classification_tree_svm.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<center>\n",
    "    <img src=\"https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/assets/logos/SN_web_lightmode.png\" width=\"300\" alt=\"cognitiveclass.ai logo\">\n",
    "</center>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Credit Card Fraud Detection using Scikit-Learn and Snap ML**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This project uses two popular classification models to recognize fraudulent credit card transactions. These models are: Decision Tree and Support Vector Machine. We use a real dataset to train each of these models. The dataset includes information about transactions made by credit cards in September 2013 by European cardholders. You will use the trained model to assess if a credit card transaction is legitimate or not."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Objectives\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Perform basic data preprocessing in Python\n",
    "* Model a classification task using the Scikit-Learn and Snap ML Python APIs\n",
    "* Train Suppport Vector Machine and Decision Tree models using Scikit-Learn and Snap ML\n",
    "* Run inference and assess the quality of the trained models\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table of Contents\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\" style=\"margin-top: 10px\">\n",
    "    <ol>\n",
    "        <li><a href=\"#introduction\">Introduction</a></li>\n",
    "        <li><a href=\"#import_libraries\">Import Libraries</a></li>\n",
    "        <li><a href=\"#dataset_analysis\">Dataset Analysis</a></li>\n",
    "        <li><a href=\"#dataset_preprocessing\">Dataset Preprocessing</a></li>\n",
    "        <li><a href=\"#dataset_split\">Dataset Train/Test Split</a></li>\n",
    "        <li><a href=\"#dt_sklearn\">Build a Decision Tree Classifier model with Scikit-Learn</a></li>\n",
    "        <li><a href=\"#dt_snap\">Build a Decision Tree Classifier model with Snap ML</a></li>\n",
    "        <li><a href=\"#dt_sklearn_snap\">Evaluate the Scikit-Learn and Snap ML Decision Tree Classifiers</a></li>\n",
    "        <li><a href=\"#svm_sklearn\">Build a Support Vector Machine model with Scikit-Learn</a></li>\n",
    "        <li><a href=\"#svm_snap\">Build a Support Vector Machine model with Snap ML</a></li>\n",
    "        <li><a href=\"#svm_sklearn_snap\">Evaluate the Scikit-Learn and Snap ML Support Vector Machine Models</a></li>\n",
    "    </ol>\n",
    "</div>\n",
    "<br>\n",
    "<hr>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"Introduction\">\n",
    "    <h2>Introduction</h2>\n",
    "    <br>The aim of this project is to build a model that predicts if a credit card transaction is fraudulent or not. We model the problem as a binary classification problem. A transaction belongs to the positive class (1) if it is a fraud, otherwise it belongs to the negative class (0).\n",
    "    <br>\n",
    "    <br>We have access to transactions that occured over a certain period of time. The majority of the transactions are normally legitimate and only a small fraction are non-legitimate. Thus, typically we have access to a dataset that is highly unbalanced. This is also the case of the current dataset: only 492 transactions out of 284,807 are fraudulent (the positive class - the frauds - accounts for 0.172% of all transactions).\n",
    "    <br>\n",
    "    <br>This is a Kaggle dataset. You can find this \"Credit Card Fraud Detection\" dataset from the following link: <a href=\"https://www.kaggle.com/mlg-ulb/creditcardfraud\">Credit Card Fraud Detection</a>.\n",
    "<br>\n",
    "    <br>To train the model, we can use part of the input dataset, while the remaining data can be utilized to assess the quality of the trained model. \n",
    "    <br>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"import_libraries\">\n",
    "    <h2>Import Libraries</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting scikit-learn==1.0.2\n",
      "  Downloading scikit_learn-1.0.2-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (24.8 MB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m24.8/24.8 MB\u001b[0m \u001b[31m47.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
      "\u001b[?25hRequirement already satisfied: numpy>=1.14.6 in /home/jupyterlab/conda/envs/python/lib/python3.7/site-packages (from scikit-learn==1.0.2) (1.21.6)\n",
      "Requirement already satisfied: scipy>=1.1.0 in /home/jupyterlab/conda/envs/python/lib/python3.7/site-packages (from scikit-learn==1.0.2) (1.7.3)\n",
      "Collecting joblib>=0.11 (from scikit-learn==1.0.2)\n",
      "  Downloading joblib-1.3.2-py3-none-any.whl (302 kB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m302.2/302.2 kB\u001b[0m \u001b[31m43.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hCollecting threadpoolctl>=2.0.0 (from scikit-learn==1.0.2)\n",
      "  Downloading threadpoolctl-3.1.0-py3-none-any.whl (14 kB)\n",
      "Installing collected packages: threadpoolctl, joblib, scikit-learn\n",
      "  Attempting uninstall: scikit-learn\n",
      "    Found existing installation: scikit-learn 0.20.1\n",
      "    Uninstalling scikit-learn-0.20.1:\n",
      "      Successfully uninstalled scikit-learn-0.20.1\n",
      "Successfully installed joblib-1.3.2 scikit-learn-1.0.2 threadpoolctl-3.1.0\n",
      "Collecting snapml\n",
      "  Downloading snapml-1.14.0-cp37-cp37m-manylinux2014_x86_64.whl (7.4 MB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.4/7.4 MB\u001b[0m \u001b[31m68.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m:00:01\u001b[0m00:01\u001b[0m\n",
      "\u001b[?25hRequirement already satisfied: scikit-learn in /home/jupyterlab/conda/envs/python/lib/python3.7/site-packages (from snapml) (1.0.2)\n",
      "Requirement already satisfied: scipy in /home/jupyterlab/conda/envs/python/lib/python3.7/site-packages (from snapml) (1.7.3)\n",
      "Requirement already satisfied: numpy>=1.18.5 in /home/jupyterlab/conda/envs/python/lib/python3.7/site-packages (from snapml) (1.21.6)\n",
      "Requirement already satisfied: joblib>=0.11 in /home/jupyterlab/conda/envs/python/lib/python3.7/site-packages (from scikit-learn->snapml) (1.3.2)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in /home/jupyterlab/conda/envs/python/lib/python3.7/site-packages (from scikit-learn->snapml) (3.1.0)\n",
      "Installing collected packages: snapml\n",
      "Successfully installed snapml-1.14.0\n"
     ]
    }
   ],
   "source": [
    "# Install scikit-learn using pip\n",
    "!pip install scikit-learn==1.0.2\n",
    "\n",
    "# Snap ML is available on PyPI.\n",
    "!pip install snapml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Import the libraries we need to use in this lab\n",
    "from __future__ import print_function\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import normalize, StandardScaler\n",
    "from sklearn.utils.class_weight import compute_sample_weight\n",
    "from sklearn.metrics import roc_auc_score\n",
    "import time\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 284807 observations in the credit card fraud dataset.\n",
      "There are 31 variables in the dataset.\n"
     ]
    }
   ],
   "source": [
    "# download the dataset\n",
    "url= \"https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBMDeveloperSkillsNetwork-ML0101EN-SkillsNetwork/labs/Module%203/data/creditcard.csv\"\n",
    "\n",
    "# read the input data\n",
    "raw_data=pd.read_csv(url)\n",
    "print(\"There are \" + str(len(raw_data)) + \" observations in the credit card fraud dataset.\")\n",
    "print(\"There are \" + str(len(raw_data.columns)) + \" variables in the dataset.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"dataset_analysis\">\n",
    "    <h2>Dataset Analysis</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We read the dataset in a Pandas dataframe and visualize its content. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>...</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>378.66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>123.50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>69.99</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9  ...       V21       V22       V23       V24       V25  \\\n",
       "0  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "1  0.085102 -0.255425  ... -0.225775 -0.638672  0.101288 -0.339846  0.167170   \n",
       "2  0.247676 -1.514654  ...  0.247998  0.771679  0.909412 -0.689281 -0.327642   \n",
       "3  0.377436 -1.387024  ... -0.108300  0.005274 -0.190321 -1.175575  0.647376   \n",
       "4 -0.270533  0.817739  ... -0.009431  0.798278 -0.137458  0.141267 -0.206010   \n",
       "\n",
       "        V26       V27       V28  Amount  Class  \n",
       "0 -0.189115  0.133558 -0.021053  149.62      0  \n",
       "1  0.125895 -0.008983  0.014724    2.69      0  \n",
       "2 -0.139097 -0.055353 -0.059752  378.66      0  \n",
       "3 -0.221929  0.062723  0.061458  123.50      0  \n",
       "4  0.502292  0.219422  0.215153   69.99      0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# display the first rows in the dataset\n",
    "raw_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In practice, a financial institution may have access to a much larger dataset of transactions. To simulate such a case, we  inflate the original one 10 times.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 2848070 observations in the inflated credit card fraud dataset.\n",
      "There are 31 variables in the dataset.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "2   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "3   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "4   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "\n",
       "         V8        V9  ...       V21       V22       V23       V24       V25  \\\n",
       "0  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "1  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "2  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "3  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "4  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "\n",
       "        V26       V27       V28  Amount  Class  \n",
       "0 -0.189115  0.133558 -0.021053  149.62    0.0  \n",
       "1 -0.189115  0.133558 -0.021053  149.62    0.0  \n",
       "2 -0.189115  0.133558 -0.021053  149.62    0.0  \n",
       "3 -0.189115  0.133558 -0.021053  149.62    0.0  \n",
       "4 -0.189115  0.133558 -0.021053  149.62    0.0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_replicas = 10\n",
    "\n",
    "# inflate the original dataset\n",
    "big_raw_data = pd.DataFrame(np.repeat(raw_data.values, n_replicas, axis=0), columns=raw_data.columns)\n",
    "\n",
    "print(\"There are \" + str(len(big_raw_data)) + \" observations in the inflated credit card fraud dataset.\")\n",
    "print(\"There are \" + str(len(big_raw_data.columns)) + \" variables in the dataset.\")\n",
    "\n",
    "# display first rows in the new dataset\n",
    "big_raw_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Each row in the dataset represents a credit card transaction. As shown above, each row has 31 variables. One variable (the last variable in the table above) is called Class and represents the target variable. We train a model that uses the other variables to predict the value of the Class variable. \n",
    "\n",
    "Note: For confidentiality reasons, the original names of most features are anonymized V1, V2 .. V28. The values of these features are the result of a PCA transformation and are numerical. The feature 'Class' is the target variable and it takes two values: 1 in case of fraud and 0 otherwise. For more information about the dataset please visit this webpage: https://www.kaggle.com/mlg-ulb/creditcardfraud.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get the set of distinct classes\n",
    "labels = big_raw_data.Class.unique()\n",
    "\n",
    "# get the count of each class\n",
    "sizes = big_raw_data.Class.value_counts().values\n",
    "\n",
    "# plot the class value counts\n",
    "fig, ax = plt.subplots()\n",
    "ax.pie(sizes, labels=labels, autopct='%1.3f%%')\n",
    "ax.set_title('Target Variable Value Counts')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As shown above, the Class variable has two values: 0 (the credit card transaction is legitimate) and 1 (the credit card transaction is fraudulent). Moreover, the dataset is highly unbalanced, the target variable classes are not represented equally. This case requires special attention when training or when evaluating the quality of a model. One way of handing this case at train time is to bias the model to pay more attention to the samples in the minority class. The models under the current study will be configured to take into account the class weights of the samples at train/fit time.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"dataset_preprocessing\">\n",
    "    <h2>Dataset Preprocessing</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this subsection we prepare the data for training. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X.shape= (2848070, 29) y.shape= (2848070,)\n"
     ]
    }
   ],
   "source": [
    "# standardize features by removing the mean and scaling to unit variance\n",
    "big_raw_data.iloc[:, 1:30] = StandardScaler().fit_transform(big_raw_data.iloc[:, 1:30])\n",
    "data_matrix = big_raw_data.values\n",
    "\n",
    "# X: feature matrix (for this analysis, we exclude the Time variable from the dataset)\n",
    "X = data_matrix[:, 1:30]\n",
    "\n",
    "# y: labels vector\n",
    "y = data_matrix[:, 30]\n",
    "\n",
    "# data normalization\n",
    "X = normalize(X, norm=\"l1\")\n",
    "\n",
    "# print the shape of the features matrix and the labels vector\n",
    "print('X.shape=', X.shape, 'y.shape=', y.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"dataset_split\">\n",
    "    <h2>Dataset Train/Test Split</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We divide the pre-processed dataset into a subset to be used for training the model (the train set) and a subset to be used for evaluating the quality of the model (the test set).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train.shape= (1993649, 29) Y_train.shape= (1993649,)\n",
      "X_test.shape= (854421, 29) Y_test.shape= (854421,)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42, stratify=y)       \n",
    "print('X_train.shape=', X_train.shape, 'Y_train.shape=', y_train.shape)\n",
    "print('X_test.shape=', X_test.shape, 'Y_test.shape=', y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"dt_sklearn\">\n",
    "    <h2>Build a Decision Tree Classifier model with Scikit-Learn</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Scikit-Learn] Training time (s):  47.35515\n"
     ]
    }
   ],
   "source": [
    "# compute the sample weights to be used as input to the train routine so that \n",
    "# it takes into account the class imbalance present in this dataset\n",
    "w_train = compute_sample_weight('balanced', y_train)\n",
    "\n",
    "# import the Decision Tree Classifier Model from scikit-learn\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "# for reproducible output across multiple function calls, set random_state to a given integer value\n",
    "sklearn_dt = DecisionTreeClassifier(max_depth=4, random_state=35)\n",
    "\n",
    "# train a Decision Tree Classifier using scikit-learn\n",
    "t0 = time.time()\n",
    "sklearn_dt.fit(X_train, y_train, sample_weight=w_train)\n",
    "sklearn_time = time.time()-t0\n",
    "print(\"[Scikit-Learn] Training time (s):  {0:.5f}\".format(sklearn_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"dt_snapml\">\n",
    "    <h2>Build a Decision Tree Classifier model with Snap ML</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Snap ML] Training time (s):  9.77214\n"
     ]
    }
   ],
   "source": [
    "# if not already computed, \n",
    "# compute the sample weights to be used as input to the train routine so that \n",
    "# it takes into account the class imbalance present in this dataset\n",
    "# w_train = compute_sample_weight('balanced', y_train)\n",
    "\n",
    "# import the Decision Tree Classifier Model from Snap ML\n",
    "from snapml import DecisionTreeClassifier\n",
    "\n",
    "# Snap ML offers multi-threaded CPU/GPU training of decision trees, unlike scikit-learn\n",
    "# to use the GPU, set the use_gpu parameter to True\n",
    "# snapml_dt = DecisionTreeClassifier(max_depth=4, random_state=45, use_gpu=True)\n",
    "\n",
    "# to set the number of CPU threads used at training time, set the n_jobs parameter\n",
    "# for reproducible output across multiple function calls, set random_state to a given integer value\n",
    "snapml_dt = DecisionTreeClassifier(max_depth=4, random_state=45, n_jobs=4)\n",
    "\n",
    "# train a Decision Tree Classifier model using Snap ML\n",
    "t0 = time.time()\n",
    "snapml_dt.fit(X_train, y_train, sample_weight=w_train)\n",
    "snapml_time = time.time()-t0\n",
    "print(\"[Snap ML] Training time (s):  {0:.5f}\".format(snapml_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"dt_sklearn_snapml\">\n",
    "    <h2>Evaluate the Scikit-Learn and Snap ML Decision Tree Classifier Models</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Decision Tree Classifier] Snap ML vs. Scikit-Learn speedup : 4.85x \n",
      "[Scikit-Learn] ROC-AUC score : 0.966\n",
      "[Snap ML] ROC-AUC score : 0.966\n"
     ]
    }
   ],
   "source": [
    "# Snap ML vs Scikit-Learn training speedup\n",
    "training_speedup = sklearn_time/snapml_time\n",
    "print('[Decision Tree Classifier] Snap ML vs. Scikit-Learn speedup : {0:.2f}x '.format(training_speedup))\n",
    "\n",
    "# run inference and compute the probabilities of the test samples \n",
    "# to belong to the class of fraudulent transactions\n",
    "sklearn_pred = sklearn_dt.predict_proba(X_test)[:,1]\n",
    "\n",
    "# evaluate the Compute Area Under the Receiver Operating Characteristic \n",
    "# Curve (ROC-AUC) score from the predictions\n",
    "sklearn_roc_auc = roc_auc_score(y_test, sklearn_pred)\n",
    "print('[Scikit-Learn] ROC-AUC score : {0:.3f}'.format(sklearn_roc_auc))\n",
    "\n",
    "# run inference and compute the probabilities of the test samples\n",
    "# to belong to the class of fraudulent transactions\n",
    "snapml_pred = snapml_dt.predict_proba(X_test)[:,1]\n",
    "\n",
    "# evaluate the Compute Area Under the Receiver Operating Characteristic\n",
    "# Curve (ROC-AUC) score from the prediction scores\n",
    "snapml_roc_auc = roc_auc_score(y_test, snapml_pred)   \n",
    "print('[Snap ML] ROC-AUC score : {0:.3f}'.format(snapml_roc_auc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As shown above both decision tree models provide the same score on the test dataset. However Snap ML runs the training routine faster than Scikit-Learn. This is one of the advantages of using Snap ML: acceleration of training of classical machine learning models, such as linear and tree-based models. For more Snap ML examples, please visit [snapml-examples](https://ibm.biz/BdPfxP).\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"svm_sklearn\">\n",
    "    <h2>Build a Support Vector Machine model with Scikit-Learn</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Scikit-Learn] Training time (s):  93.52\n"
     ]
    }
   ],
   "source": [
    "# import the linear Support Vector Machine (SVM) model from Scikit-Learn\n",
    "from sklearn.svm import LinearSVC\n",
    "\n",
    "# instatiate a scikit-learn SVM model\n",
    "# to indicate the class imbalance at fit time, set class_weight='balanced'\n",
    "# for reproducible output across multiple function calls, set random_state to a given integer value\n",
    "sklearn_svm = LinearSVC(class_weight='balanced', random_state=31, loss=\"hinge\", fit_intercept=False)\n",
    "\n",
    "# train a linear Support Vector Machine model using Scikit-Learn\n",
    "t0 = time.time()\n",
    "sklearn_svm.fit(X_train, y_train)\n",
    "sklearn_time = time.time() - t0\n",
    "print(\"[Scikit-Learn] Training time (s):  {0:.2f}\".format(sklearn_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"svm_snap\">\n",
    "    <h2>Build a Support Vector Machine model with Snap ML</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Snap ML] Training time (s):  43.72\n"
     ]
    }
   ],
   "source": [
    "# import the Support Vector Machine model (SVM) from Snap ML\n",
    "from snapml import SupportVectorMachine\n",
    "\n",
    "# in contrast to scikit-learn's LinearSVC, Snap ML offers multi-threaded CPU/GPU training of SVMs\n",
    "# to use the GPU, set the use_gpu parameter to True\n",
    "# snapml_svm = SupportVectorMachine(class_weight='balanced', random_state=25, use_gpu=True, fit_intercept=False)\n",
    "\n",
    "# to set the number of threads used at training time, one needs to set the n_jobs parameter\n",
    "snapml_svm = SupportVectorMachine(class_weight='balanced', random_state=25, n_jobs=4, fit_intercept=False)\n",
    "# print(snapml_svm.get_params())\n",
    "\n",
    "# train an SVM model using Snap ML\n",
    "t0 = time.time()\n",
    "model = snapml_svm.fit(X_train, y_train)\n",
    "snapml_time = time.time() - t0\n",
    "print(\"[Snap ML] Training time (s):  {0:.2f}\".format(snapml_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div id=\"svm_sklearn_snap\">\n",
    "    <h2>Evaluate the Scikit-Learn and Snap ML Support Vector Machine Models</h2>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Support Vector Machine] Snap ML vs. Scikit-Learn training speedup : 2.14x \n",
      "[Scikit-Learn] ROC-AUC score:   0.984\n",
      "[Snap ML] ROC-AUC score:   0.985\n"
     ]
    }
   ],
   "source": [
    "# compute the Snap ML vs Scikit-Learn training speedup\n",
    "training_speedup = sklearn_time/snapml_time\n",
    "print('[Support Vector Machine] Snap ML vs. Scikit-Learn training speedup : {0:.2f}x '.format(training_speedup))\n",
    "\n",
    "# run inference using the Scikit-Learn model\n",
    "# get the confidence scores for the test samples\n",
    "sklearn_pred = sklearn_svm.decision_function(X_test)\n",
    "\n",
    "# evaluate accuracy on test set\n",
    "acc_sklearn  = roc_auc_score(y_test, sklearn_pred)\n",
    "print(\"[Scikit-Learn] ROC-AUC score:   {0:.3f}\".format(acc_sklearn))\n",
    "\n",
    "# run inference using the Snap ML model\n",
    "# get the confidence scores for the test samples\n",
    "snapml_pred = snapml_svm.decision_function(X_test)\n",
    "\n",
    "# evaluate accuracy on test set\n",
    "acc_snapml  = roc_auc_score(y_test, snapml_pred)\n",
    "print(\"[Snap ML] ROC-AUC score:   {0:.3f}\".format(acc_snapml))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As shown above both SVM models provide the same score on the test dataset. However, as in the case of decision trees, Snap ML runs the training routine faster than Scikit-Learn. For more Snap ML examples, please visit [snapml-examples](https://ibm.biz/BdPfxP). Moreover, as shown above, not only is Snap ML seemlessly accelerating scikit-learn applications, but the library's Python API is also compatible with scikit-learn metrics and data preprocessors.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Authors\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Andreea Anghel \n",
    "\n",
    "<a href=\"http://www.linkedin.com/in/idaraabasiudoh\" target=\"_blank\">Idara-Abasi Udoh</a>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Other Contributors\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Joseph Santarcangelo\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <h3 align=\"center\">  Copyright &copy; 2021 IBM Corporation.  <h3/>\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python",
   "language": "python",
   "name": "conda-env-python-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.12"
  },
  "prev_pub_hash": "3f8f93d5880dec3928c1a1a0753bb6ff77b98a9e4f7dde3af10b097f0c020d5c"
 },
 "nbformat": 4,
 "nbformat_minor": 4
}