examples/Naive_Bayes_Iris_demo/PyDP_Naive_Bayes_Comparison.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "altered-forwarding",
   "metadata": {},
   "source": [
    "## Comparing results on various datasets using PyDP Naive Bayes vs Scikit-Learn's Naive Bayes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "wicked-salmon",
   "metadata": {},
   "source": [
    "This notebook has implemented both the Naive Bayes algorithm on various datasets such as \"adult\", \"mushroom\", \"nursery\", \"digit\" and \"skin\".\n",
    "\n",
    "Different Training and Testing Accuracies along with the plots aids in understanding the difference in results.\n",
    "\n",
    "\n",
    "\n",
    "Sources:\n",
    "- https://github.com/OpenMined/PyDP/blob/dev/src/pydp/ml/naive_bayes.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "instrumental-characteristic",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "sns.set_theme(style=\"white\")\n",
    "\n",
    "np.random.seed(0)\n",
    "\n",
    "from sklearn.model_selection import KFold"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "spatial-blame",
   "metadata": {},
   "source": [
    "## Datasets\n",
    "\n",
    "Load all the datasets i.e. \"adult\", \"mushroom\", \"nursery\", \"digit\", \"skin\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "excessive-sentence",
   "metadata": {},
   "outputs": [],
   "source": [
    "import data_utils\n",
    "\n",
    "datasets = {\n",
    "    \"adult\": data_utils.load_adult_dataset(),\n",
    "    \"mushroom\": data_utils.load_mushroom_dataset(),\n",
    "    \"nursery\": data_utils.load_nursery_dataset(),\n",
    "    \"digit\": data_utils.load_digit_dataset(),\n",
    "    \"skin\": data_utils.load_skin_dataset(),\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "genuine-withdrawal",
   "metadata": {},
   "source": [
    "## Finding average training and testing accuracies of Scikit-learn's Gaussian Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ordinary-vision",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/.cache/pypoetry/virtualenvs/pydp-FONabfq3-py3.6/lib/python3.6/site-packages/sklearn/model_selection/_split.py:297: FutureWarning: Setting a random_state has no effect since shuffle is False. This will raise an error in 0.24. You should leave random_state to its default (None), or set shuffle=True.\n",
      "  FutureWarning\n"
     ]
    }
   ],
   "source": [
    "# Train GaussianNB from Sklearn for basline 10-fold average training and testing accuracies\n",
    "\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "\n",
    "clf = GaussianNB()\n",
    "\n",
    "kfold = KFold(n_splits=10, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "tamil-mauritius",
   "metadata": {},
   "source": [
    "``kfold``: The data is split using scikit-learn's K-Folds cross validator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "double-category",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.7951810130150003\n",
      "Average Testing Accuracy\n",
      "0.7953074892935194\n",
      "Average Training Accuracy\n",
      "0.965356642758566\n",
      "Average Testing Accuracy\n",
      "0.9367470719042167\n",
      "Average Training Accuracy\n",
      "0.83596536351166\n",
      "Average Testing Accuracy\n",
      "0.7715277777777778\n",
      "Average Training Accuracy\n",
      "0.7993080268880981\n",
      "Average Testing Accuracy\n",
      "0.7909252669039146\n",
      "Average Training Accuracy\n",
      "0.9253783621397431\n",
      "Average Testing Accuracy\n",
      "0.8928012046502467\n"
     ]
    }
   ],
   "source": [
    "def get_fit_result(clf, kfold, df, X, y):\n",
    "    \n",
    "    training_accuracies = []\n",
    "    testing_accuracies = []\n",
    "\n",
    "    for train, test in kfold.split(df.values):\n",
    "\n",
    "        X_train = X.values[train]\n",
    "        y_train = y.values[train]\n",
    "\n",
    "        X_test = X.values[test]\n",
    "        y_test = y.values[test]\n",
    "\n",
    "        clf.fit(X_train, y_train)\n",
    "\n",
    "        training_accuracy = clf.score(X_train, y_train)\n",
    "        testing_accuracy = clf.score(X_test, y_test)\n",
    "\n",
    "        training_accuracies.append(training_acucracy)\n",
    "        testing_accuracies.append(testing_acucracy)\n",
    "\n",
    "    mean_train_acc = np.mean(training_accuracies)\n",
    "    mean_test_acc = np.mean(testing_accuracies)\n",
    "    \n",
    "    print(\"Average Training Accuracy\")\n",
    "    print(mean_train_acc)\n",
    "    print(\"Average Testing Accuracy\")\n",
    "    print(mean_test_acc)\n",
    "    \n",
    "    return mean_train_acc , mean_test_acc\n",
    "\n",
    "baselines = {}\n",
    "\n",
    "for name, dataset in datasets.items():\n",
    "    baseline_result = get_fit_result(clf, kfold, dataset[0], dataset[1], dataset[2])\n",
    "    baselines[name] = baseline_result"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "incident-voice",
   "metadata": {},
   "source": [
    "``clf``: Gaussian Naive Bayes from scikit-learn is used as a classifier on five different datasets\n",
    "\n",
    "``training_accuracies``: A dict containing all accuracies of a k-fold cross validator on five different datasets\n",
    "\n",
    "``testing_accuracies``: A dict containing all accuracies of a k-fold cross validator on five different datasets\n",
    "\n",
    "``mean_train_acc``: Average of training accuracy of a particular dataset\n",
    "\n",
    "``mean_test_acc``: Average of testing accuracy of a particular dataset\n",
    "\n",
    "``baseline_result``: Stores a tuple containing average training and testing accuracies\n",
    "\n",
    "``baselines``: A dict containing names of five different datasets and their corresponding training and testing accuracies as a tuple"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "surprising-venice",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.758798009399016\n",
      "Average Testing Accuracy\n",
      "0.7583307621686498\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.6911780070076436\n",
      "Average Testing Accuracy\n",
      "0.6914400961376707\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.6901169080234085\n",
      "Average Testing Accuracy\n",
      "0.6876304976090054\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.6233382362524147\n",
      "Average Testing Accuracy\n",
      "0.6268954450025988\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.3987363325774006\n",
      "Average Testing Accuracy\n",
      "0.4002249831962758\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.5204769154304216\n",
      "Average Testing Accuracy\n",
      "0.5171452301940482\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.402092611837536\n",
      "Average Testing Accuracy\n",
      "0.4049677919189739\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.8886817195215866\n",
      "Average Testing Accuracy\n",
      "0.7972282309029988\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.7093375079726092\n",
      "Average Testing Accuracy\n",
      "0.5656341834354304\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.4725374065397694\n",
      "Average Testing Accuracy\n",
      "0.4938155526875466\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.5275844329622909\n",
      "Average Testing Accuracy\n",
      "0.3643876598864511\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.49563870402112825\n",
      "Average Testing Accuracy\n",
      "0.48739631238676917\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.4919848919990171\n",
      "Average Testing Accuracy\n",
      "0.5118597725386121\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.48387284253187623\n",
      "Average Testing Accuracy\n",
      "0.5104179921109555\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.8704389574759945\n",
      "Average Testing Accuracy\n",
      "0.7926697530864197\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.6382115912208504\n",
      "Average Testing Accuracy\n",
      "0.5773148148148148\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.31644375857338825\n",
      "Average Testing Accuracy\n",
      "0.33595679012345675\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.29188100137174205\n",
      "Average Testing Accuracy\n",
      "0.28479938271604943\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.24055212620027439\n",
      "Average Testing Accuracy\n",
      "0.22013888888888888\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.23736282578875167\n",
      "Average Testing Accuracy\n",
      "0.23888888888888887\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.27031893004115226\n",
      "Average Testing Accuracy\n",
      "0.2575617283950617\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.418564650059312\n",
      "Average Testing Accuracy\n",
      "0.40444839857651244\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.13792012653222618\n",
      "Average Testing Accuracy\n",
      "0.12526690391459078\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.12829181494661918\n",
      "Average Testing Accuracy\n",
      "0.12508896797153027\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.10294582839066826\n",
      "Average Testing Accuracy\n",
      "0.10302491103202846\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.11633056544088574\n",
      "Average Testing Accuracy\n",
      "0.12473309608540925\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.10719652036378016\n",
      "Average Testing Accuracy\n",
      "0.10960854092526691\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.09881376037959669\n",
      "Average Testing Accuracy\n",
      "0.10302491103202846\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.925819529775682\n",
      "Average Testing Accuracy\n",
      "0.8835830578232817\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.8076923786311779\n",
      "Average Testing Accuracy\n",
      "0.760947822462993\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.808439158584523\n",
      "Average Testing Accuracy\n",
      "0.7467100057511867\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.8223266226870212\n",
      "Average Testing Accuracy\n",
      "0.7822570603157569\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.8051505821441538\n",
      "Average Testing Accuracy\n",
      "0.7411806473637917\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.8144894637957835\n",
      "Average Testing Accuracy\n",
      "0.7200298869547174\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gilang/PyDP/src/pydp/ml/naive_bayes.py:103: PrivacyLeakWarning: Bounds have not been specified and will be calculated on the data provided. This will result in additional privacy leakage. To ensure differential privacy and no additional privacy leakage, specify bounds for each dimension.\n",
      "  PrivacyLeakWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Training Accuracy\n",
      "0.7047217261818646\n",
      "Average Testing Accuracy\n",
      "0.6622325237282457\n"
     ]
    }
   ],
   "source": [
    "from pydp.ml.naive_bayes import GaussianNB\n",
    "\n",
    "\n",
    "epsilons = [10.0, 1.0, 0.1, 0.05, 0.01, 0.005, 0.001]\n",
    "\n",
    "dp_results = {}\n",
    "\n",
    "for name, dataset in datasets.items():\n",
    "    \n",
    "    mean_train_accs = []\n",
    "    mean_test_accs = []\n",
    "    \n",
    "    for epsilon in epsilons:\n",
    "        clf = GaussianNB(epsilon=epsilon)\n",
    "\n",
    "        training_accuracy, test_accuracy = get_fit_result(clf, kfold, dataset[0], dataset[1], dataset[2])\n",
    "\n",
    "        mean_train_accs.append(training_accuracy * 100)\n",
    "        mean_test_accs.append(test_accuracy * 100 )\n",
    "        \n",
    "    dp_results[name] = mean_train_accs, mean_test_accs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "precise-childhood",
   "metadata": {},
   "source": [
    "``clf``: Gaussian Naive Bayes from PyDP is used as a classifier on five different datasets\n",
    "\n",
    "``mean_train_accs``: Differentially private average of training accuracy of a particular dataset\n",
    "\n",
    "``mean_test_accs``: Differentially private average of testing accuracy of a particular dataset\n",
    "\n",
    "``dp_results``: A dict containing names of five different datasets and their corresponding differentially private training and testing accuracies as a tuple"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "legitimate-routine",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_result(ax, dataset_name, baseline_train, baseline_test, dp_train, dp_test, epsilons):\n",
    "    ax.plot(epsilons, dp_train, marker=\"o\", label=\"Train (DP)\")\n",
    "    ax.plot(epsilons, dp_test, marker=\"x\", label=\"Test (DP)\")\n",
    "\n",
    "    # Set limit on y and axis\n",
    "    ax.set_xlim(0.1, -3.)\n",
    "    ax.set_ylim(0, 100)\n",
    "\n",
    "    # Set title\n",
    "    ax.set_title(f\"{dataset_name.capitalize()} Dataset:Average Accuracy over 10-fold Cross Validation\")\n",
    "\n",
    "    # Set label on y and axis\n",
    "    ax.set_ylabel(\"Train Accuracy (linear scale)\")\n",
    "    ax.set_xlabel(\"Epsilons (log scale 10^)\")\n",
    "\n",
    "    # Turn off grid background\n",
    "    ax.grid(False)\n",
    "\n",
    "    # Adjust x axis to display only specific numbers\n",
    "    ax.set_xticks(np.arange(ax.get_xticks().min(), ax.get_xticks().max(), 1.0))\n",
    "\n",
    "    # Draw the read line\n",
    "    ax.axhline(baseline_train * 100, color=\"r\", linewidth=2, label=\"Baseline Train\")\n",
    "    ax.axhline(baseline_test * 100, color=\"c\", linewidth=2, label=\"Baseline Test\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "clean-circular",
   "metadata": {},
   "source": [
    "## Plotting corresponding train and test accuracies over 10 folds cross validation using scikit-learn and PyDP's Naive Bayes algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "conscious-somerset",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(3,2,figsize=(15,15), constrained_layout=True)\n",
    "\n",
    "fig.delaxes(axs[2,1])\n",
    "\n",
    "epsilons = np.log10(epsilons)\n",
    "\n",
    "dataset_name = \"adult\"\n",
    "plot_result(\n",
    "    axs[0,0], \n",
    "    dataset_name, \n",
    "    baselines[dataset_name][0],\n",
    "    baselines[dataset_name][1], \n",
    "    dp_results[dataset_name][0], \n",
    "    dp_results[dataset_name][1], \n",
    "    epsilons\n",
    ")\n",
    "\n",
    "dataset_name = \"mushroom\"\n",
    "plot_result(\n",
    "    axs[0,1], \n",
    "    dataset_name, \n",
    "    baselines[dataset_name][0],\n",
    "    baselines[dataset_name][1], \n",
    "    dp_results[dataset_name][0], \n",
    "    dp_results[dataset_name][1], \n",
    "    epsilons\n",
    ")\n",
    "\n",
    "dataset_name = \"nursery\"\n",
    "plot_result(\n",
    "    axs[1,0], \n",
    "    dataset_name, \n",
    "    baselines[dataset_name][0],\n",
    "    baselines[dataset_name][1], \n",
    "    dp_results[dataset_name][0], \n",
    "    dp_results[dataset_name][1], \n",
    "    epsilons\n",
    ")\n",
    "\n",
    "dataset_name = \"digit\"\n",
    "plot_result(\n",
    "    axs[1,1], \n",
    "    dataset_name, \n",
    "    baselines[dataset_name][0],\n",
    "    baselines[dataset_name][1], \n",
    "    dp_results[dataset_name][0], \n",
    "    dp_results[dataset_name][1], \n",
    "    epsilons\n",
    ")\n",
    "\n",
    "dataset_name = \"skin\"\n",
    "plot_result(\n",
    "    axs[2,0], \n",
    "    dataset_name, \n",
    "    baselines[dataset_name][0],\n",
    "    baselines[dataset_name][1], \n",
    "    dp_results[dataset_name][0], \n",
    "    dp_results[dataset_name][1], \n",
    "    epsilons\n",
    ")\n",
    "\n",
    "\n",
    "handles, labels = axs[2,0].get_legend_handles_labels()\n",
    "fig.legend(handles, labels, loc='lower right')\n",
    "\n",
    "plt.savefig(\"./result.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "historical-migration",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "boxed-torture",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}