row in the right DataFrame whose 'on' key\n",
    "            # is greater than or equal to the left's key.\n",
    "            direction=\"forward\",\n",
    "        )\n",
    "\n",
    "        json_wav_mapping = json_wav_mapping.dropna(how=\"any\", axis=0)\n",
    "        json_wav_mapping = json_wav_mapping[\n",
    "            ~(\n",
    "                json_wav_mapping[\"wav_timestamp\"]\n",
    "                < json_wav_mapping[\"json_timestamp\"].shift(1)\n",
    "            )\n",
    "        ].reset_index(drop=True)\n",
    "\n",
    "        # Asserts to check that the merge is performed correctly\n",
    "        assert all(\n",
    "            (\n",
    "                json_wav_mapping[\"json_timestamp\"].shift(1)\n",
    "                < json_wav_mapping[\"wav_timestamp\"]\n",
    "            ).values[1:]\n",
    "        )\n",
    "        wav_json_mapping_dict = (\n",
    "            json_wav_mapping[[\"json_files\", \"wav_files\"]]\n",
    "            .set_index(\"wav_files\")\n",
    "            .to_dict()[\"json_files\"]\n",
    "        )\n",
    "        return wav_json_mapping_dict\n",
    "\n",
    "    # Iterate over each database folder\n",
    "    for db_folder in [folder.joinpath(x) for x in [\"marloes\", \"Radboud\", \"PiSCES\"]]:\n",
    "        wav_json_mapping_dict = get_wav_json_mapping(db_folder)\n",
    "\n",
    "        for wav_file in db_folder.glob(\"*.wav\"):\n",
    "            pic_name, time_str = wav_file.name.split(\"__\")\n",
    "            metadata_dict = parse_pic_name_ts(db_folder, pic_name, time_str)\n",
    "            metadata_dict[\"pic_name\"] = pic_name\n",
    "            metadata_dict[\"time_str\"] = time_str.split(\".\")[0]\n",
    "            metadata_dict[\"wav_size\"] = wav_file.stat().st_size\n",
    "            data, sample_rate = librosa.load(path=wav_file, sr=None)\n",
    "            metadata_dict[\"wav_sample_rate\"] = sample_rate\n",
    "            metadata_dict[\"wav_duration_s\"] = round(data.shape[0] / sample_rate, 2)\n",
    "            ts = pd.Timestamp(wav_file.stat().st_mtime, unit=\"s\", tz=\"Europe/Brussels\")\n",
    "            metadata_dict[\"wav_timestamp\"] = ts + ts.utcoffset()\n",
    "            metadata_dict[\"ID\"] = db_folder.parent.name.split(\"__\")[-1]\n",
    "            metadata_dict[\"DB\"] = db_folder.name[:1].upper() + db_folder.name[1:]\n",
    "\n",
    "            # also read the json file to fetch the arousal and valence values\n",
    "            if wav_file.name in wav_json_mapping_dict:\n",
    "                with open(\n",
    "                    db_folder.joinpath(wav_json_mapping_dict[wav_file.name]), \"r\"\n",
    "                ) as f:\n",
    "                    json_dct = json.load(f)\n",
    "                    metadata_dict[\"arousal\"] = int(json_dct.get(\"arousal\"))\n",
    "                    metadata_dict[\"valence\"] = int(json_dct.get(\"valence\"))\n",
    "\n",
    "            session_dict_list.append(metadata_dict)\n",
    "\n",
    "df_session = pd.DataFrame(session_dict_list)\n",
    "\n",
    "\n",
    "# Sanity check - to see whether the wav_time_obj corresponds with the time_str\n",
    "# there must be < 10 minutes of difference between the two\n",
    "assert all(\n",
    "    (\n",
    "        df_session.apply(\n",
    "            lambda r: r.wav_timestamp.replace(\n",
    "                **dict(\n",
    "                    zip(\n",
    "                        [\"hour\", \"minute\", \"second\"],\n",
    "                        list(map(int, r.time_str.split(\"_\"))),\n",
    "                    )\n",
    "                )\n",
    "            ),\n",
    "            axis=1,\n",
    "        )\n",
    "        - df_session.wav_timestamp\n",
    "    ).abs()\n",
    "    < pd.Timedelta(minutes=10)\n",
    ")\n",
    "\n",
    "\n",
    "del folder, db_folder, wav_file, ts, metadata_dict, session_dict_list, time_str, f\n",
    "del parse_pic_name_ts, pic_name, wav_json_mapping_dict, json_dct, data, sample_rate\n",
    "\n",
    "display(df_session.sample(3))\n",
    "\n",
    "print(\"-\" * 88)\n",
    "\n",
    "df_session_tot = df_metadata_relevant.merge(\n",
    "    df_session, how=\"left\", left_on=\"ID\", right_on=\"ID\"\n",
    ")\n",
    "display(df_session_tot.sample(3))\n",
    "\n",
    "del df_session\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4908801d",
   "metadata": {},
   "source": [
    "### Add the `DB_no` and `session_no` column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "4e71e49a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[  1  10   3   5   2   8 100  12 200]\n",
      "[0 4 2 3 5 1 7 6 8]\n",
      "[1 6 3 4 2 5 8 7 9]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "# This is a simplified testcase to see whether the applied logic will work\n",
    "test_arr = [1, 10, 3, 5, 2, 8, 100, 12, 200]\n",
    "print(np.array(test_arr))\n",
    "print(np.argsort(test_arr))\n",
    "print(1 + np.argsort(np.argsort(test_arr)))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "2fd220cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_session_tot[\"DB_no\"] = df_session_tot.groupby([\"ID\", \"DB\"])[\"wav_timestamp\"].apply(\n",
    "    lambda s: 1 + s.argsort().argsort()\n",
    ")\n",
    "df_session_tot[\"session_no\"] = df_session_tot.groupby([\"ID\"])[\"wav_timestamp\"].apply(\n",
    "    lambda s: 1 + s.argsort().argsort()\n",
    ")\n",
    "\n",
    "df_session_tot[\"pic_no\"] = None\n",
    "pic_mask = df_session_tot.DB != \"Marloes\"\n",
    "df_session_tot.loc[pic_mask, \"pic_no\"] = (\n",
    "    df_session_tot[pic_mask]\n",
    "    .groupby([\"ID\"])\n",
    "    .wav_timestamp.apply(lambda s: 1 + s.argsort().argsort())\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "cf97fcac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    3257\n",
       "True        3\n",
       "dtype: int64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sort df session chronologically per user\n",
    "df_session_tot = df_session_tot.sort_values(\n",
    "    by=[\"ID\", \"wav_timestamp\"], ignore_index=True\n",
    ")\n",
    "\n",
    "# In some rare cases, users got the same image twice; we remove the sessions where\n",
    "# they got that image for the second time\n",
    "duplicated_mask = df_session_tot.duplicated(subset=[\"ID\", \"pic_name\"], keep=\"first\") & (\n",
    "    df_session_tot.DB != \"Marloes\"\n",
    ")\n",
    "display(duplicated_mask.value_counts())\n",
    "df_session_tot = df_session_tot[~duplicated_mask].copy().reset_index(drop=True)\n",
    "del duplicated_mask\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "46bd5517",
   "metadata": {},
   "source": [
    "## **Session analysis**: duration based visualizations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "1b1018e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:ylabel='Frequency'>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(\n",
    "    (df_session_tot.wav_timestamp - df_session_tot.timestamp).dt.total_seconds() / 60\n",
    ").plot.hist(color=color_palette, figsize=(10, 5), bins=30)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "659abb03",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-1.0, 120.0)"
      ]
     },
     "execution_count": 32,
     "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": [
    "# A histogram of the recording durations (without checking for voicing)\n",
    "fig, ax = plt.subplots(figsize=(20, 8))\n",
    "sns.set_context(\"talk\")\n",
    "sns.histplot(\n",
    "    data=df_session_tot,\n",
    "    hue=\"DB\",\n",
    "    x=\"wav_duration_s\",\n",
    "    palette=color_palette[1:2] + color_palette[:1] + color_palette[2:3],\n",
    "    # palette='Set3',\n",
    "    ax=ax,\n",
    "    alpha=0.4,\n",
    "    bins=80,\n",
    "    binrange=[0, 120],\n",
    ")\n",
    "\n",
    "ax.axvline(x=15, ls=\":\", c=\"black\")\n",
    "ax.set_xlim(-1, 120)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "daf7f59c",
   "metadata": {},
   "source": [
    "### Analyzing the sample rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "cc1de622",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_session_tot.wav_sample_rate.value_counts().plot.bar(color=color_palette)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "229c9360",
   "metadata": {},
   "source": [
    "since the recordings were performed on different days, the sample rate varies."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5b3b9300",
   "metadata": {},
   "source": [
    "### Investigating the file duration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "f2cb430d",
   "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></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ID</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>prolific_token</th>\n",
       "      <th>DB</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9a725424-aafb-4d9b-8be7-acad80a46614</th>\n",
       "      <th>2022-07-27 14:55:01.266930+02:00</th>\n",
       "      <th>60fe5f65bf40c582873fefc2</th>\n",
       "      <th>Radboud</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">a2dbee10-fc31-42ec-89f8-9b4e5fa74b7f</th>\n",
       "      <th rowspan=\"2\" valign=\"top\">2022-06-27 22:18:01.896264+02:00</th>\n",
       "      <th rowspan=\"2\" valign=\"top\">5f2be3f5d81eaf0009520f8b</th>\n",
       "      <th>PiSCES</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Radboud</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c30db4c2-a62a-4d69-8f1b-893a1f4b8c9d</th>\n",
       "      <th>2022-07-16 21:49:33.730743+02:00</th>\n",
       "      <th>5e360d610d43625c7da0a09d</th>\n",
       "      <th>Radboud</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                         0\n",
       "ID                                   timestamp                        prolific_token           DB         \n",
       "9a725424-aafb-4d9b-8be7-acad80a46614 2022-07-27 14:55:01.266930+02:00 60fe5f65bf40c582873fefc2 Radboud   1\n",
       "a2dbee10-fc31-42ec-89f8-9b4e5fa74b7f 2022-06-27 22:18:01.896264+02:00 5f2be3f5d81eaf0009520f8b PiSCES   14\n",
       "                                                                                               Radboud  10\n",
       "c30db4c2-a62a-4d69-8f1b-893a1f4b8c9d 2022-07-16 21:49:33.730743+02:00 5e360d610d43625c7da0a09d Radboud   1"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_session_tot[\n",
    "    (df_session_tot.wav_size > 400)\n",
    "    & (df_session_tot.wav_duration_s < 15)\n",
    "    & (df_session_tot.DB != \"Marloes\")\n",
    "].groupby([\"ID\", \"timestamp\", \"prolific_token\", \"DB\"]).size().to_frame()\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "da8b7699",
   "metadata": {},
   "source": [
    "`a2dbee10-fc31-42ec-89f8-9b4e5fa74b7f` has a lot of short segments => we will omit this users audio data. The valence/arousal data is still of interest though.\n",
    "\n",
    "After auditory inspection, we could conclude that this user did not put any effort into the study."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9962e7d6",
   "metadata": {},
   "source": [
    "### Parsing events"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "5b649f26",
   "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>age</th>\n",
       "      <th>device</th>\n",
       "      <th>education</th>\n",
       "      <th>prolific_token</th>\n",
       "      <th>sex</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>ID</th>\n",
       "      <th>Marloes</th>\n",
       "      <th>PiSCES</th>\n",
       "      <th>Radboud</th>\n",
       "      <th>pic_name</th>\n",
       "      <th>time_str</th>\n",
       "      <th>wav_size</th>\n",
       "      <th>wav_sample_rate</th>\n",
       "      <th>wav_duration_s</th>\n",
       "      <th>wav_timestamp</th>\n",
       "      <th>DB</th>\n",
       "      <th>arousal</th>\n",
       "      <th>valence</th>\n",
       "      <th>DB_no</th>\n",
       "      <th>session_no</th>\n",
       "      <th>pic_no</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2836</th>\n",
       "      <td>20</td>\n",
       "      <td>pc-mic</td>\n",
       "      <td>so</td>\n",
       "      <td>5f58064bbb4b673607a4c7f4</td>\n",
       "      <td>male</td>\n",
       "      <td>2022-07-26 21:46:34.776329+02:00</td>\n",
       "      <td>e03516f6-1af7-4c72-ad49-41b6db4733d7</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>Rafd090_46_Caucasian_male_neutral_frontal</td>\n",
       "      <td>22_10_27</td>\n",
       "      <td>5087276</td>\n",
       "      <td>48000</td>\n",
       "      <td>52.99</td>\n",
       "      <td>2022-07-26 22:10:28+02:00</td>\n",
       "      <td>Radboud</td>\n",
       "      <td>77</td>\n",
       "      <td>53</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>894</th>\n",
       "      <td>29</td>\n",
       "      <td>headset</td>\n",
       "      <td>so</td>\n",
       "      <td>5beb4883e0a39400015430f5</td>\n",
       "      <td>male</td>\n",
       "      <td>2022-06-29 14:11:06.409834+02:00</td>\n",
       "      <td>42c842a5-7051-44c3-af42-cf824fded959</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>Picture 98</td>\n",
       "      <td>14_54_49</td>\n",
       "      <td>7249964</td>\n",
       "      <td>48000</td>\n",
       "      <td>75.52</td>\n",
       "      <td>2022-06-29 14:54:48+02:00</td>\n",
       "      <td>PiSCES</td>\n",
       "      <td>60</td>\n",
       "      <td>69</td>\n",
       "      <td>13</td>\n",
       "      <td>30</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2995</th>\n",
       "      <td>27</td>\n",
       "      <td>pc-mic</td>\n",
       "      <td>unif</td>\n",
       "      <td>n.a.</td>\n",
       "      <td>female</td>\n",
       "      <td>2022-07-16 20:29:30.890814+02:00</td>\n",
       "      <td>effe47df-0f91-489d-8b7d-547f6fed85dd</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>Rafd090_33_Caucasian_male_neutral_frontal</td>\n",
       "      <td>21_24_56</td>\n",
       "      <td>4464684</td>\n",
       "      <td>44100</td>\n",
       "      <td>50.62</td>\n",
       "      <td>2022-07-16 21:25:14+02:00</td>\n",
       "      <td>Radboud</td>\n",
       "      <td>51</td>\n",
       "      <td>55</td>\n",
       "      <td>11</td>\n",
       "      <td>27</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      age   device education            prolific_token     sex  \\\n",
       "2836   20   pc-mic        so  5f58064bbb4b673607a4c7f4    male   \n",
       "894    29  headset        so  5beb4883e0a39400015430f5    male   \n",
       "2995   27   pc-mic      unif                      n.a.  female   \n",
       "\n",
       "                            timestamp                                    ID  \\\n",
       "2836 2022-07-26 21:46:34.776329+02:00  e03516f6-1af7-4c72-ad49-41b6db4733d7   \n",
       "894  2022-06-29 14:11:06.409834+02:00  42c842a5-7051-44c3-af42-cf824fded959   \n",
       "2995 2022-07-16 20:29:30.890814+02:00  effe47df-0f91-489d-8b7d-547f6fed85dd   \n",
       "\n",
       "      Marloes  PiSCES  Radboud                                   pic_name  \\\n",
       "2836        7      15       15  Rafd090_46_Caucasian_male_neutral_frontal   \n",
       "894         7      15       15                                 Picture 98   \n",
       "2995        7      15       15  Rafd090_33_Caucasian_male_neutral_frontal   \n",
       "\n",
       "      time_str  wav_size  wav_sample_rate  wav_duration_s  \\\n",
       "2836  22_10_27   5087276            48000           52.99   \n",
       "894   14_54_49   7249964            48000           75.52   \n",
       "2995  21_24_56   4464684            44100           50.62   \n",
       "\n",
       "                 wav_timestamp       DB  arousal  valence  DB_no  session_no  \\\n",
       "2836 2022-07-26 22:10:28+02:00  Radboud       77       53      6          15   \n",
       "894  2022-06-29 14:54:48+02:00   PiSCES       60       69     13          30   \n",
       "2995 2022-07-16 21:25:14+02:00  Radboud       51       55     11          27   \n",
       "\n",
       "     pic_no  \n",
       "2836     12  \n",
       "894      25  \n",
       "2995     22  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_session_tot.sample(3)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "740946f2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3257, 22)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(3257, 22)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_session_tot_cleaned = df_session_tot.copy()\n",
    "df_session_tot_cleaned = df_session_tot_cleaned.sort_values(\n",
    "    by=[\"ID\", \"wav_timestamp\"], ignore_index=True\n",
    ")\n",
    "print(df_session_tot.shape)\n",
    "display(df_session_tot_cleaned.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "ba847bbb",
   "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>session_no</th>\n",
       "      <th>wav_timestamp</th>\n",
       "      <th>DB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3147</th>\n",
       "      <td>1</td>\n",
       "      <td>2020-12-01 17:29:06+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3148</th>\n",
       "      <td>2</td>\n",
       "      <td>2020-12-01 17:29:42+01:00</td>\n",
       "      <td>Radboud</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3149</th>\n",
       "      <td>3</td>\n",
       "      <td>2020-12-01 17:30:44+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3150</th>\n",
       "      <td>4</td>\n",
       "      <td>2020-12-01 17:31:22+01:00</td>\n",
       "      <td>Radboud</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3151</th>\n",
       "      <td>5</td>\n",
       "      <td>2020-12-01 17:32:34+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3152</th>\n",
       "      <td>6</td>\n",
       "      <td>2020-12-01 17:33:10+01:00</td>\n",
       "      <td>Marloes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3153</th>\n",
       "      <td>7</td>\n",
       "      <td>2020-12-01 17:33:52+01:00</td>\n",
       "      <td>Radboud</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3154</th>\n",
       "      <td>8</td>\n",
       "      <td>2020-12-01 17:34:44+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3155</th>\n",
       "      <td>9</td>\n",
       "      <td>2020-12-01 17:36:02+01:00</td>\n",
       "      <td>Radboud</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3156</th>\n",
       "      <td>10</td>\n",
       "      <td>2020-12-01 17:36:40+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      session_no             wav_timestamp       DB\n",
       "3147           1 2020-12-01 17:29:06+01:00   PiSCES\n",
       "3148           2 2020-12-01 17:29:42+01:00  Radboud\n",
       "3149           3 2020-12-01 17:30:44+01:00   PiSCES\n",
       "3150           4 2020-12-01 17:31:22+01:00  Radboud\n",
       "3151           5 2020-12-01 17:32:34+01:00   PiSCES\n",
       "3152           6 2020-12-01 17:33:10+01:00  Marloes\n",
       "3153           7 2020-12-01 17:33:52+01:00  Radboud\n",
       "3154           8 2020-12-01 17:34:44+01:00   PiSCES\n",
       "3155           9 2020-12-01 17:36:02+01:00  Radboud\n",
       "3156          10 2020-12-01 17:36:40+01:00   PiSCES"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>session_no</th>\n",
       "      <th>wav_timestamp</th>\n",
       "      <th>DB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3147</th>\n",
       "      <td>2</td>\n",
       "      <td>2020-12-01 17:29:06+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3148</th>\n",
       "      <td>3</td>\n",
       "      <td>2020-12-01 17:29:42+01:00</td>\n",
       "      <td>Radboud</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3149</th>\n",
       "      <td>4</td>\n",
       "      <td>2020-12-01 17:30:44+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3150</th>\n",
       "      <td>5</td>\n",
       "      <td>2020-12-01 17:31:22+01:00</td>\n",
       "      <td>Radboud</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3151</th>\n",
       "      <td>6</td>\n",
       "      <td>2020-12-01 17:32:34+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3152</th>\n",
       "      <td>7</td>\n",
       "      <td>2020-12-01 17:33:10+01:00</td>\n",
       "      <td>Marloes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3153</th>\n",
       "      <td>8</td>\n",
       "      <td>2020-12-01 17:33:52+01:00</td>\n",
       "      <td>Radboud</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3154</th>\n",
       "      <td>9</td>\n",
       "      <td>2020-12-01 17:34:44+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3155</th>\n",
       "      <td>10</td>\n",
       "      <td>2020-12-01 17:36:02+01:00</td>\n",
       "      <td>Radboud</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3156</th>\n",
       "      <td>11</td>\n",
       "      <td>2020-12-01 17:36:40+01:00</td>\n",
       "      <td>PiSCES</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      session_no             wav_timestamp       DB\n",
       "3147           2 2020-12-01 17:29:06+01:00   PiSCES\n",
       "3148           3 2020-12-01 17:29:42+01:00  Radboud\n",
       "3149           4 2020-12-01 17:30:44+01:00   PiSCES\n",
       "3150           5 2020-12-01 17:31:22+01:00  Radboud\n",
       "3151           6 2020-12-01 17:32:34+01:00   PiSCES\n",
       "3152           7 2020-12-01 17:33:10+01:00  Marloes\n",
       "3153           8 2020-12-01 17:33:52+01:00  Radboud\n",
       "3154           9 2020-12-01 17:34:44+01:00   PiSCES\n",
       "3155          10 2020-12-01 17:36:02+01:00  Radboud\n",
       "3156          11 2020-12-01 17:36:40+01:00   PiSCES"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# custom parsing\n",
    "display(\n",
    "    df_session_tot_cleaned[\n",
    "        df_session_tot_cleaned.ID == \"f57eb3a1-bc73-4fb8-9c40-8a31a6b67e34\"\n",
    "    ]\n",
    "    .sort_values(by=\"wav_timestamp\")[[\"session_no\", \"wav_timestamp\", \"DB\"]]\n",
    "    .iloc[:10]\n",
    ")\n",
    "\n",
    "# this user starts with pisces and has a total of 6 marloes instead of 7\n",
    "# for some reason, the first marloes was not stored -> adjust the session_no and DB_no\n",
    "df_session_tot_cleaned.loc[\n",
    "    df_session_tot_cleaned.ID == \"f57eb3a1-bc73-4fb8-9c40-8a31a6b67e34\", \"session_no\"\n",
    "] += 1\n",
    "df_session_tot_cleaned.loc[\n",
    "    (df_session_tot_cleaned.ID == (\"f57eb3a1-bc73-4fb8-9c40-8a31a6b67e34\"))\n",
    "    & (df_session_tot_cleaned.DB == \"Marloes\"),\n",
    "    \"DB_no\",\n",
    "] += 1\n",
    "display(\n",
    "    df_session_tot_cleaned[\n",
    "        df_session_tot_cleaned.ID == \"f57eb3a1-bc73-4fb8-9c40-8a31a6b67e34\"\n",
    "    ]\n",
    "    .sort_values(by=\"wav_timestamp\")[[\"session_no\", \"wav_timestamp\", \"DB\"]]\n",
    "    .iloc[:10]\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d85a956b",
   "metadata": {},
   "source": [
    "### Saving the session dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "34e61a94",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_session_tot_cleaned.to_parquet(\n",
    "    loc_data_dir.joinpath(\"df_session_tot_cleaned.parquet\")\n",
    ")\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.13"
  },
  "vscode": {
   "interpreter": {
    "hash": "14accab11a8dde1829a82d5477aee5050c56a20d58c0fbbbf6574e6407175af4"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}