Artifacts/algorithms/fixed-text/kd-analysis-fixed-text.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from numpy import mean, absolute\n",
    "from scipy.spatial import distance\n",
    "from scipy.linalg import pinv\n",
    "\n",
    "from sklearn.metrics import roc_curve\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.pyplot import Rectangle\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "###########################################################################\n",
    "# 1. Anomaly detectors\n",
    "###########################################################################\n",
    "\n",
    "# The euclideanTrain and euclideanScore functions comprise the\n",
    "# Euclidean anomaly detector.  During training, the detector takes a\n",
    "# set of password-timing vectors (encoded as rows in a matrix) and\n",
    "# calculates the mean vector.  This mean vector is returned as the\n",
    "# detection model.  During scoring, the detector takes the detection\n",
    "# model and a new set of password-timing vectors (also encoded as rows\n",
    "# in a matrix) and calculates the squared Euclidean distance between\n",
    "# the mean vector and each of the new password-timing vectors.  These\n",
    "# scores are returned in a vector whose length is equal to the number\n",
    "# of password-timing vectors in the scoring matrix.\n",
    "\n",
    "def euclideanTrain(YTrain):\n",
    "    dmod = {'mean': np.mean(YTrain, axis=0)}\n",
    "    return dmod\n",
    "\n",
    "\n",
    "def euclideanScore(dmod, YScore):\n",
    "    p = len(dmod['mean'])\n",
    "    n = YScore.shape[0]\n",
    "\n",
    "    if YScore.shape[1] != p:\n",
    "        raise ValueError(\"Training/test feature length mismatch\")\n",
    "\n",
    "    meanMatrix = np.tile(dmod['mean'], (n, 1))\n",
    "\n",
    "    scores = np.sum(np.square(YScore - meanMatrix), axis=1)\n",
    "\n",
    "    return scores\n",
    "\n",
    "\n",
    "# The manhattanTrain and manhattanScore functions comprise the\n",
    "# Manhattan anomaly detector.  During training, the detector takes a\n",
    "# set of password-timing vectors (encoded as rows in a matrix) and\n",
    "# calculates the mean vector.  This mean vector is returned as the\n",
    "# detection model.  During scoring, the detector takes the detection\n",
    "# model and a new set of password-timing vectors (also encoded as rows\n",
    "# in a matrix) and calculates the Manhattan distance between the mean\n",
    "# vector and each of the new password-timing vectors.  These scores\n",
    "# are returned in a vector whose length is equal to the number of\n",
    "# password-timing vectors in the scoring matrix.\n",
    "\n",
    "def manhattanTrain(YTrain):\n",
    "    dmod = {'mean': np.mean(YTrain, axis=0)}\n",
    "    return dmod\n",
    "\n",
    "\n",
    "def manhattanScore(dmod, YScore):\n",
    "    p = len(dmod['mean'])\n",
    "    n = YScore.shape[0]\n",
    "\n",
    "    if YScore.shape[1] != p:\n",
    "        raise ValueError(\"Training/test feature length mismatch\")\n",
    "\n",
    "    meanMatrix = np.tile(dmod['mean'], (n, 1))\n",
    "\n",
    "    scores = np.sum(np.abs(YScore - meanMatrix), axis=1)\n",
    "\n",
    "    return scores\n",
    "\n",
    "\n",
    "# Scaled Manhattan distance\n",
    "def mad(data, axis=None):\n",
    "    return mean(absolute(data - mean(data, axis)), axis)\n",
    "\n",
    "def manhattanScaledTrain(YTrain):\n",
    "    dmod = {'mean': np.mean(YTrain, axis=0), 'mad': mad(YTrain,0)}\n",
    "    return dmod\n",
    "\n",
    "\n",
    "def manhattanScaledScore(dmod, YScore):\n",
    "    p = len(dmod['mean'])\n",
    "    n = YScore.shape[0]\n",
    "    \n",
    "    if YScore.shape[1] != p:\n",
    "        raise ValueError(\"Training/test feature length mismatch\")\n",
    "\n",
    "    meanMatrix = np.tile(dmod['mean'], (n, 1))\n",
    "    scores = np.sum(np.abs(YScore - meanMatrix)/dmod[\"mad\"].T, axis=1)\n",
    "    return scores\n",
    "\n",
    "\n",
    "# The mahalanobisTrain and mahalanobisScore functions comprise the\n",
    "# Mahalanobis anomaly detector.  During training, the detector takes a\n",
    "# set of password-timing vectors (encoded as rows in a matrix) and\n",
    "# calculates the mean vector and also the inverse of the covariance\n",
    "# matrix.  This vector and matrix are returned as the detection model.\n",
    "# During scoring, the detector takes the detection model and a new set\n",
    "# of password-timing vectors (also encoded as rows in a matrix) and\n",
    "# calculates the squared Mahalanobis distance between the mean vector\n",
    "# and each of the new password-timing vectors.  These scores are\n",
    "# returned in a vector whose length is equal to the number of\n",
    "# password-timing vectors in the scoring matrix.\n",
    "\n",
    "def mahalanobisTrain(YTrain):\n",
    "    dmod = {'mean': np.mean(YTrain, axis=0), 'covInv': pinv(\n",
    "        np.cov(YTrain, rowvar=False))}\n",
    "    return dmod\n",
    "\n",
    "\n",
    "def mahalanobisR(X, meanCol, IC):\n",
    "    m = []\n",
    "    for i in range(X.shape[0]):\n",
    "        m.append(distance.mahalanobis(X[i], meanCol, IC) ** 2)\n",
    "    return(m)\n",
    "\n",
    "\n",
    "def mahalanobisScore(dmod, YScore):\n",
    "    p = len(dmod['mean'])\n",
    "    n = YScore.shape[0]\n",
    "\n",
    "    if YScore.shape[1] != p:\n",
    "        raise ValueError(\"Training/test feature length mismatch\")\n",
    "\n",
    "    scores = mahalanobisR(YScore, dmod['mean'], dmod['covInv'])\n",
    "\n",
    "    return scores\n",
    "\n",
    "\n",
    "detectorSet = {\n",
    "    'Euclidean': {\n",
    "        'train': euclideanTrain,\n",
    "        'score': euclideanScore\n",
    "    },\n",
    "    'Manhattan': {\n",
    "        'train': manhattanTrain,\n",
    "        'score': manhattanScore\n",
    "    },\n",
    "    'ManhattanScaled':{\n",
    "        'train': manhattanScaledTrain,\n",
    "        'score': manhattanScaledScore\n",
    "    },\n",
    "    'Mahalanobis': {\n",
    "        'train': mahalanobisTrain,\n",
    "        'score': mahalanobisScore\n",
    "    }\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "###########################################################################\n",
    "# 2. Evaluation procedures\n",
    "###########################################################################\n",
    "\n",
    "# The calculateEqualError function takes a set of user scores and\n",
    "# impostor scores, makes an ROC curve using the ROCR functionality,\n",
    "# and then geometrically calculates the point at which the miss and\n",
    "# false-alarm (i.e., false-negative and false-positive) rates are\n",
    "# equal.\n",
    "\n",
    "\n",
    "def calculateEqualError(userScores, impostorScores):\n",
    "\n",
    "    predictions = np.concatenate((userScores, impostorScores), axis=None)\n",
    "    labels = [0] * len(userScores) + [1] * len(impostorScores)\n",
    "\n",
    "    fpr, tpr, thresholds = roc_curve(labels, predictions)\n",
    "    eer = fpr[np.nanargmin(np.absolute((1 - tpr) - fpr))]\n",
    "\n",
    "    # print(eer)\n",
    "    # print(thresholds)\n",
    "    #dists = tpr - fpr\n",
    "    # print(dists)\n",
    "    #idx1 = next(i for i, dist in enumerate(dists) if dist == min(dists[dists >= 0]))\n",
    "    #idx2 = next(i for i, dist in enumerate(dists) if dist == max(dists[dists < 0]))\n",
    "    # print(idx1)\n",
    "    #assert len([idx1]) == 1\n",
    "    #assert len([idx2]) == 1\n",
    "    #assert abs(idx1 - idx2) == 1\n",
    "\n",
    "    #x = [tpr[idx1], fpr[idx1]]\n",
    "    #y = [tpr[idx2], fpr[idx2]]\n",
    "    #a = (x[0] - x[1]) / (y[1] - x[1] - y[0] + x[0])\n",
    "    #eer = x[0] + a * (y[0] - x[0])\n",
    "\n",
    "    return eer\n",
    "\n",
    "\n",
    "def evaluateSubject(train, eval, evalSubject, detectorTrain, detectorScore):\n",
    "    # Extract the training, user scoring, and impostor scoring matrices\n",
    "    # for the subject. The training matrix is the first 200 password\n",
    "    # repetitions for the subject, corresponding to the first 4 sessions\n",
    "    # of passwords. The user scoring matrix is the last 200 password\n",
    "    # repetitions, and the impostor scoring matrix is the first 5\n",
    "    # repetitions from all the other subjects.\n",
    "\n",
    "    YTrain = train.loc[(train['subject'] == evalSubject) & (\n",
    "        train['sessionIndex'] <= 4), ~train.columns.isin(['subject', 'sessionIndex', 'rep'])].values\n",
    "\n",
    "    YScore0 = eval.loc[(eval['subject'] == evalSubject) & (\n",
    "        eval['sessionIndex'] > 4), ~eval.columns.isin(['subject', 'sessionIndex', 'rep'])].values\n",
    "\n",
    "    YScore1 = train.loc[(train['subject'] != evalSubject) & (train['sessionIndex'] == 1) & (\n",
    "        train['rep'] <= 5), ~train.columns.isin(['subject', 'sessionIndex', 'rep'])].values\n",
    "\n",
    "    # Run the training and scoring\n",
    "    dmod = detectorTrain(YTrain)\n",
    "    userScores = detectorScore(dmod, YScore0)\n",
    "    impostorScores = detectorScore(dmod, YScore1)\n",
    "\n",
    "    eer = calculateEqualError(userScores, impostorScores)\n",
    "\n",
    "    return eer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "###########################################################################\n",
    "# 3. Main procedure\n",
    "###########################################################################\n",
    "def eval(train, eval) -> pd.DataFrame:\n",
    "    # Retrieve the list of subjects\n",
    "    subjects = sorted(train[\"subject\"].unique())\n",
    "\n",
    "    training_set = train\n",
    "    eval_set = eval\n",
    "\n",
    "\n",
    "    # For each of the detectors, evaluate the detector on each subject,\n",
    "    # and record the equal-error rates in a dictionary.\n",
    "\n",
    "    eers = {}\n",
    "\n",
    "    for detectorName, detector in detectorSet.items():\n",
    "\n",
    "        #print(f\"Evaluating the {detectorName} detector\")\n",
    "        detectorTrain = detector['train']\n",
    "        detectorScore = detector['score']\n",
    "\n",
    "        eers[detectorName] = np.full(len(subjects), np.nan)\n",
    "\n",
    "        for i, subject in enumerate(subjects):\n",
    "            eer = evaluateSubject(training_set, eval_set, subject,\n",
    "                                detectorTrain=detectorTrain,\n",
    "                                detectorScore=detectorScore)\n",
    "\n",
    "            eers[detectorName][i] = eer\n",
    "            #print(f\"  {i+1}/{len(subjects)}: {eer}\", end='\\r')\n",
    "\n",
    "        #print(f\"  average equal-error: {np.mean(eers[detectorName])}\")\n",
    "\n",
    "    #print(\"Tabulating results:\")\n",
    "\n",
    "    eers = pd.DataFrame(eers, index=subjects)\n",
    "\n",
    "    res = pd.DataFrame({'err.mean': eers.mean(), 'err.std': eers.std()})\n",
    "    res = res.round(decimals=3)\n",
    "\n",
    "    return res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Original\n",
    "original = pd.read_csv('./DSL-StrongPasswordData.csv')\n",
    "original_out = eval(original, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Chromium Isolated\n",
    "chromium_isolated = pd.read_csv('./chromium/password_data_chromium_isolated_warmup/full_data_set.csv')\n",
    "chromium_isolated_ot = eval(original, chromium_isolated)\n",
    "chromium_isolated_tt = eval(chromium_isolated, chromium_isolated)\n",
    "chromium_isolated_to = eval(chromium_isolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Chromium Uniosolated\n",
    "chromium_unisolated = pd.read_csv('./chromium/password_data_chromium_unisolated_warmup/full_data_set.csv')\n",
    "chromium_unisolated_ot = eval(original, chromium_unisolated)\n",
    "chromium_unisolated_tt = eval(chromium_unisolated, chromium_unisolated)\n",
    "chromium_unisolated_to = eval(chromium_unisolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Chrome Isolated\n",
    "chrome_isolated = pd.read_csv('./chrome/password_data_chrome_isolated_warmup/full_data_set.csv')\n",
    "chrome_isolated_ot = eval(original, chrome_isolated)\n",
    "chrome_isolated_tt = eval(chrome_isolated, chrome_isolated)\n",
    "chrome_isolated_to = eval(chrome_isolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Chrome Uniosolated\n",
    "chrome_unisolated = pd.read_csv('./chrome/password_data_chrome_unisolated_warmup/full_data_set.csv')\n",
    "chrome_unisolated_ot = eval(original, chrome_unisolated)\n",
    "chrome_unisolated_tt = eval(chrome_unisolated, chrome_unisolated)\n",
    "chrome_unisolated_to = eval(chrome_unisolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Edge Isolated\n",
    "edge_isolated = pd.read_csv('./edge/password_data_edge_isolated_warmup/full_data_set.csv')\n",
    "edge_isolated_ot = eval(original, edge_isolated)\n",
    "edge_isolated_tt = eval(edge_isolated, edge_isolated)\n",
    "edge_isolated_to = eval(edge_isolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Edge Uniosolated\n",
    "edge_unisolated = pd.read_csv('./edge/password_data_edge_unisolated_warmup/full_data_set.csv')\n",
    "edge_unisolated_ot = eval(original, edge_isolated)\n",
    "edge_unisolated_tt = eval(edge_isolated, edge_isolated)\n",
    "edge_unisolated_to = eval(edge_isolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Safari Isolated\n",
    "safari_isolated = pd.read_csv('./safari/password_data_safari_isolated_warmup/full_data_set.csv')\n",
    "safari_isolated_ot = eval(original, safari_isolated)\n",
    "safari_isolated_tt = eval(safari_isolated, safari_isolated)\n",
    "safari_isolated_to = eval(safari_isolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Safari Uniosolated\n",
    "safari_unisolated = pd.read_csv('./safari/password_data_safari_unisolated_warmup/full_data_set.csv')\n",
    "safari_unisolated_ot = eval(original, safari_unisolated)\n",
    "safari_unisolated_tt = eval(safari_unisolated, safari_unisolated)\n",
    "safari_unisolated_to = eval(safari_unisolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox Isolated\n",
    "firefox_isolated = pd.read_csv('./firefox/password_data_firefox_isolated_warmup/full_data_set.csv')\n",
    "firefox_isolated_ot = eval(original, firefox_isolated)\n",
    "firefox_isolated_tt = eval(firefox_isolated, firefox_isolated)\n",
    "firefox_isolated_to = eval(firefox_isolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox Uniosolated\n",
    "firefox_unisolated = pd.read_csv('./firefox/password_data_firefox_unisolated_warmup/full_data_set.csv')\n",
    "firefox_unisolated_ot = eval(original, firefox_unisolated)\n",
    "firefox_unisolated_tt = eval(firefox_unisolated, firefox_unisolated)\n",
    "firefox_unisolated_to = eval(firefox_unisolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Tor Uniosolated\n",
    "tor_unisolated = pd.read_csv('./tor/password_data_tor_unisolated_warmup/full_data_set.csv')\n",
    "tor_unisolated_ot = eval(original, tor_unisolated)\n",
    "tor_unisolated_tt = eval(tor_unisolated, tor_unisolated)\n",
    "tor_unisolated_to = eval(tor_unisolated, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting\n",
    "firefox_unisolated_resist = pd.read_csv('./firefox_resistFP/firefox_unisolated_resistFP/full_data_set.csv')\n",
    "firefox_unisolated_resist_out_ot = eval(original, firefox_unisolated_resist)\n",
    "firefox_unisolated_resist_out_tt = eval(firefox_unisolated_resist, firefox_unisolated_resist)\n",
    "firefox_unisolated_resist_out_to = eval(firefox_unisolated_resist, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 20ms\n",
    "firefox_unisolated_resist_20ms = pd.read_csv('./firefox_resistFP_20/firefox_unisolated_resistFP_20ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_20ms_out_ot = eval(original, firefox_unisolated_resist_20ms)\n",
    "firefox_unisolated_resist_20ms_out_tt = eval(firefox_unisolated_resist_20ms, firefox_unisolated_resist_20ms)\n",
    "firefox_unisolated_resist_20ms_out_to = eval(firefox_unisolated_resist_20ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 33ms\n",
    "firefox_unisolated_resist_33ms = pd.read_csv('./firefox_resistFP_33/firefox_unisolated_resistFP_33ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_33ms_out_ot = eval(original, firefox_unisolated_resist_33ms)\n",
    "firefox_unisolated_resist_33ms_out_tt = eval(firefox_unisolated_resist_33ms, firefox_unisolated_resist_33ms)\n",
    "firefox_unisolated_resist_33ms_out_to = eval(firefox_unisolated_resist_33ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 40ms\n",
    "firefox_unisolated_resist_40ms = pd.read_csv('./firefox_resistFP_40/firefox_unisolated_resistFP_40ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_40ms_out_ot = eval(original, firefox_unisolated_resist_40ms)\n",
    "firefox_unisolated_resist_40ms_out_tt = eval(firefox_unisolated_resist_40ms, firefox_unisolated_resist_40ms)\n",
    "firefox_unisolated_resist_40ms_out_to = eval(firefox_unisolated_resist_40ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 60ms\n",
    "firefox_unisolated_resist_60ms = pd.read_csv('./firefox_resistFP_60/firefox_unisolated_resistFP_60ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_60ms_out_ot = eval(original, firefox_unisolated_resist_60ms)\n",
    "firefox_unisolated_resist_60ms_out_tt = eval(firefox_unisolated_resist_60ms, firefox_unisolated_resist_60ms)\n",
    "firefox_unisolated_resist_60ms_out_to = eval(firefox_unisolated_resist_60ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 80ms\n",
    "firefox_unisolated_resist_80ms = pd.read_csv('./firefox_resistFP_80/firefox_unisolated_resistFP_80ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_80ms_out_ot = eval(original, firefox_unisolated_resist_80ms)\n",
    "firefox_unisolated_resist_80ms_out_tt = eval(firefox_unisolated_resist_80ms, firefox_unisolated_resist_80ms)\n",
    "firefox_unisolated_resist_80ms_out_to = eval(firefox_unisolated_resist_80ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 100ms\n",
    "firefox_unisolated_resist_100ms = pd.read_csv('./firefox_resistFP_100/firefox_unisolated_resistFP_100ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_100ms_out_ot = eval(original, firefox_unisolated_resist_100ms)\n",
    "firefox_unisolated_resist_100ms_out_tt = eval(firefox_unisolated_resist_100ms, firefox_unisolated_resist_100ms)\n",
    "firefox_unisolated_resist_100ms_out_to = eval(firefox_unisolated_resist_100ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 120ms\n",
    "firefox_unisolated_resist_120ms = pd.read_csv('./firefox_resistFP_120/firefox_unisolated_resistFP_120ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_120ms_out_ot = eval(original, firefox_unisolated_resist_120ms)\n",
    "firefox_unisolated_resist_120ms_out_tt = eval(firefox_unisolated_resist_120ms, firefox_unisolated_resist_120ms)\n",
    "firefox_unisolated_resist_120ms_out_to = eval(firefox_unisolated_resist_120ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 140ms\n",
    "firefox_unisolated_resist_140ms = pd.read_csv('./firefox_resistFP_140/firefox_unisolated_resistFP_140ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_140ms_out_ot = eval(original,firefox_unisolated_resist_140ms)\n",
    "firefox_unisolated_resist_140ms_out_tt = eval(firefox_unisolated_resist_140ms, firefox_unisolated_resist_140ms)\n",
    "firefox_unisolated_resist_140ms_out_to = eval(firefox_unisolated_resist_140ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 160ms\n",
    "firefox_unisolated_resist_160ms = pd.read_csv('./firefox_resistFP_160/firefox_unisolated_resistFP_160ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_160ms_out_ot = eval(original,firefox_unisolated_resist_160ms)\n",
    "firefox_unisolated_resist_160ms_out_tt = eval(firefox_unisolated_resist_160ms, firefox_unisolated_resist_160ms)\n",
    "firefox_unisolated_resist_160ms_out_to = eval(firefox_unisolated_resist_160ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 180ms\n",
    "firefox_unisolated_resist_180ms = pd.read_csv('./firefox_resistFP_180/firefox_unisolated_resistFP_180ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_180ms_out_ot = eval(original, firefox_unisolated_resist_180ms)\n",
    "firefox_unisolated_resist_180ms_out_tt = eval(firefox_unisolated_resist_180ms, firefox_unisolated_resist_180ms)\n",
    "firefox_unisolated_resist_180ms_out_to = eval(firefox_unisolated_resist_180ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 200ms\n",
    "firefox_unisolated_resist_200ms = pd.read_csv('./firefox_resistFP_200/firefox_unisolated_resistFP_200ms/full_data_set.csv')\n",
    "firefox_unisolated_resist_200ms_out_ot = eval(original, firefox_unisolated_resist_200ms)\n",
    "firefox_unisolated_resist_200ms_out_tt = eval(firefox_unisolated_resist_200ms, firefox_unisolated_resist_200ms)\n",
    "firefox_unisolated_resist_200ms_out_to = eval(firefox_unisolated_resist_200ms, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting NJ\n",
    "firefox_unisolated_resist_NJ = pd.read_csv('./firefox_resistFP_NJ/firefox_unisolated_resistFP_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_NJ_out_ot = eval(original, firefox_unisolated_resist_NJ)\n",
    "firefox_unisolated_resist_NJ_out_tt = eval(firefox_unisolated_resist_NJ, firefox_unisolated_resist_NJ)\n",
    "firefox_unisolated_resist_NJ_out_to = eval(firefox_unisolated_resist_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 20ms NJ\n",
    "firefox_unisolated_resist_20ms_NJ = pd.read_csv('./firefox_resistFP_20_NJ/firefox_unisolated_resistFP_20ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_20ms_NJ_out_ot = eval(original, firefox_unisolated_resist_20ms_NJ)\n",
    "firefox_unisolated_resist_20ms_NJ_out_tt = eval(firefox_unisolated_resist_20ms_NJ, firefox_unisolated_resist_20ms_NJ)\n",
    "firefox_unisolated_resist_20ms_NJ_out_to = eval(firefox_unisolated_resist_20ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 33ms NJ\n",
    "firefox_unisolated_resist_33ms_NJ = pd.read_csv('./firefox_resistFP_33_NJ/firefox_unisolated_resistFP_33ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_33ms_NJ_out_ot = eval(original, firefox_unisolated_resist_33ms_NJ)\n",
    "firefox_unisolated_resist_33ms_NJ_out_tt = eval(firefox_unisolated_resist_33ms_NJ, firefox_unisolated_resist_33ms_NJ)\n",
    "firefox_unisolated_resist_33ms_NJ_out_to = eval(firefox_unisolated_resist_33ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 40ms NJ\n",
    "firefox_unisolated_resist_40ms_NJ = pd.read_csv('./firefox_resistFP_40_NJ/firefox_unisolated_resistFP_40ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_40ms_NJ_out_ot = eval(original, firefox_unisolated_resist_40ms_NJ)\n",
    "firefox_unisolated_resist_40ms_NJ_out_tt = eval(firefox_unisolated_resist_40ms_NJ, firefox_unisolated_resist_40ms_NJ)\n",
    "firefox_unisolated_resist_40ms_NJ_out_to = eval(firefox_unisolated_resist_40ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 60ms NJ\n",
    "firefox_unisolated_resist_60ms_NJ = pd.read_csv('./firefox_resistFP_60_NJ/firefox_unisolated_resistFP_60ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_60ms_NJ_out_ot = eval(original, firefox_unisolated_resist_60ms_NJ)\n",
    "firefox_unisolated_resist_60ms_NJ_out_tt = eval(firefox_unisolated_resist_60ms_NJ, firefox_unisolated_resist_60ms_NJ)\n",
    "firefox_unisolated_resist_60ms_NJ_out_to = eval(firefox_unisolated_resist_60ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 80ms NJ\n",
    "firefox_unisolated_resist_80ms_NJ = pd.read_csv('./firefox_resistFP_80_NJ/firefox_unisolated_resistFP_80ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_80ms_NJ_out_ot = eval(original, firefox_unisolated_resist_80ms_NJ)\n",
    "firefox_unisolated_resist_80ms_NJ_out_tt = eval(firefox_unisolated_resist_80ms_NJ, firefox_unisolated_resist_80ms_NJ)\n",
    "firefox_unisolated_resist_80ms_NJ_out_to = eval(firefox_unisolated_resist_80ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 100ms NJ\n",
    "firefox_unisolated_resist_100ms_NJ = pd.read_csv('./firefox_resistFP_100_NJ/firefox_unisolated_resistFP_100ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_100ms_NJ_out_ot = eval(original, firefox_unisolated_resist_100ms_NJ)\n",
    "firefox_unisolated_resist_100ms_NJ_out_tt = eval(firefox_unisolated_resist_100ms_NJ, firefox_unisolated_resist_100ms_NJ)\n",
    "firefox_unisolated_resist_100ms_NJ_out_to = eval(firefox_unisolated_resist_100ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated| resistFingerprinting 120ms NJ\n",
    "firefox_unisolated_resist_120ms_NJ = pd.read_csv('./firefox_resistFP_120_NJ/firefox_unisolated_resistFP_120ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_120ms_NJ_out_ot = eval(original, firefox_unisolated_resist_120ms_NJ)\n",
    "firefox_unisolated_resist_120ms_NJ_out_tt = eval(firefox_unisolated_resist_120ms_NJ, firefox_unisolated_resist_120ms_NJ)\n",
    "firefox_unisolated_resist_120ms_NJ_out_to = eval(firefox_unisolated_resist_120ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 140ms NJ\n",
    "firefox_unisolated_resist_140ms_NJ = pd.read_csv('./firefox_resistFP_140_NJ/firefox_unisolated_resistFP_140ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_140ms_NJ_out_ot = eval(original, firefox_unisolated_resist_140ms_NJ)\n",
    "firefox_unisolated_resist_140ms_NJ_out_tt = eval(firefox_unisolated_resist_140ms_NJ, firefox_unisolated_resist_140ms_NJ)\n",
    "firefox_unisolated_resist_140ms_NJ_out_to = eval(firefox_unisolated_resist_140ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 160ms NJ\n",
    "firefox_unisolated_resist_160ms_NJ = pd.read_csv('./firefox_resistFP_160_NJ/firefox_unisolated_resistFP_160ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_160ms_NJ_out_ot = eval(original, firefox_unisolated_resist_160ms_NJ)\n",
    "firefox_unisolated_resist_160ms_NJ_out_tt = eval(firefox_unisolated_resist_160ms_NJ, firefox_unisolated_resist_160ms_NJ)\n",
    "firefox_unisolated_resist_160ms_NJ_out_to = eval(firefox_unisolated_resist_160ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 180ms NJ\n",
    "firefox_unisolated_resist_180ms_NJ = pd.read_csv('./firefox_resistFP_180_NJ/firefox_unisolated_resistFP_180ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_180ms_NJ_out_ot = eval(original, firefox_unisolated_resist_180ms_NJ)\n",
    "firefox_unisolated_resist_180ms_NJ_out_tt = eval(firefox_unisolated_resist_180ms_NJ, firefox_unisolated_resist_180ms_NJ)\n",
    "firefox_unisolated_resist_180ms_NJ_out_to = eval(firefox_unisolated_resist_180ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Firefox unisolated | resistFingerprinting 200ms NJ\n",
    "firefox_unisolated_resist_200ms_NJ = pd.read_csv('./firefox_resistFP_200_NJ/firefox_unisolated_resistFP_200ms_NJ/full_data_set.csv')\n",
    "firefox_unisolated_resist_200ms_NJ_out_ot = eval(original, firefox_unisolated_resist_200ms_NJ)\n",
    "firefox_unisolated_resist_200ms_NJ_out_tt = eval(firefox_unisolated_resist_200ms_NJ, firefox_unisolated_resist_200ms_NJ)\n",
    "firefox_unisolated_resist_200ms_NJ_out_to = eval(firefox_unisolated_resist_200ms_NJ, original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# write data | tt\n",
    "\n",
    "df_tt = {\n",
    "    \"browser\": [\n",
    "        \"Original\",\n",
    "        \"Chromium Isolated\", \"Chromium Unisolated\",\n",
    "        \"Chrome Isolated\", \"Chrome Unisolated\",\n",
    "        \"Edge Isolated\", \"Edge Unisolated\",\n",
    "        \"Firefox Isolated\", \"Firefox Unisolated\",\n",
    "        \"Safari Isolated\", \"Safari Unisolated\",\n",
    "        \"Tor Isolated\",\n",
    "\n",
    "        \"Firefox RF\", \"Firefox RF 20ms\",\n",
    "        \"Firefox RF 33ms\", \"Firefox RF 40ms\",\n",
    "        \"Firefox RF 60ms\", \"Firefox RF 80ms\",\n",
    "        \"Firefox RF 100ms\", \"Firefox RF 120ms\",\n",
    "        \"Firefox RF 140ms\", \"Firefox RF 160ms\",\n",
    "        \"Firefox RF 180ms\", \"Firefox RF 200ms\",\n",
    "\n",
    "        \"Firefox RF NJ\", \"Firefox RF NJ 20ms\",\n",
    "        \"Firefox RF NJ 33ms\", \"Firefox RF NJ 40ms\",\n",
    "        \"Firefox RF NJ 60ms\", \"Firefox RF NJ 80ms\",\n",
    "        \"Firefox RF NJ 100ms\", \"Firefox RF NJ 120ms\",\n",
    "        \"Firefox RF NJ 140ms\", \"Firefox RF NJ 160ms\",\n",
    "        \"Firefox RF NJ 180ms\", \"Firefox RF NJ 200ms\",\n",
    "    ],\n",
    "    \"Euclidean(mean)\": [\n",
    "        original_out.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        chromium_isolated_tt.loc[\"Euclidean\"][\"err.mean\"], chromium_unisolated_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        chrome_isolated_tt.loc[\"Euclidean\"][\"err.mean\"], chrome_unisolated_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        edge_isolated_tt.loc[\"Euclidean\"][\"err.mean\"], edge_unisolated_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_isolated_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        safari_isolated_tt.loc[\"Euclidean\"][\"err.mean\"], safari_unisolated_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        tor_unisolated_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_tt.loc[\"Euclidean\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Euclidean(std)\": [\n",
    "        original_out.loc[\"Euclidean\"][\"err.std\"],\n",
    "        chromium_isolated_tt.loc[\"Euclidean\"][\"err.std\"], chromium_unisolated_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        chrome_isolated_tt.loc[\"Euclidean\"][\"err.std\"], chrome_unisolated_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        edge_isolated_tt.loc[\"Euclidean\"][\"err.std\"], edge_unisolated_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_isolated_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        safari_isolated_tt.loc[\"Euclidean\"][\"err.std\"], safari_unisolated_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        tor_unisolated_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_20ms_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_40ms_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_80ms_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_120ms_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_160ms_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_200ms_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_tt.loc[\"Euclidean\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"Manhattan(mean)\": [\n",
    "        original_out.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        chromium_isolated_tt.loc[\"Manhattan\"][\"err.mean\"], chromium_unisolated_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        chrome_isolated_tt.loc[\"Manhattan\"][\"err.mean\"], chrome_unisolated_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        edge_isolated_tt.loc[\"Manhattan\"][\"err.mean\"], edge_unisolated_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_isolated_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        safari_isolated_tt.loc[\"Manhattan\"][\"err.mean\"], safari_unisolated_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        tor_unisolated_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_tt.loc[\"Manhattan\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Manhattan(std)\": [\n",
    "        original_out.loc[\"Manhattan\"][\"err.std\"],\n",
    "        chromium_isolated_tt.loc[\"Manhattan\"][\"err.std\"], chromium_unisolated_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        chrome_isolated_tt.loc[\"Manhattan\"][\"err.std\"], chrome_unisolated_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        edge_isolated_tt.loc[\"Manhattan\"][\"err.std\"], edge_unisolated_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_isolated_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        safari_isolated_tt.loc[\"Manhattan\"][\"err.std\"], safari_unisolated_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        tor_unisolated_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_20ms_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_40ms_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_80ms_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_120ms_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_160ms_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_200ms_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_tt.loc[\"Manhattan\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"ManhattanScaled(mean)\": [\n",
    "        original_out.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        chromium_isolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"], chromium_unisolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        chrome_isolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"], chrome_unisolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        edge_isolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"], edge_unisolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_isolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        safari_isolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"], safari_unisolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        tor_unisolated_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"ManhattanScaled(std)\": [\n",
    "        original_out.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        chromium_isolated_tt.loc[\"ManhattanScaled\"][\"err.std\"], chromium_unisolated_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        chrome_isolated_tt.loc[\"ManhattanScaled\"][\"err.std\"], chrome_unisolated_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        edge_isolated_tt.loc[\"ManhattanScaled\"][\"err.std\"], edge_unisolated_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_isolated_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        safari_isolated_tt.loc[\"ManhattanScaled\"][\"err.std\"], safari_unisolated_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        tor_unisolated_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_20ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_40ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_80ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_120ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_160ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_200ms_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_tt.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"Mahalanobis(mean)\": [\n",
    "        original_out.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        chromium_isolated_tt.loc[\"Mahalanobis\"][\"err.mean\"], chromium_unisolated_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        chrome_isolated_tt.loc[\"Mahalanobis\"][\"err.mean\"], chrome_unisolated_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        edge_isolated_tt.loc[\"Mahalanobis\"][\"err.mean\"], edge_unisolated_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_isolated_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        safari_isolated_tt.loc[\"Mahalanobis\"][\"err.mean\"], safari_unisolated_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        tor_unisolated_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_tt.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Mahalanobis(std)\": [\n",
    "        original_out.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        chromium_isolated_tt.loc[\"Mahalanobis\"][\"err.std\"], chromium_unisolated_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        chrome_isolated_tt.loc[\"Mahalanobis\"][\"err.std\"], chrome_unisolated_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        edge_isolated_tt.loc[\"Mahalanobis\"][\"err.std\"], edge_unisolated_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_isolated_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        safari_isolated_tt.loc[\"Mahalanobis\"][\"err.std\"], safari_unisolated_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        tor_unisolated_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_20ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_40ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_80ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_120ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_160ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_200ms_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_tt.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "    ],\n",
    "}\n",
    "\n",
    "df = pd.DataFrame(df_tt, copy=False)\n",
    "df.to_csv(\"./password_browser_browser.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# write data | to\n",
    "\n",
    "df_to = {\n",
    "    \"browser\": [\n",
    "        \"Original\",\n",
    "        \"Chromium Isolated\", \"Chromium Unisolated\",\n",
    "        \"Chrome Isolated\", \"Chrome Unisolated\",\n",
    "        \"Edge Isolated\", \"Edge Unisolated\",\n",
    "        \"Firefox Isolated\", \"Firefox Unisolated\",\n",
    "        \"Safari Isolated\", \"Safari Unisolated\",\n",
    "        \"Tor Isolated\",\n",
    "\n",
    "        \"Firefox RF\", \"Firefox RF 20ms\",\n",
    "        \"Firefox RF 33ms\", \"Firefox RF 40ms\",\n",
    "        \"Firefox RF 60ms\", \"Firefox RF 80ms\",\n",
    "        \"Firefox RF 100ms\", \"Firefox RF 120ms\",\n",
    "        \"Firefox RF 140ms\", \"Firefox RF 160ms\",\n",
    "        \"Firefox RF 180ms\", \"Firefox RF 200ms\",\n",
    "\n",
    "        \"Firefox RF NJ\", \"Firefox RF NJ 20ms\",\n",
    "        \"Firefox RF NJ 33ms\", \"Firefox RF NJ 40ms\",\n",
    "        \"Firefox RF NJ 60ms\", \"Firefox RF NJ 80ms\",\n",
    "        \"Firefox RF NJ 100ms\", \"Firefox RF NJ 120ms\",\n",
    "        \"Firefox RF NJ 140ms\", \"Firefox RF NJ 160ms\",\n",
    "        \"Firefox RF NJ 180ms\", \"Firefox RF NJ 200ms\",\n",
    "    ],\n",
    "    \"Euclidean(mean)\": [\n",
    "        original_out.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        chromium_isolated_to.loc[\"Euclidean\"][\"err.mean\"], chromium_unisolated_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        chrome_isolated_to.loc[\"Euclidean\"][\"err.mean\"], chrome_unisolated_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        edge_isolated_to.loc[\"Euclidean\"][\"err.mean\"], edge_unisolated_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_isolated_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        safari_isolated_to.loc[\"Euclidean\"][\"err.mean\"], safari_unisolated_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        tor_unisolated_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_to.loc[\"Euclidean\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Euclidean(std)\": [\n",
    "        original_out.loc[\"Euclidean\"][\"err.std\"],\n",
    "        chromium_isolated_to.loc[\"Euclidean\"][\"err.std\"], chromium_unisolated_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        chrome_isolated_to.loc[\"Euclidean\"][\"err.std\"], chrome_unisolated_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        edge_isolated_to.loc[\"Euclidean\"][\"err.std\"], edge_unisolated_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_isolated_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        safari_isolated_to.loc[\"Euclidean\"][\"err.std\"], safari_unisolated_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        tor_unisolated_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_20ms_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_40ms_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_80ms_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_120ms_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_160ms_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_200ms_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_to.loc[\"Euclidean\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"Manhattan(mean)\": [\n",
    "        original_out.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        chromium_isolated_to.loc[\"Manhattan\"][\"err.mean\"], chromium_unisolated_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        chrome_isolated_to.loc[\"Manhattan\"][\"err.mean\"], chrome_unisolated_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        edge_isolated_to.loc[\"Manhattan\"][\"err.mean\"], edge_unisolated_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_isolated_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        safari_isolated_to.loc[\"Manhattan\"][\"err.mean\"], safari_unisolated_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        tor_unisolated_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_to.loc[\"Manhattan\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Manhattan(std)\": [\n",
    "        original_out.loc[\"Manhattan\"][\"err.std\"],\n",
    "        chromium_isolated_to.loc[\"Manhattan\"][\"err.std\"], chromium_unisolated_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        chrome_isolated_to.loc[\"Manhattan\"][\"err.std\"], chrome_unisolated_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        edge_isolated_to.loc[\"Manhattan\"][\"err.std\"], edge_unisolated_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_isolated_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        safari_isolated_to.loc[\"Manhattan\"][\"err.std\"], safari_unisolated_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        tor_unisolated_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_20ms_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_40ms_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_80ms_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_120ms_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_160ms_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_200ms_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_to.loc[\"Manhattan\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"ManhattanScaled(mean)\": [\n",
    "        original_out.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        chromium_isolated_to.loc[\"ManhattanScaled\"][\"err.mean\"], chromium_unisolated_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        chrome_isolated_to.loc[\"ManhattanScaled\"][\"err.mean\"], chrome_unisolated_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        edge_isolated_to.loc[\"ManhattanScaled\"][\"err.mean\"], edge_unisolated_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_isolated_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        safari_isolated_to.loc[\"ManhattanScaled\"][\"err.mean\"], safari_unisolated_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        tor_unisolated_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"ManhattanScaled(std)\": [\n",
    "        original_out.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        chromium_isolated_to.loc[\"ManhattanScaled\"][\"err.std\"], chromium_unisolated_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        chrome_isolated_to.loc[\"ManhattanScaled\"][\"err.std\"], chrome_unisolated_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        edge_isolated_to.loc[\"ManhattanScaled\"][\"err.std\"], edge_unisolated_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_isolated_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        safari_isolated_to.loc[\"ManhattanScaled\"][\"err.std\"], safari_unisolated_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        tor_unisolated_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_20ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_40ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_80ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_120ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_160ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_200ms_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_to.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"Mahalanobis(mean)\": [\n",
    "        original_out.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        chromium_isolated_to.loc[\"Mahalanobis\"][\"err.mean\"], chromium_unisolated_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        chrome_isolated_to.loc[\"Mahalanobis\"][\"err.mean\"], chrome_unisolated_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        edge_isolated_to.loc[\"Mahalanobis\"][\"err.mean\"], edge_unisolated_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_isolated_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        safari_isolated_to.loc[\"Mahalanobis\"][\"err.mean\"], safari_unisolated_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        tor_unisolated_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_to.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Mahalanobis(std)\": [\n",
    "        original_out.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        chromium_isolated_to.loc[\"Mahalanobis\"][\"err.std\"], chromium_unisolated_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        chrome_isolated_to.loc[\"Mahalanobis\"][\"err.std\"], chrome_unisolated_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        edge_isolated_to.loc[\"Mahalanobis\"][\"err.std\"], edge_unisolated_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_isolated_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        safari_isolated_to.loc[\"Mahalanobis\"][\"err.std\"], safari_unisolated_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        tor_unisolated_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_20ms_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_40ms_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_80ms_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_120ms_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_160ms_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_200ms_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_to.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "    ],\n",
    "}\n",
    "\n",
    "df = pd.DataFrame(df_to, copy=False)\n",
    "df.to_csv(\"./password_browser_original.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# write data | ot\n",
    "\n",
    "df_ot = {\n",
    "    \"browser\": [\n",
    "        \"Original\",\n",
    "        \"Chromium Isolated\", \"Chromium Unisolated\",\n",
    "        \"Chrome Isolated\", \"Chrome Unisolated\",\n",
    "        \"Edge Isolated\", \"Edge Unisolated\",\n",
    "        \"Firefox Isolated\", \"Firefox Unisolated\",\n",
    "        \"Safari Isolated\", \"Safari Unisolated\",\n",
    "        \"Tor Isolated\",\n",
    "\n",
    "        \"Firefox RF\", \"Firefox RF 20ms\",\n",
    "        \"Firefox RF 33ms\", \"Firefox RF 40ms\",\n",
    "        \"Firefox RF 60ms\", \"Firefox RF 80ms\",\n",
    "        \"Firefox RF 100ms\", \"Firefox RF 120ms\",\n",
    "        \"Firefox RF 140ms\", \"Firefox RF 160ms\",\n",
    "        \"Firefox RF 180ms\", \"Firefox RF 200ms\",\n",
    "\n",
    "        \"Firefox RF NJ\", \"Firefox RF NJ 20ms\",\n",
    "        \"Firefox RF NJ 33ms\", \"Firefox RF NJ 40ms\",\n",
    "        \"Firefox RF NJ 60ms\", \"Firefox RF NJ 80ms\",\n",
    "        \"Firefox RF NJ 100ms\", \"Firefox RF NJ 120ms\",\n",
    "        \"Firefox RF NJ 140ms\", \"Firefox RF NJ 160ms\",\n",
    "        \"Firefox RF NJ 180ms\", \"Firefox RF NJ 200ms\",\n",
    "    ],\n",
    "    \"Euclidean(mean)\": [\n",
    "        original_out.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        chromium_isolated_ot.loc[\"Euclidean\"][\"err.mean\"], chromium_unisolated_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        chrome_isolated_ot.loc[\"Euclidean\"][\"err.mean\"], chrome_unisolated_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        edge_isolated_ot.loc[\"Euclidean\"][\"err.mean\"], edge_unisolated_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_isolated_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        safari_isolated_ot.loc[\"Euclidean\"][\"err.mean\"], safari_unisolated_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        tor_unisolated_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_ot.loc[\"Euclidean\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Euclidean(std)\": [\n",
    "        original_out.loc[\"Euclidean\"][\"err.std\"],\n",
    "        chromium_isolated_ot.loc[\"Euclidean\"][\"err.std\"], chromium_unisolated_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        chrome_isolated_ot.loc[\"Euclidean\"][\"err.std\"], chrome_unisolated_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        edge_isolated_ot.loc[\"Euclidean\"][\"err.std\"], edge_unisolated_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_isolated_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        safari_isolated_ot.loc[\"Euclidean\"][\"err.std\"], safari_unisolated_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        tor_unisolated_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_20ms_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_40ms_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_80ms_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_120ms_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_160ms_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_200ms_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_ot.loc[\"Euclidean\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"Manhattan(mean)\": [\n",
    "        original_out.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        chromium_isolated_ot.loc[\"Manhattan\"][\"err.mean\"], chromium_unisolated_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        chrome_isolated_ot.loc[\"Manhattan\"][\"err.mean\"], chrome_unisolated_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        edge_isolated_ot.loc[\"Manhattan\"][\"err.mean\"], edge_unisolated_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_isolated_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        safari_isolated_ot.loc[\"Manhattan\"][\"err.mean\"], safari_unisolated_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        tor_unisolated_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_ot.loc[\"Manhattan\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Manhattan(std)\": [\n",
    "        original_out.loc[\"Manhattan\"][\"err.std\"],\n",
    "        chromium_isolated_ot.loc[\"Manhattan\"][\"err.std\"], chromium_unisolated_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        chrome_isolated_ot.loc[\"Manhattan\"][\"err.std\"], chrome_unisolated_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        edge_isolated_ot.loc[\"Manhattan\"][\"err.std\"], edge_unisolated_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_isolated_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        safari_isolated_ot.loc[\"Manhattan\"][\"err.std\"], safari_unisolated_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        tor_unisolated_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_20ms_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_40ms_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_80ms_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_120ms_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_160ms_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_200ms_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_ot.loc[\"Manhattan\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"ManhattanScaled(mean)\": [\n",
    "        original_out.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        chromium_isolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"], chromium_unisolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        chrome_isolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"], chrome_unisolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        edge_isolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"], edge_unisolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_isolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        safari_isolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"], safari_unisolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        tor_unisolated_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"ManhattanScaled(std)\": [\n",
    "        original_out.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        chromium_isolated_ot.loc[\"ManhattanScaled\"][\"err.std\"], chromium_unisolated_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        chrome_isolated_ot.loc[\"ManhattanScaled\"][\"err.std\"], chrome_unisolated_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        edge_isolated_ot.loc[\"ManhattanScaled\"][\"err.std\"], edge_unisolated_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_isolated_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        safari_isolated_ot.loc[\"ManhattanScaled\"][\"err.std\"], safari_unisolated_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        tor_unisolated_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_20ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_40ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_80ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_120ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_160ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_200ms_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_ot.loc[\"ManhattanScaled\"][\"err.std\"],\n",
    "    ],\n",
    "    \n",
    "    \"Mahalanobis(mean)\": [\n",
    "        original_out.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        chromium_isolated_ot.loc[\"Mahalanobis\"][\"err.mean\"], chromium_unisolated_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        chrome_isolated_ot.loc[\"Mahalanobis\"][\"err.mean\"], chrome_unisolated_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        edge_isolated_ot.loc[\"Mahalanobis\"][\"err.mean\"], edge_unisolated_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_isolated_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        safari_isolated_ot.loc[\"Mahalanobis\"][\"err.mean\"], safari_unisolated_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        tor_unisolated_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_20ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_40ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_80ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_120ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_160ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_200ms_out_ot.loc[\"ManhattanScaled\"][\"err.mean\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_20ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_40ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_80ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_120ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_160ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"], firefox_unisolated_resist_200ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.mean\"],\n",
    "    ],\n",
    "    \"Mahalanobis(std)\": [\n",
    "        original_out.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        chromium_isolated_ot.loc[\"Mahalanobis\"][\"err.std\"], chromium_unisolated_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        chrome_isolated_ot.loc[\"Mahalanobis\"][\"err.std\"], chrome_unisolated_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        edge_isolated_ot.loc[\"Mahalanobis\"][\"err.std\"], edge_unisolated_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_isolated_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        safari_isolated_ot.loc[\"Mahalanobis\"][\"err.std\"], safari_unisolated_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        tor_unisolated_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_20ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_40ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_80ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_120ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_160ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_200ms_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "\n",
    "        firefox_unisolated_resist_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_20ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_33ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_40ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_60ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_80ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_100ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_120ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_140ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_160ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "        firefox_unisolated_resist_180ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"], firefox_unisolated_resist_200ms_NJ_out_ot.loc[\"Mahalanobis\"][\"err.std\"],\n",
    "    ],\n",
    "}\n",
    "\n",
    "df = pd.DataFrame(df_ot, copy=False)\n",
    "df.to_csv(\"./password_original_browser.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plot data\n",
    "# plot authentification\n",
    "x = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]\n",
    "labels = [\n",
    "    'Original', '16.66ms\\n(Default)', \n",
    "    '20ms', \n",
    "    #'33.33ms', \n",
    "    '40ms', '60ms', \n",
    "    '80ms', '100ms', \n",
    "    '120ms',  '140ms', \n",
    "    '160ms', '180ms',\n",
    "    '200ms'\n",
    "]\n",
    "categories = [\n",
    "    \"Enrollment: Original, Test: Modified\"\n",
    "    \"Enrollment: Modified, Test: Modified\"\n",
    "    \"Enrollment: Modified, Test: Original\"\n",
    "]\n",
    "\n",
    "y_ticks = np.arange(0,110,10)\n",
    "\n",
    "# ot\n",
    "\n",
    "y_euclidean_ot = [\n",
    "    original_out.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_ot.iloc[0]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_manhattan_ot = [\n",
    "    original_out.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_ot.iloc[1]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_manhattan_scaled_ot = [\n",
    "    original_out.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_ot.iloc[2]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_mahalanobis_ot = [\n",
    "    original_out.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_ot.iloc[3]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "# tt\n",
    "\n",
    "y_euclidean_tt = [\n",
    "    original_out.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_tt.iloc[0]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_manhattan_tt = [\n",
    "    original_out.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_tt.iloc[1]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_manhattan_scaled_tt = [\n",
    "    original_out.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_tt.iloc[2]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_mahalanobis_tt = [\n",
    "    original_out.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_tt.iloc[3]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "# to\n",
    "\n",
    "y_euclidean_to = [\n",
    "    original_out.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_to.iloc[0]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_manhattan_to = [\n",
    "    original_out.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_to.iloc[1]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_manhattan_scaled_to = [\n",
    "    original_out.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_to.iloc[2]['err.mean'] * 100,\n",
    "]\n",
    "\n",
    "y_mahalanobis_to = [\n",
    "    original_out.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_20ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    #firefox_unisolated_resist_33ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_40ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_60ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_80ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_100ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_120ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_140ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_160ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_180ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "    firefox_unisolated_resist_200ms_out_to.iloc[3]['err.mean'] * 100,\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plot | legend right\n",
    "\n",
    "font_size = 26\n",
    "font_size_ticks = font_size- 2\n",
    "\n",
    "x = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]\n",
    "labels = [\n",
    "    'Orig.', '16.66\\n(Default)', \n",
    "    '20', \n",
    "    #'33.33ms', \n",
    "    '40', '60', \n",
    "    '80', '100', \n",
    "    '120', '140', \n",
    "    '160', '180',\n",
    "    '200'\n",
    "]\n",
    "y_ticks = np.arange(0,120,20)\n",
    "\n",
    "categories = [\"Euclidean\",\"Manhattan\",\"Manhattan Scaled\",\"Mahalanobis\"]\n",
    "\n",
    "figure = plt.figure(layout='constrained', figsize=(28, 5))\n",
    "ax = figure.add_subplot(111)\n",
    "\n",
    "linewidth = 4.0\n",
    "\n",
    "p1, = ax.plot(x, y_euclidean_ot,  label='',color=\"violet\", linewidth=linewidth)\n",
    "p2, = ax.plot(x, y_manhattan_ot,  label='', color=\"gray\", linewidth=linewidth)\n",
    "p3, = ax.plot(x, y_manhattan_scaled_ot, label='', color=\"green\", linewidth=linewidth)\n",
    "p4, = ax.plot(x, y_mahalanobis_ot, label='', color=\"blue\", linewidth=linewidth)\n",
    "\n",
    "p5, = ax.plot(x, y_euclidean_tt,  label='', linestyle=':',color=\"violet\", linewidth=linewidth)\n",
    "p6, = ax.plot(x, y_manhattan_tt,  label='', linestyle=':', color=\"gray\", linewidth=linewidth)\n",
    "p7, = ax.plot(x, y_manhattan_scaled_tt, label='', linestyle=':', color=\"green\", linewidth=linewidth)\n",
    "p8, = ax.plot(x, y_mahalanobis_tt, label='', linestyle=':', color=\"blue\", linewidth=linewidth)\n",
    "\n",
    "p9, = ax.plot(x, y_euclidean_to,  label='', linestyle='dashed', color=\"violet\", linewidth=linewidth)\n",
    "p10, = ax.plot(x, y_manhattan_to,  label='', linestyle='dashed', color=\"gray\", linewidth=linewidth)\n",
    "p11, = ax.plot(x, y_manhattan_scaled_to, label='', linestyle='dashed', color=\"green\", linewidth=linewidth)\n",
    "p12, = ax.plot(x, y_mahalanobis_to, label='', linestyle='dashed', color=\"blue\", linewidth=linewidth)\n",
    "\n",
    "\n",
    "# create blank rectangle\n",
    "extra = Rectangle((0, 0), 1, 1, fc=\"w\", fill=False, edgecolor='none', linewidth=0)\n",
    "\n",
    "# create organized list containing all handles for table. Extra represent empty space\n",
    "legend_handle = [\n",
    "    extra,extra,extra,extra,extra,\n",
    "    extra,extra,extra,extra,extra,\n",
    "    extra,extra,extra,extra,extra, extra,p1, p2, p3, p4, extra, p5, p6, p7, p8, extra, p9, p10, p11, p12]\n",
    "\n",
    "# define the labels\n",
    "labels_legend = [\n",
    "    r\"$\\bf{Enrollment}$: Original, $\\bf{Test}$: Modified\", \"   Euclidean\", \"   Manhattan\", \"   Manhattan Scaled\", \"   Mahalanobis\", \n",
    "    r\"$\\bf{Enrollment}$: Modified, $\\bf{Test}$: Modified\", \"   Euclidean\", \"   Manhattan\", \"   Manhattan Scaled\", \"   Mahalanobis\", \n",
    "    r\"$\\bf{Enrollment}$: Modified, $\\bf{Test}$: Original\", \"   Euclidean\", \"   Manhattan\", \"   Manhattan Scaled\", \"   Mahalanobis\",\n",
    "    ]\n",
    "label_empty = [\" \", \" \", \" \", \" \", \" \"]\n",
    "\n",
    "# organize labels for table construction\n",
    "legend_labels = np.concatenate([labels_legend, label_empty * 3])\n",
    "\n",
    "# create legend\n",
    "figure.legend(legend_handle, legend_labels, ncol = 2, loc='outside right center', handletextpad = -0.5, fontsize=font_size_ticks, frameon=False)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "plt.title(\"\", fontsize=font_size)\n",
    "plt.xlabel(\"Timer precision setting for Resist Fingerprinting [ms]\", fontsize=font_size)\n",
    "plt.ylabel(\"Equal error rate [%]\", fontsize=font_size)\n",
    "plt.xticks(x, labels, rotation=0, fontsize=font_size_ticks)\n",
    "plt.yticks(y_ticks, fontsize=font_size_ticks)\n",
    "plt.grid(axis=\"y\", alpha=0.4)\n",
    "plt.savefig('./password_firefox.pdf', bbox_inches=\"tight\", pad_inches=0.1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plot | legend below\n",
    "\n",
    "font_size = 20\n",
    "font_size_ticks = font_size- 2\n",
    "\n",
    "x = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]\n",
    "labels = [\n",
    "    'Orig.', '16.66\\n(Default)', \n",
    "    '20', \n",
    "    #'33.33ms', \n",
    "    '40', '60', \n",
    "    '80', '100', \n",
    "    '120', '140', \n",
    "    '160', '180',\n",
    "    '200'\n",
    "]\n",
    "y_ticks = np.arange(0,120,20)\n",
    "\n",
    "categories = [\"Euclidean\",\"Manhattan\",\"Manhattan Scaled\",\"Mahalanobis\"]\n",
    "\n",
    "figure = plt.figure(layout='constrained', figsize=(18, 5))\n",
    "ax = figure.add_subplot(111)\n",
    "\n",
    "linewidth = 4.0\n",
    "\n",
    "p1, = ax.plot(x, y_euclidean_ot,  label='',color=\"violet\", linewidth=linewidth)\n",
    "p2, = ax.plot(x, y_manhattan_ot,  label='', color=\"gray\", linewidth=linewidth)\n",
    "p3, = ax.plot(x, y_manhattan_scaled_ot, label='', color=\"green\", linewidth=linewidth)\n",
    "p4, = ax.plot(x, y_mahalanobis_ot, label='', color=\"blue\", linewidth=linewidth)\n",
    "\n",
    "p5, = ax.plot(x, y_euclidean_tt,  label='', linestyle=':',color=\"violet\", linewidth=linewidth)\n",
    "p6, = ax.plot(x, y_manhattan_tt,  label='', linestyle=':', color=\"gray\", linewidth=linewidth)\n",
    "p7, = ax.plot(x, y_manhattan_scaled_tt, label='', linestyle=':', color=\"green\", linewidth=linewidth)\n",
    "p8, = ax.plot(x, y_mahalanobis_tt, label='', linestyle=':', color=\"blue\", linewidth=linewidth)\n",
    "\n",
    "p9, = ax.plot(x, y_euclidean_to,  label='', linestyle='dashed', color=\"violet\", linewidth=linewidth)\n",
    "p10, = ax.plot(x, y_manhattan_to,  label='', linestyle='dashed', color=\"gray\", linewidth=linewidth)\n",
    "p11, = ax.plot(x, y_manhattan_scaled_to, label='', linestyle='dashed', color=\"green\", linewidth=linewidth)\n",
    "p12, = ax.plot(x, y_mahalanobis_to, label='', linestyle='dashed', color=\"blue\", linewidth=linewidth)\n",
    "\n",
    "\n",
    "# create blank rectangle\n",
    "extra = Rectangle((0, 0), 0, 0, fc=\"w\", fill=False, edgecolor='none', linewidth=0)\n",
    "\n",
    "# create organized list containing all handles for table. Extra represent empty space\n",
    "legend_handle = [\n",
    "    extra,p1,p2,p3,p4,\n",
    "    extra,p5,p6,p7,p8,\n",
    "    extra,p9,p10,p11,p12]\n",
    "\n",
    "# define the labels\n",
    "labels_legend = [\n",
    "    r\"$\\bf{Enrollment}$: Original, \" + \"\\n\" + r\"$\\bf{Test}$: Modified\", \"Euclidean\", \"Manhattan\", \"Manhattan Scaled\", \"Mahalanobis\",\n",
    "    r\"$\\bf{Enrollment}$: Modified, \" + \"\\n\" + r\"$\\bf{Test}$: Modified\", \"Euclidean\", \"Manhattan\", \"Manhattan Scaled\", \"Mahalanobis\",\n",
    "    r\"$\\bf{Enrollment}$: Modified, \" + \"\\n\" + r\"$\\bf{Test}$: Original\", \"Euclidean\", \"Manhattan\", \"Manhattan Scaled\", \"Mahalanobis\",\n",
    "]\n",
    "\n",
    "# organize labels for table construction\n",
    "legend_labels = np.concatenate([labels_legend])\n",
    "\n",
    "# create legend\n",
    "figure.legend(legend_handle, legend_labels, ncol = 3, loc='outside lower center', fontsize=font_size, frameon=False, bbox_to_anchor=(0.5, -0.52))\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "plt.title(\"\", fontsize=font_size)\n",
    "plt.xlabel(\"Timer precision setting for Resist Fingerprinting [ms]\", fontsize=font_size)\n",
    "plt.ylabel(\"Equal error rate [%]\", fontsize=font_size)\n",
    "plt.xticks(x, labels, rotation=0, fontsize=font_size_ticks)\n",
    "plt.yticks(y_ticks, fontsize=font_size_ticks)\n",
    "plt.grid(axis=\"y\", alpha=0.4)\n",
    "plt.savefig('./password_firefox2.pdf', bbox_inches=\"tight\", pad_inches=0.1)\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.0"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}