lectures/02_decision-trees.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "![](img/330-banner.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "# Lecture 2: Terminology, Baselines, Decision Trees\n",
    "\n",
    "UBC 2021-22\n",
    "\n",
    "Instructor: Varada Kolhatkar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "import glob\n",
    "import os\n",
    "import re\n",
    "import sys\n",
    "from collections import Counter, defaultdict\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "sys.path.append(\"code/.\")\n",
    "import graphviz\n",
    "import IPython\n",
    "import mglearn\n",
    "from IPython.display import HTML, display\n",
    "from plotting_functions import *\n",
    "from sklearn.dummy import DummyClassifier\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.linear_model import LinearRegression, LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.pipeline import Pipeline, make_pipeline\n",
    "from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor, export_graphviz\n",
    "from utils import *\n",
    "\n",
    "plt.rcParams[\"font.size\"] = 16\n",
    "pd.set_option(\"display.max_colwidth\", 200)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Announcements \n",
    "\n",
    "- Things due this week \n",
    "    - Syllabus quiz: Sept 14, 11:59pm\n",
    "    - Homework 1 (hw1): Sept 14, 11:59pm \n",
    "- Homework 2 (hw2) has been released (Due: Sept 20, 11:59pm)\n",
    "- You can find the tentative due dates for all deliverables [here](https://ubc-cs.github.io/cpsc330/README.html#deliverable-due-dates-tentative). \n",
    "- Please monitor Piazza (especially pinned posts and instructor posts) for announcements. \n",
    "- Sorry for the setup difficulties. That said, you are likely to go through this in future and figuring out these things on your own is a valuable skill."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Getting to know you survey"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Your concerns\n",
    "\n",
    "- familiarity with Python\n",
    "    - Sorry about this but we won't be spending any time on teaching Python. We won't be doing fancy stuff with Python in this course but if you struggled with hw1, it's a sign that you are likely to struggle later in the course. \n",
    "- Workload and falling behind\n",
    "    - We'll try to do some practice exercises in class so that you can keep up with the material. \n",
    "    - If you feel like we're going too fast or too slow (unlikely), please feel free to reach out. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Summary of fun facts 😀 \n",
    "\n",
    "We have very interesting students in this class!! We have \n",
    "- poker, tennis, judo, basket ball, spikeball players\n",
    "- rock climbers, sky divers, swimmers, latte artists\n",
    "- saxophone, guitar players, tuba players  \n",
    "- people who have podcasts, who have done musicals, who used to be a practicing veterinarian, who have aircraft pilot license, who are multilingual\n",
    "- people with diverse backgrounds: English literature, physiology, statistics, CS, biomedical engineering, mechanical engineering, ...\n",
    "- coffee, cat, dog, tortoise, shopping lovers\n",
    "- some anti-fun people \n",
    "- ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Some of my favourite ones: \n",
    "> I've traveled to 50+ countries.\n",
    "‼️‼️\n",
    "\n",
    "> I can sing the periodic element table song"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Quick recap: True or False?  \n",
    "\n",
    "- There are different types of machine learning. \n",
    "- Predicting spam and predicting housing prices are both examples of supervised machine learning. \n",
    "- For problems such as spelling correction, translation, face recognition, spam identification, if you are a domain expert, it's usually faster and scalable to come up with a robust set of rules manually rather than building a machine learning model. \n",
    "- Google News is likely be using machine learning to organize news. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Learning outcomes \n",
    "From this lecture, you will be able to \n",
    "\n",
    "- identify whether a given problem could be solved using supervised machine learning or not; \n",
    "- differentiate between supervised and unsupervised machine learning;\n",
    "- explain machine learning terminology such as features, targets, predictions, training, and error;\n",
    "- differentiate between classification and regression problems;\n",
    "- use `DummyClassifier` and `DummyRegressor` as baselines for machine learning problems;\n",
    "- explain the `fit` and `predict` paradigm and use `score` method of ML models; \n",
    "- broadly describe how decision tree prediction works;\n",
    "- use `DecisionTreeClassifier` and `DecisionTreeRegressor` to build decision trees using `scikit-learn`; \n",
    "- visualize decision trees; \n",
    "- explain the difference between parameters and hyperparameters; \n",
    "- explain the concept of decision boundaries;\n",
    "- explain the relation between model complexity and decision boundaries."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Big picture and datasets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this lecture, we'll talk about our first machine learning model: Decision trees. We will also familiarize ourselves with some common terminology in supervised machine learning."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Toy datasets \n",
    "Later in the course we will use larger datasets from Kaggle, for instance. But for our first couple of lectures, we will be working with the following three toy datasets:  \n",
    "\n",
    "- [Quiz2 grade prediction classification dataset](data/quiz2-grade-toy-classification.csv)\n",
    "- [Quiz2 grade prediction regression dataset](data/quiz2-grade-toy-regression.csv)\n",
    "- [Canada USA cities dataset](canada_usa_cities.csv)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "```{note} \n",
    "If it's not necessary for you to understand the code, I will put it in one of the files under the `code` directory to avoid clutter in this notebook. For example, most of the plotting code is going to be in `code/plotting_functions.py`. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "<br><br><br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Terminology [[video](https://youtu.be/YNT8n4cXu4A)]\n",
    "\n",
    "You will see a lot of variable terminology in machine learning and statistics. Let's familiarize ourselves with some of the basic terminology used in ML. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{seealso}\n",
    "Check out [the accompanying video](https://youtu.be/YNT8n4cXu4A) on this material. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "I'll be using the following grade prediction toy dataset to demonstrate the terminology. Imagine that you are taking a course with four home work assignments and two quizzes. You and your friends are quite nervous about your quiz2 grades and you want to know how will you do based on your previous performance and some other attributes. So you decide to collect some data from your friends from last year and train a supervised machine learning model for quiz2 grade prediction. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(21, 8)\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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "      <th>quiz2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>83</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1   quiz2\n",
       "0              1                 1    92    93    84    91     92      A+\n",
       "1              1                 0    94    90    80    83     91  not A+\n",
       "2              0                 0    78    85    83    80     80  not A+\n",
       "3              0                 1    91    94    92    91     89      A+\n",
       "4              0                 1    77    83    90    92     85      A+"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classification_df = pd.read_csv(\"data/quiz2-grade-toy-classification.csv\")\n",
    "print(classification_df.shape)\n",
    "classification_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Recap: Supervised machine learning\n",
    "\n",
    "![](img/sup-learning.png)\n",
    "<!-- <img src=\"img/sup-learning.png\" height=\"800\" width=\"800\">  -->"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Tabular data\n",
    "In supervised machine learning, the input data is typically organized in a **tabular** format, where rows are **examples** and columns are **features**. One of the columns is typically the **target**. \n",
    "\n",
    "![](img/sup-ml-terminology.png)\n",
    "\n",
    "<!-- <img src=\"img/sup-ml-terminology.png\" height=\"1000\" width=\"1000\">  -->"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "**Features** \n",
    ": Features are relevant characteristics of the problem, usually suggested by experts. Features are typically denoted by $X$ and the number of features is usually denoted by $d$.  \n",
    "\n",
    "**Target**\n",
    ": Target is the feature we want to predict (typically denoted by $y$). \n",
    "\n",
    "**Example** \n",
    ": A row of feature values. When people refer to an example, it may or may not include the target corresponding to the feature values, depending upon the context. The number of examples is usually denoted by $n$. \n",
    "\n",
    "**Training**\n",
    ": The process of learning the mapping between the features ($X$) and the target ($y$). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Example: Tabular data for grade prediction\n",
    "\n",
    "The tabular data usually contains both: the features (`X`) and the target (`y`). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "      <th>quiz2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>83</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1   quiz2\n",
       "0              1                 1    92    93    84    91     92      A+\n",
       "1              1                 0    94    90    80    83     91  not A+\n",
       "2              0                 0    78    85    83    80     80  not A+\n",
       "3              0                 1    91    94    92    91     89      A+\n",
       "4              0                 1    77    83    90    92     85      A+"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classification_df = pd.read_csv(\"data/quiz2-grade-toy-classification.csv\")\n",
    "classification_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "So the first step in training a supervised machine learning model is separating `X` and `y`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>83</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1\n",
       "0              1                 1    92    93    84    91     92\n",
       "1              1                 0    94    90    80    83     91\n",
       "2              0                 0    78    85    83    80     80\n",
       "3              0                 1    91    94    92    91     89\n",
       "4              0                 1    77    83    90    92     85"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = classification_df.drop(columns=[\"quiz2\"])\n",
    "y = classification_df[\"quiz2\"]\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        A+\n",
       "1    not A+\n",
       "2    not A+\n",
       "3        A+\n",
       "4        A+\n",
       "Name: quiz2, dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Example: Tabular data for the housing price prediction\n",
    "\n",
    "Here is an example of tabular data for housing price prediction. You can download the data from [here](https://www.kaggle.com/harlfoxem/housesalesprediction). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>price</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>sqft_living</th>\n",
       "      <th>sqft_lot</th>\n",
       "      <th>floors</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>view</th>\n",
       "      <th>condition</th>\n",
       "      <th>grade</th>\n",
       "      <th>sqft_above</th>\n",
       "      <th>sqft_basement</th>\n",
       "      <th>yr_built</th>\n",
       "      <th>yr_renovated</th>\n",
       "      <th>zipcode</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>sqft_living15</th>\n",
       "      <th>sqft_lot15</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>221900.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1180</td>\n",
       "      <td>5650</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>1180.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1955</td>\n",
       "      <td>0</td>\n",
       "      <td>98178</td>\n",
       "      <td>47.5112</td>\n",
       "      <td>-122.257</td>\n",
       "      <td>1340</td>\n",
       "      <td>5650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>538000.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2.25</td>\n",
       "      <td>2570</td>\n",
       "      <td>7242</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>2170.0</td>\n",
       "      <td>400</td>\n",
       "      <td>1951</td>\n",
       "      <td>1991</td>\n",
       "      <td>98125</td>\n",
       "      <td>47.7210</td>\n",
       "      <td>-122.319</td>\n",
       "      <td>1690</td>\n",
       "      <td>7639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>180000.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>770</td>\n",
       "      <td>10000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>770.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1933</td>\n",
       "      <td>0</td>\n",
       "      <td>98028</td>\n",
       "      <td>47.7379</td>\n",
       "      <td>-122.233</td>\n",
       "      <td>2720</td>\n",
       "      <td>8062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>604000.0</td>\n",
       "      <td>4</td>\n",
       "      <td>3.00</td>\n",
       "      <td>1960</td>\n",
       "      <td>5000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>1050.0</td>\n",
       "      <td>910</td>\n",
       "      <td>1965</td>\n",
       "      <td>0</td>\n",
       "      <td>98136</td>\n",
       "      <td>47.5208</td>\n",
       "      <td>-122.393</td>\n",
       "      <td>1360</td>\n",
       "      <td>5000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>510000.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2.00</td>\n",
       "      <td>1680</td>\n",
       "      <td>8080</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>8</td>\n",
       "      <td>1680.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1987</td>\n",
       "      <td>0</td>\n",
       "      <td>98074</td>\n",
       "      <td>47.6168</td>\n",
       "      <td>-122.045</td>\n",
       "      <td>1800</td>\n",
       "      <td>7503</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_df = pd.read_csv(\"data/kc_house_data.csv\")\n",
    "housing_df.drop([\"id\", \"date\"], axis=1, inplace=True)\n",
    "HTML(housing_df.head().to_html(index=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "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>bedrooms</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>sqft_living</th>\n",
       "      <th>sqft_lot</th>\n",
       "      <th>floors</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>view</th>\n",
       "      <th>condition</th>\n",
       "      <th>grade</th>\n",
       "      <th>sqft_above</th>\n",
       "      <th>sqft_basement</th>\n",
       "      <th>yr_built</th>\n",
       "      <th>yr_renovated</th>\n",
       "      <th>zipcode</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>sqft_living15</th>\n",
       "      <th>sqft_lot15</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1180</td>\n",
       "      <td>5650</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>1180.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1955</td>\n",
       "      <td>0</td>\n",
       "      <td>98178</td>\n",
       "      <td>47.5112</td>\n",
       "      <td>-122.257</td>\n",
       "      <td>1340</td>\n",
       "      <td>5650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>2.25</td>\n",
       "      <td>2570</td>\n",
       "      <td>7242</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>2170.0</td>\n",
       "      <td>400</td>\n",
       "      <td>1951</td>\n",
       "      <td>1991</td>\n",
       "      <td>98125</td>\n",
       "      <td>47.7210</td>\n",
       "      <td>-122.319</td>\n",
       "      <td>1690</td>\n",
       "      <td>7639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>770</td>\n",
       "      <td>10000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>770.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1933</td>\n",
       "      <td>0</td>\n",
       "      <td>98028</td>\n",
       "      <td>47.7379</td>\n",
       "      <td>-122.233</td>\n",
       "      <td>2720</td>\n",
       "      <td>8062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>3.00</td>\n",
       "      <td>1960</td>\n",
       "      <td>5000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>1050.0</td>\n",
       "      <td>910</td>\n",
       "      <td>1965</td>\n",
       "      <td>0</td>\n",
       "      <td>98136</td>\n",
       "      <td>47.5208</td>\n",
       "      <td>-122.393</td>\n",
       "      <td>1360</td>\n",
       "      <td>5000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>2.00</td>\n",
       "      <td>1680</td>\n",
       "      <td>8080</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>8</td>\n",
       "      <td>1680.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1987</td>\n",
       "      <td>0</td>\n",
       "      <td>98074</td>\n",
       "      <td>47.6168</td>\n",
       "      <td>-122.045</td>\n",
       "      <td>1800</td>\n",
       "      <td>7503</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   bedrooms  bathrooms  sqft_living  sqft_lot  floors  waterfront  view  \\\n",
       "0         3       1.00         1180      5650     1.0           0     0   \n",
       "1         3       2.25         2570      7242     2.0           0     0   \n",
       "2         2       1.00          770     10000     1.0           0     0   \n",
       "3         4       3.00         1960      5000     1.0           0     0   \n",
       "4         3       2.00         1680      8080     1.0           0     0   \n",
       "\n",
       "   condition  grade  sqft_above  sqft_basement  yr_built  yr_renovated  \\\n",
       "0          3      7      1180.0              0      1955             0   \n",
       "1          3      7      2170.0            400      1951          1991   \n",
       "2          3      6       770.0              0      1933             0   \n",
       "3          5      7      1050.0            910      1965             0   \n",
       "4          3      8      1680.0              0      1987             0   \n",
       "\n",
       "   zipcode      lat     long  sqft_living15  sqft_lot15  \n",
       "0    98178  47.5112 -122.257           1340        5650  \n",
       "1    98125  47.7210 -122.319           1690        7639  \n",
       "2    98028  47.7379 -122.233           2720        8062  \n",
       "3    98136  47.5208 -122.393           1360        5000  \n",
       "4    98074  47.6168 -122.045           1800        7503  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = housing_df.drop(columns=[\"price\"])\n",
    "y = housing_df[\"price\"]\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    221900.0\n",
       "1    538000.0\n",
       "2    180000.0\n",
       "3    604000.0\n",
       "4    510000.0\n",
       "Name: price, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "```{admonition} Attention\n",
    ":class: important\n",
    "To a machine, column names (features) have no meaning. Only feature values and how they vary across examples mean something. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Alternative terminology for examples, features, targets, and training\n",
    "\n",
    "- **examples** = rows = samples = records = instances \n",
    "- **features** = inputs = predictors = explanatory variables = regressors = independent variables = covariates\n",
    "- **targets** = outputs = outcomes = response variable = dependent variable = labels (if categorical).\n",
    "- **training** = learning = fitting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{seealso} \n",
    "Check out [the MDS terminology document](https://ubc-mds.github.io/resources_pages/terminology/). \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Supervised learning vs. Unsupervised learning\n",
    "\n",
    "In **supervised learning**, training data comprises a set of features ($X$) and their corresponding targets ($y$). We wish to find a **model function $f$** that relates $X$ to $y$. Then use that model function **to predict the targets** of new examples. \n",
    "\n",
    "\n",
    "![](img/sup-learning.png)\n",
    "\n",
    "<!-- <img src=\"img/sup-learning.png\" height=\"900\" width=\"900\"> -->\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "In **unsupervised learning** training data consists of observations ($X$) **without any corresponding targets**. Unsupervised learning could be used to **group similar things together** in $X$ or to provide **concise summary** of the data. We'll learn more about this topic in later videos.\n",
    "\n",
    "![](img/unsup-learning.png)\n",
    "\n",
    "<!-- <img src=\"img/unsup-learning.png\" alt=\"\" height=\"900\" width=\"900\"> -->"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Supervised machine learning is about function approximation, i.e., finding the mapping function between `X` and `y` whereas unsupervised machine learning is about concisely describing the data.   \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Classification vs. Regression \n",
    "In supervised machine learning, there are two main kinds of learning problems based on what they are trying to predict.\n",
    "- **Classification problem**: predicting among two or more discrete classes\n",
    "    - Example1: Predict whether a patient has a liver disease or not\n",
    "    - Example2: Predict whether a student would get an A+ or not in quiz2.  \n",
    "- **Regression problem**: predicting a continuous value\n",
    "    - Example1: Predict housing prices \n",
    "    - Example2: Predict a student's score in quiz2."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "![](img/classification-vs-regression.png)\n",
    "<!-- <img src=\"img/classification-vs-regression.png\" height=\"1500\" width=\"1500\">  -->"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "      <th>quiz2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1   quiz2\n",
       "0              1                 1    92    93    84    91     92      A+\n",
       "1              1                 0    94    90    80    83     91  not A+\n",
       "2              0                 0    78    85    83    80     80  not A+\n",
       "3              0                 1    91    94    92    91     89      A+"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# quiz2 classification toy data\n",
    "classification_df = pd.read_csv(\"data/quiz2-grade-toy-classification.csv\")\n",
    "classification_df.head(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "      <th>quiz2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "      <td>92</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1  quiz2\n",
       "0              1                 1    92    93    84    91     92     90\n",
       "1              1                 0    94    90    80    83     91     84\n",
       "2              0                 0    78    85    83    80     80     82\n",
       "3              0                 1    91    94    92    91     89     92"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# quiz2 regression toy data\n",
    "regression_df = pd.read_csv(\"data/quiz2-grade-toy-regression.csv\")\n",
    "regression_df.head(4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### ❓❓ Questions for you"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    },
    "tags": []
   },
   "source": [
    "```{admonition} Exercise 2.1: $X$ and $y$  \n",
    "1. How many examples and features are there in the housing price data above? You can use `df.shape` to get number of rows and columns in a dataframe. \n",
    "2. For each of the following examples what would be the relevant features and what would be the target?\n",
    "    1. Sentiment analysis\n",
    "    2. Fraud detection \n",
    "    3. Face recognition \n",
    "```    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    },
    "tags": [
     "remove_cell"
    ]
   },
   "source": [
    "```{admonition} Exercise 2.1.1: Solutions!\n",
    ":class: tip, dropdown\n",
    "TBD\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "```{admonition} Exercise 2.2: Supervised vs. unsupervised \n",
    "\n",
    "Which of these are examples of supervised learning?\n",
    "\n",
    "1. Finding groups of similar properties in a real estate data set.\n",
    "2. Predicting real estate prices based on house features like number of rooms, learning from past sales as examples.\n",
    "3. Grouping articles on different topics from different news sources (something like the Google News app). \n",
    "4. Detecting credit card fraud based on examples of fraudulent and non-fraudulent transactions.\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    },
    "tags": [
     "remove_cell"
    ]
   },
   "source": [
    "```{admonition} Exercise 2.1.2: Solutions!\n",
    ":class: tip, dropdown\n",
    "TBD\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "```{admonition} Exercise 2.3: Classification vs. Regression\n",
    "\n",
    "Which of these are examples of classification and which ones are of regression?\n",
    "\n",
    "1. Predicting the price of a house based on features such as number of bedrooms and the year built.\n",
    "2. Predicting if a house will sell or not based on features like the price of the house, number of rooms, etc.\n",
    "3. Predicting percentage grade in CPSC 330 based on past grades.\n",
    "4. Predicting whether you should bicycle tomorrow or not based on the weather forecast.\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "```{admonition} Exercise 2.1.3: Solutions!\n",
    ":class: tip, dropdown\n",
    "TBD\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Baselines [[video](https://youtu.be/6eT5cLL-2Vc)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{seealso}\n",
    "Check out [the accompanying video](https://youtu.be/6eT5cLL-2Vc) on this material. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    },
    "tags": []
   },
   "source": [
    "### Supervised learning (Reminder)\n",
    "\n",
    "- Training data $\\rightarrow$ Machine learning algorithm $\\rightarrow$ ML model \n",
    "- Unseen test data + ML model $\\rightarrow$ predictions\n",
    "![](img/sup-learning.png)\n",
    "<!-- <img src=\"img/sup-learning.png\" height=\"1000\" width=\"1000\">  -->"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "\n",
    "Let's build a very simple supervised machine learning model for quiz2 grade prediction problem. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "      <th>quiz2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>83</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1   quiz2\n",
       "0              1                 1    92    93    84    91     92      A+\n",
       "1              1                 0    94    90    80    83     91  not A+\n",
       "2              0                 0    78    85    83    80     80  not A+\n",
       "3              0                 1    91    94    92    91     89      A+\n",
       "4              0                 1    77    83    90    92     85      A+"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classification_df = pd.read_csv(\"data/quiz2-grade-toy-classification.csv\")\n",
    "classification_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "not A+    11\n",
       "A+        10\n",
       "Name: quiz2, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classification_df['quiz2'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seems like \"not A+\" occurs more frequently than \"A+\". What if we predict \"not A+\" all the time? "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Baselines \n",
    "\n",
    "**Baseline**\n",
    ": A simple machine learning algorithm based on simple rules of thumb. \n",
    "\n",
    "- For example, most frequent baseline always predicts the most frequent label in the training set. \n",
    "- Baselines provide a way to sanity check your machine learning model.    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### `DummyClassifier` \n",
    "\n",
    "- `sklearn`'s baseline model for classification  \n",
    "- Let's train `DummyClassifier` on the grade prediction dataset. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Steps to train a classifier using `sklearn` \n",
    "\n",
    "1. Read the data\n",
    "2. Create $X$ and $y$\n",
    "3. Create a classifier object\n",
    "4. `fit` the classifier\n",
    "5. `predict` on new examples\n",
    "6. `score` the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Reading the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "      <th>quiz2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "      <td>not A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>83</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "      <td>A+</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1   quiz2\n",
       "0              1                 1    92    93    84    91     92      A+\n",
       "1              1                 0    94    90    80    83     91  not A+\n",
       "2              0                 0    78    85    83    80     80  not A+\n",
       "3              0                 1    91    94    92    91     89      A+\n",
       "4              0                 1    77    83    90    92     85      A+"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classification_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Create $X$ and $y$\n",
    "\n",
    "- $X$ &rarr; Feature vectors\n",
    "- $y$ &rarr; Target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [],
   "source": [
    "X = classification_df.drop(columns=[\"quiz2\"])\n",
    "y = classification_df[\"quiz2\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Create a classifier object\n",
    "\n",
    "- `import` the appropriate classifier \n",
    "- Create an object of the classifier "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.dummy import DummyClassifier # import the classifier\n",
    "\n",
    "dummy_clf = DummyClassifier(strategy=\"most_frequent\") # Create a classifier object"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### `fit` the classifier\n",
    "\n",
    "- The \"learning\" is carried out when we call `fit` on the classifier object. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [],
   "source": [
    "dummy_clf.fit(X, y); # fit the classifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### `predict` the target of given examples\n",
    "\n",
    "- We can predict the target of examples by calling `predict` on the classifier object. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['not A+', 'not A+', 'not A+', 'not A+', 'not A+', 'not A+',\n",
       "       'not A+', 'not A+', 'not A+', 'not A+', 'not A+', 'not A+',\n",
       "       'not A+', 'not A+', 'not A+', 'not A+', 'not A+', 'not A+',\n",
       "       'not A+', 'not A+', 'not A+'], dtype='<U6')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dummy_clf.predict(X) # predict using the trained classifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### `score` your model\n",
    "\n",
    "- How do you know how well your model is doing?\n",
    "- For classification problems, by default, `score` gives the **accuracy** of the model, i.e., proportion of correctly predicted targets.  \n",
    "\n",
    "    $accuracy = \\frac{\\text{correct predictions}}{\\text{total examples}}$   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The accuracy of the model on the training data: 0.524\n"
     ]
    }
   ],
   "source": [
    "print(\"The accuracy of the model on the training data: %0.3f\" % (dummy_clf.score(X, y)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "- Sometimes you will also see people reporting **error**, which is usually $1 - accuracy$ \n",
    "- `score` \n",
    "    - calls `predict` on `X` \n",
    "    - compares predictions with `y` (true targets)\n",
    "    - returns the accuracy in case of classification.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The error of the model on the training data: 0.476\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    \"The error of the model on the training data: %0.3f\" % (1 - dummy_clf.score(X, y))\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### `fit`, `predict` , and `score` summary\n",
    "\n",
    "Here is the general pattern when we build ML models using `sklearn`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5238095238095238\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array(['not A+', 'not A+'], dtype='<U6')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create `X` and `y` from the given data\n",
    "X = classification_df.drop(columns=[\"quiz2\"])\n",
    "y = classification_df[\"quiz2\"]\n",
    "\n",
    "clf = DummyClassifier(strategy=\"most_frequent\") # Create a class object\n",
    "clf.fit(X, y) # Train/fit the model\n",
    "print(clf.score(X, y)) # Assess the model\n",
    "\n",
    "new_examples = [[0, 1, 92, 90, 95, 93, 92], [1, 1, 92, 93, 94, 92]]\n",
    "clf.predict(new_examples) # Predict on some new data using the trained model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "```{note} \n",
    "You'll be exploring dummy classifier in your lab!\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### [`DummyRegressor`](https://scikit-learn.org/0.15/modules/generated/sklearn.dummy.DummyRegressor.html)\n",
    "\n",
    "You can also do the same thing for regression problems using `DummyRegressor`, which predicts mean, median, or constant value of the training set for all examples. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "- Let's build a regression baseline model using `sklearn`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([86.28571429, 86.28571429])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.dummy import DummyRegressor\n",
    "\n",
    "regression_df = pd.read_csv(\"data/quiz2-grade-toy-regression.csv\") # Read data \n",
    "X = regression_df.drop(columns=[\"quiz2\"]) # Create `X` and `y` from the given data\n",
    "y = regression_df[\"quiz2\"]\n",
    "reg = DummyRegressor() # Create a class object\n",
    "reg.fit(X, y) # Train/fit the model\n",
    "reg.score(X, y) # Assess the model\n",
    "new_examples = [[0, 1, 92, 90, 95, 93, 92], [1, 1, 92, 93, 94, 92]]\n",
    "reg.predict(new_examples) # Predict on some new data using the trained model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "- The `fit` and `predict` paradigms similar to classification. The `score` method in the context of regression returns somethings called [$R^2$ score](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.r2_score.html#sklearn.metrics.r2_score). (More on this in later videos.)     \n",
    "    - The maximum $R^2$ is 1 for perfect predictions. \n",
    "    - For `DummyRegressor` it returns the mean of the `y` values.   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg.score(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### ❓❓ Questions for you"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{admonition} Exercise 2.4\n",
    "1. Order the steps below to build ML models using `sklearn`. \n",
    "    a. `score` to evaluate the performance of a given model\n",
    "    b. `predict` on new examples \n",
    "    c. Creating a model instance\n",
    "    d. Creating `X` and `y` \n",
    "    e. `fit`\n",
    "2. `predict` takes only `X` as argument whereas `fit` and `score` take both `X` and `y` as arguments. True or False. \n",
    "3. Have you ever played [20-questions game](https://en.wikipedia.org/wiki/Twenty_questions)? If yes, think about how do you decide what question to ask next? \n",
    "\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{admonition} Exercise 2.2.1: Solutions!\n",
    ":class: tip, dropdown\n",
    "TBD\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br><br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Decision trees [[video](https://youtu.be/Hcf19Ij35rA)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{seealso}\n",
    "Check out [the accompanying video](https://youtu.be/Hcf19Ij35rA) on this material. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Writing a traditional program to predict quiz2 grade\n",
    "\n",
    "- Can we do better than the baseline? \n",
    "- Forget about ML for a second. If you are asked to write a program to predict whether a student gets an A+ or not in quiz2, how would you go for it?  \n",
    "- For simplicity, let's binarize the feature values. \n",
    "\n",
    "![](img/quiz2-grade-toy.png)\n",
    "\n",
    "<!-- <img src=\"img/quiz2-grade-toy.png\" height=\"700\" width=\"700\">  -->\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "- Is there a pattern that distinguishes yes's from no's and what does the pattern say about today? \n",
    "- How about a rule-based algorithm with a number of *if else* statements?  \n",
    "    ```\n",
    "    if class_attendance == 1 and quiz1 == 1:\n",
    "        quiz2 == \"A+\"\n",
    "    elif class_attendance == 1 and lab3 == 1 and lab4 == 1:\n",
    "        quiz2 == \"A+\"\n",
    "    ...\n",
    "    ```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "- How many possible rule combinations there could be with the given 7 binary features? \n",
    "    - Gets unwieldy pretty quickly "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Decision tree algorithm \n",
    "\n",
    "- A machine learning algorithm to derive such rules from data in a principled way.  \n",
    "- Have you ever played [20-questions game](https://en.wikipedia.org/wiki/Twenty_questions)? Decision trees are based on the same idea! \n",
    "- Let's `fit` a decision tree using `scikit-learn` and `predict` with it.\n",
    "- Recall that `scikit-learn` uses the term `fit` for training or learning and uses `predict` for prediction. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Building decision trees with `sklearn`\n",
    "\n",
    "Let's binarize our toy dataset for simplicity. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1\n",
       "0              1                 1     1     1     0     1      1\n",
       "1              1                 0     1     1     0     0      1\n",
       "2              0                 0     0     0     0     0      0\n",
       "3              0                 1     1     1     1     1      0\n",
       "4              0                 1     0     0     1     1      0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classification_df = pd.read_csv(\"data/quiz2-grade-toy-classification.csv\")\n",
    "X = classification_df.drop(columns=[\"quiz2\"])\n",
    "y = classification_df[\"quiz2\"]\n",
    "\n",
    "X_binary = X.copy()\n",
    "columns = [\"lab1\", \"lab2\", \"lab3\", \"lab4\", \"quiz1\"]\n",
    "for col in columns:\n",
    "    X_binary[col] = X_binary[col].apply(lambda x: 1 if x >= 90 else 0)\n",
    "X_binary.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        A+\n",
       "1    not A+\n",
       "2    not A+\n",
       "3        A+\n",
       "4        A+\n",
       "Name: quiz2, dtype: object"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### `DummyClassifier` on quiz2 grade prediction toy dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5238095238095238"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dummy_clf = DummyClassifier(strategy=\"most_frequent\")\n",
    "dummy_clf.fit(X_binary, y)\n",
    "dummy_clf.score(X_binary, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "#### `DecisionTreeClassifier` on quiz2 grade prediction toy dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9047619047619048"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "model = DecisionTreeClassifier() # Create a decision tree\n",
    "model.fit(X_binary, y) # Fit a decision tree\n",
    "model.score(X_binary, y) # Assess the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The decision tree classifier is giving much higher accuracy than the dummy classifier. That's good news! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.48.0 (0)\n",
       " -->\n",
       "<!-- Title: Tree Pages: 1 -->\n",
       "<svg width=\"679pt\" height=\"562pt\"\n",
       " viewBox=\"0.00 0.00 678.50 562.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 558)\">\n",
       "<title>Tree</title>\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-558 674.5,-558 674.5,4 -4,4\"/>\n",
       "<!-- 0 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>0</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"305,-554 208,-554 208,-516 305,-516 305,-554\"/>\n",
       "<text text-anchor=\"middle\" x=\"256.5\" y=\"-538.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"256.5\" y=\"-523.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>1</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"239,-480 142,-480 142,-442 239,-442 239,-480\"/>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-464.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-449.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;1 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>0&#45;&gt;1</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M239.85,-515.83C232.09,-507.37 222.72,-497.15 214.24,-487.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"216.72,-485.42 207.38,-480.41 211.56,-490.15 216.72,-485.42\"/>\n",
       "<text text-anchor=\"middle\" x=\"206.29\" y=\"-501.69\" font-family=\"Times,serif\" font-size=\"14.00\">True</text>\n",
       "</g>\n",
       "<!-- 8 -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>8</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"363,-480 282,-480 282,-442 363,-442 363,-480\"/>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-464.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab2 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-449.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;8 -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>0&#45;&gt;8</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M273.15,-515.83C280.91,-507.37 290.28,-497.15 298.76,-487.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"301.44,-490.15 305.62,-480.41 296.28,-485.42 301.44,-490.15\"/>\n",
       "<text text-anchor=\"middle\" x=\"306.71\" y=\"-501.69\" font-family=\"Times,serif\" font-size=\"14.00\">False</text>\n",
       "</g>\n",
       "<!-- 2 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>2</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"97,-405 0,-405 0,-369 97,-369 97,-405\"/>\n",
       "<text text-anchor=\"middle\" x=\"48.5\" y=\"-383.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1&#45;&gt;2 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>1&#45;&gt;2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M154.67,-441.83C135.43,-432.08 111.58,-419.99 91.37,-409.74\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"92.87,-406.57 82.37,-405.17 89.7,-412.82 92.87,-406.57\"/>\n",
       "</g>\n",
       "<!-- 3 -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>3</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"266,-406 115,-406 115,-368 266,-368 266,-406\"/>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-390.8\" font-family=\"Times,serif\" font-size=\"14.00\">class_attendance &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-375.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 1&#45;&gt;3 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>1&#45;&gt;3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M190.5,-441.83C190.5,-434.13 190.5,-424.97 190.5,-416.42\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"194,-416.41 190.5,-406.41 187,-416.41 194,-416.41\"/>\n",
       "</g>\n",
       "<!-- 4 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>4</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"192,-332 95,-332 95,-294 192,-294 192,-332\"/>\n",
       "<text text-anchor=\"middle\" x=\"143.5\" y=\"-316.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"143.5\" y=\"-301.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;4 -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>3&#45;&gt;4</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M178.64,-367.83C173.34,-359.71 166.98,-349.96 161.13,-341.01\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"163.92,-338.87 155.52,-332.41 158.06,-342.7 163.92,-338.87\"/>\n",
       "</g>\n",
       "<!-- 7 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>7</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"286.5,-331 210.5,-331 210.5,-295 286.5,-295 286.5,-331\"/>\n",
       "<text text-anchor=\"middle\" x=\"248.5\" y=\"-309.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;7 -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>3&#45;&gt;7</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M205.14,-367.83C212.09,-359.2 220.52,-348.73 228.09,-339.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"230.92,-341.4 234.47,-331.42 225.47,-337.01 230.92,-341.4\"/>\n",
       "</g>\n",
       "<!-- 5 -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>5</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"129.5,-257 53.5,-257 53.5,-221 129.5,-221 129.5,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"91.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 4&#45;&gt;5 -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>4&#45;&gt;5</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M130.38,-293.83C124.2,-285.28 116.74,-274.94 110,-265.62\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"112.77,-263.48 104.08,-257.42 107.1,-267.57 112.77,-263.48\"/>\n",
       "</g>\n",
       "<!-- 6 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>6</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"245,-257 148,-257 148,-221 245,-221 245,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"196.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 4&#45;&gt;6 -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>4&#45;&gt;6</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M156.87,-293.83C163.17,-285.28 170.78,-274.94 177.64,-265.62\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"180.57,-267.55 183.68,-257.42 174.93,-263.4 180.57,-267.55\"/>\n",
       "</g>\n",
       "<!-- 9 -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>9</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"360.5,-405 284.5,-405 284.5,-369 360.5,-369 360.5,-405\"/>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-383.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;9 -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>8&#45;&gt;9</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M322.5,-441.83C322.5,-433.89 322.5,-424.41 322.5,-415.63\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"326,-415.42 322.5,-405.42 319,-415.42 326,-415.42\"/>\n",
       "</g>\n",
       "<!-- 10 -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>10</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"466.5,-406 378.5,-406 378.5,-368 466.5,-368 466.5,-406\"/>\n",
       "<text text-anchor=\"middle\" x=\"422.5\" y=\"-390.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"422.5\" y=\"-375.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;10 -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>8&#45;&gt;10</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M347.73,-441.83C360.29,-432.79 375.63,-421.75 389.14,-412.02\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"391.2,-414.85 397.27,-406.16 387.11,-409.17 391.2,-414.85\"/>\n",
       "</g>\n",
       "<!-- 11 -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>11</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"413,-332 316,-332 316,-294 413,-294 413,-332\"/>\n",
       "<text text-anchor=\"middle\" x=\"364.5\" y=\"-316.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"364.5\" y=\"-301.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;11 -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>10&#45;&gt;11</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M407.86,-367.83C401.18,-359.54 393.14,-349.56 385.81,-340.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"388.33,-338.01 379.33,-332.41 382.88,-342.4 388.33,-338.01\"/>\n",
       "</g>\n",
       "<!-- 20 -->\n",
       "<g id=\"node21\" class=\"node\">\n",
       "<title>20</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"507.5,-331 431.5,-331 431.5,-295 507.5,-295 507.5,-331\"/>\n",
       "<text text-anchor=\"middle\" x=\"469.5\" y=\"-309.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;20 -->\n",
       "<g id=\"edge20\" class=\"edge\">\n",
       "<title>10&#45;&gt;20</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M434.36,-367.83C439.88,-359.37 446.55,-349.15 452.59,-339.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"455.59,-341.7 458.13,-331.42 449.73,-337.88 455.59,-341.7\"/>\n",
       "</g>\n",
       "<!-- 12 -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>12</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"365.5,-257 289.5,-257 289.5,-221 365.5,-221 365.5,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"327.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 11&#45;&gt;12 -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>11&#45;&gt;12</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M355.16,-293.83C350.91,-285.55 345.78,-275.57 341.11,-266.47\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"344.13,-264.71 336.45,-257.42 337.91,-267.91 344.13,-264.71\"/>\n",
       "</g>\n",
       "<!-- 13 -->\n",
       "<g id=\"node14\" class=\"node\">\n",
       "<title>13</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"535,-258 384,-258 384,-220 535,-220 535,-258\"/>\n",
       "<text text-anchor=\"middle\" x=\"459.5\" y=\"-242.8\" font-family=\"Times,serif\" font-size=\"14.00\">class_attendance &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"459.5\" y=\"-227.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 11&#45;&gt;13 -->\n",
       "<g id=\"edge13\" class=\"edge\">\n",
       "<title>11&#45;&gt;13</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M388.47,-293.83C400.29,-284.88 414.7,-273.96 427.44,-264.3\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"429.68,-266.99 435.54,-258.16 425.45,-261.41 429.68,-266.99\"/>\n",
       "</g>\n",
       "<!-- 14 -->\n",
       "<g id=\"node15\" class=\"node\">\n",
       "<title>14</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"451,-183 354,-183 354,-147 451,-147 451,-183\"/>\n",
       "<text text-anchor=\"middle\" x=\"402.5\" y=\"-161.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 13&#45;&gt;14 -->\n",
       "<g id=\"edge14\" class=\"edge\">\n",
       "<title>13&#45;&gt;14</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M445.12,-219.83C438.28,-211.2 429.99,-200.73 422.55,-191.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"425.24,-189.09 416.29,-183.42 419.75,-193.43 425.24,-189.09\"/>\n",
       "</g>\n",
       "<!-- 15 -->\n",
       "<g id=\"node16\" class=\"node\">\n",
       "<title>15</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"566,-184 469,-184 469,-146 566,-146 566,-184\"/>\n",
       "<text text-anchor=\"middle\" x=\"517.5\" y=\"-168.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab4 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"517.5\" y=\"-153.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 13&#45;&gt;15 -->\n",
       "<g id=\"edge15\" class=\"edge\">\n",
       "<title>13&#45;&gt;15</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M474.14,-219.83C480.82,-211.54 488.86,-201.56 496.19,-192.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"499.12,-194.4 502.67,-184.41 493.67,-190.01 499.12,-194.4\"/>\n",
       "</g>\n",
       "<!-- 16 -->\n",
       "<g id=\"node17\" class=\"node\">\n",
       "<title>16</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"498,-109 401,-109 401,-73 498,-73 498,-109\"/>\n",
       "<text text-anchor=\"middle\" x=\"449.5\" y=\"-87.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 15&#45;&gt;16 -->\n",
       "<g id=\"edge16\" class=\"edge\">\n",
       "<title>15&#45;&gt;16</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M500.34,-145.83C492.02,-137.02 481.9,-126.31 472.89,-116.76\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"475.36,-114.28 465.95,-109.42 470.27,-119.09 475.36,-114.28\"/>\n",
       "</g>\n",
       "<!-- 17 -->\n",
       "<g id=\"node18\" class=\"node\">\n",
       "<title>17</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"655,-110 516,-110 516,-72 655,-72 655,-110\"/>\n",
       "<text text-anchor=\"middle\" x=\"585.5\" y=\"-94.8\" font-family=\"Times,serif\" font-size=\"14.00\">ml_experience &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"585.5\" y=\"-79.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 15&#45;&gt;17 -->\n",
       "<g id=\"edge17\" class=\"edge\">\n",
       "<title>15&#45;&gt;17</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M534.66,-145.83C542.65,-137.37 552.3,-127.15 561.04,-117.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"563.79,-120.09 568.11,-110.41 558.7,-115.28 563.79,-120.09\"/>\n",
       "</g>\n",
       "<!-- 18 -->\n",
       "<g id=\"node19\" class=\"node\">\n",
       "<title>18</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"576.5,-36 500.5,-36 500.5,0 576.5,0 576.5,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"538.5\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 17&#45;&gt;18 -->\n",
       "<g id=\"edge18\" class=\"edge\">\n",
       "<title>17&#45;&gt;18</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M573.4,-71.72C567.93,-63.46 561.38,-53.56 555.43,-44.58\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"558.19,-42.41 549.76,-36 552.36,-46.27 558.19,-42.41\"/>\n",
       "</g>\n",
       "<!-- 19 -->\n",
       "<g id=\"node20\" class=\"node\">\n",
       "<title>19</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"670.5,-36 594.5,-36 594.5,0 670.5,0 670.5,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"632.5\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 17&#45;&gt;19 -->\n",
       "<g id=\"edge19\" class=\"edge\">\n",
       "<title>17&#45;&gt;19</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M597.6,-71.72C603.07,-63.46 609.62,-53.56 615.57,-44.58\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"618.64,-46.27 621.24,-36 612.81,-42.41 618.64,-46.27\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.files.Source at 0x13b1c0040>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_tree(X_binary.columns, model) # model visualization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Some terminology related to trees \n",
    "\n",
    "Here is a commonly used terminology in a typical representation of decision trees. \n",
    "\n",
    "**A root node**\n",
    ": represents the first condition to check or question to ask\n",
    "\n",
    "**A branch**\n",
    ": connects a node (condition) to the next node (condition) in the tree. Each branch typically represents either true or false. \n",
    "\n",
    "**An internal node** \n",
    ": represents conditions within the tree\n",
    "\n",
    "**A leaf node**\n",
    ": represents the predicted class/value when the path from root to the leaf node is followed. \n",
    "\n",
    "**Tree depth**\n",
    ": The number of edges on the path from the root node to the farthest away leaf node."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### How does `predict` work? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1\n",
       "0              0                 1     0     0     1     1      1"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_example = np.array([[0, 1, 0, 0, 1, 1, 1]])\n",
    "pd.DataFrame(data=new_example, columns=X.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.48.0 (0)\n",
       " -->\n",
       "<!-- Title: Tree Pages: 1 -->\n",
       "<svg width=\"679pt\" height=\"562pt\"\n",
       " viewBox=\"0.00 0.00 678.50 562.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 558)\">\n",
       "<title>Tree</title>\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-558 674.5,-558 674.5,4 -4,4\"/>\n",
       "<!-- 0 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>0</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"305,-554 208,-554 208,-516 305,-516 305,-554\"/>\n",
       "<text text-anchor=\"middle\" x=\"256.5\" y=\"-538.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"256.5\" y=\"-523.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>1</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"239,-480 142,-480 142,-442 239,-442 239,-480\"/>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-464.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-449.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;1 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>0&#45;&gt;1</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M239.85,-515.83C232.09,-507.37 222.72,-497.15 214.24,-487.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"216.72,-485.42 207.38,-480.41 211.56,-490.15 216.72,-485.42\"/>\n",
       "<text text-anchor=\"middle\" x=\"206.29\" y=\"-501.69\" font-family=\"Times,serif\" font-size=\"14.00\">True</text>\n",
       "</g>\n",
       "<!-- 8 -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>8</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"363,-480 282,-480 282,-442 363,-442 363,-480\"/>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-464.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab2 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-449.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;8 -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>0&#45;&gt;8</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M273.15,-515.83C280.91,-507.37 290.28,-497.15 298.76,-487.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"301.44,-490.15 305.62,-480.41 296.28,-485.42 301.44,-490.15\"/>\n",
       "<text text-anchor=\"middle\" x=\"306.71\" y=\"-501.69\" font-family=\"Times,serif\" font-size=\"14.00\">False</text>\n",
       "</g>\n",
       "<!-- 2 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>2</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"97,-405 0,-405 0,-369 97,-369 97,-405\"/>\n",
       "<text text-anchor=\"middle\" x=\"48.5\" y=\"-383.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1&#45;&gt;2 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>1&#45;&gt;2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M154.67,-441.83C135.43,-432.08 111.58,-419.99 91.37,-409.74\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"92.87,-406.57 82.37,-405.17 89.7,-412.82 92.87,-406.57\"/>\n",
       "</g>\n",
       "<!-- 3 -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>3</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"266,-406 115,-406 115,-368 266,-368 266,-406\"/>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-390.8\" font-family=\"Times,serif\" font-size=\"14.00\">class_attendance &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-375.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 1&#45;&gt;3 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>1&#45;&gt;3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M190.5,-441.83C190.5,-434.13 190.5,-424.97 190.5,-416.42\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"194,-416.41 190.5,-406.41 187,-416.41 194,-416.41\"/>\n",
       "</g>\n",
       "<!-- 4 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>4</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"192,-332 95,-332 95,-294 192,-294 192,-332\"/>\n",
       "<text text-anchor=\"middle\" x=\"143.5\" y=\"-316.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"143.5\" y=\"-301.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;4 -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>3&#45;&gt;4</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M178.64,-367.83C173.34,-359.71 166.98,-349.96 161.13,-341.01\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"163.92,-338.87 155.52,-332.41 158.06,-342.7 163.92,-338.87\"/>\n",
       "</g>\n",
       "<!-- 7 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>7</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"286.5,-331 210.5,-331 210.5,-295 286.5,-295 286.5,-331\"/>\n",
       "<text text-anchor=\"middle\" x=\"248.5\" y=\"-309.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;7 -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>3&#45;&gt;7</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M205.14,-367.83C212.09,-359.2 220.52,-348.73 228.09,-339.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"230.92,-341.4 234.47,-331.42 225.47,-337.01 230.92,-341.4\"/>\n",
       "</g>\n",
       "<!-- 5 -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>5</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"129.5,-257 53.5,-257 53.5,-221 129.5,-221 129.5,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"91.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 4&#45;&gt;5 -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>4&#45;&gt;5</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M130.38,-293.83C124.2,-285.28 116.74,-274.94 110,-265.62\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"112.77,-263.48 104.08,-257.42 107.1,-267.57 112.77,-263.48\"/>\n",
       "</g>\n",
       "<!-- 6 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>6</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"245,-257 148,-257 148,-221 245,-221 245,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"196.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 4&#45;&gt;6 -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>4&#45;&gt;6</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M156.87,-293.83C163.17,-285.28 170.78,-274.94 177.64,-265.62\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"180.57,-267.55 183.68,-257.42 174.93,-263.4 180.57,-267.55\"/>\n",
       "</g>\n",
       "<!-- 9 -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>9</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"360.5,-405 284.5,-405 284.5,-369 360.5,-369 360.5,-405\"/>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-383.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;9 -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>8&#45;&gt;9</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M322.5,-441.83C322.5,-433.89 322.5,-424.41 322.5,-415.63\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"326,-415.42 322.5,-405.42 319,-415.42 326,-415.42\"/>\n",
       "</g>\n",
       "<!-- 10 -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>10</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"466.5,-406 378.5,-406 378.5,-368 466.5,-368 466.5,-406\"/>\n",
       "<text text-anchor=\"middle\" x=\"422.5\" y=\"-390.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"422.5\" y=\"-375.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;10 -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>8&#45;&gt;10</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M347.73,-441.83C360.29,-432.79 375.63,-421.75 389.14,-412.02\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"391.2,-414.85 397.27,-406.16 387.11,-409.17 391.2,-414.85\"/>\n",
       "</g>\n",
       "<!-- 11 -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>11</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"413,-332 316,-332 316,-294 413,-294 413,-332\"/>\n",
       "<text text-anchor=\"middle\" x=\"364.5\" y=\"-316.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"364.5\" y=\"-301.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;11 -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>10&#45;&gt;11</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M407.86,-367.83C401.18,-359.54 393.14,-349.56 385.81,-340.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"388.33,-338.01 379.33,-332.41 382.88,-342.4 388.33,-338.01\"/>\n",
       "</g>\n",
       "<!-- 20 -->\n",
       "<g id=\"node21\" class=\"node\">\n",
       "<title>20</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"507.5,-331 431.5,-331 431.5,-295 507.5,-295 507.5,-331\"/>\n",
       "<text text-anchor=\"middle\" x=\"469.5\" y=\"-309.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;20 -->\n",
       "<g id=\"edge20\" class=\"edge\">\n",
       "<title>10&#45;&gt;20</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M434.36,-367.83C439.88,-359.37 446.55,-349.15 452.59,-339.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"455.59,-341.7 458.13,-331.42 449.73,-337.88 455.59,-341.7\"/>\n",
       "</g>\n",
       "<!-- 12 -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>12</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"365.5,-257 289.5,-257 289.5,-221 365.5,-221 365.5,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"327.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 11&#45;&gt;12 -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>11&#45;&gt;12</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M355.16,-293.83C350.91,-285.55 345.78,-275.57 341.11,-266.47\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"344.13,-264.71 336.45,-257.42 337.91,-267.91 344.13,-264.71\"/>\n",
       "</g>\n",
       "<!-- 13 -->\n",
       "<g id=\"node14\" class=\"node\">\n",
       "<title>13</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"535,-258 384,-258 384,-220 535,-220 535,-258\"/>\n",
       "<text text-anchor=\"middle\" x=\"459.5\" y=\"-242.8\" font-family=\"Times,serif\" font-size=\"14.00\">class_attendance &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"459.5\" y=\"-227.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 11&#45;&gt;13 -->\n",
       "<g id=\"edge13\" class=\"edge\">\n",
       "<title>11&#45;&gt;13</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M388.47,-293.83C400.29,-284.88 414.7,-273.96 427.44,-264.3\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"429.68,-266.99 435.54,-258.16 425.45,-261.41 429.68,-266.99\"/>\n",
       "</g>\n",
       "<!-- 14 -->\n",
       "<g id=\"node15\" class=\"node\">\n",
       "<title>14</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"451,-183 354,-183 354,-147 451,-147 451,-183\"/>\n",
       "<text text-anchor=\"middle\" x=\"402.5\" y=\"-161.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 13&#45;&gt;14 -->\n",
       "<g id=\"edge14\" class=\"edge\">\n",
       "<title>13&#45;&gt;14</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M445.12,-219.83C438.28,-211.2 429.99,-200.73 422.55,-191.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"425.24,-189.09 416.29,-183.42 419.75,-193.43 425.24,-189.09\"/>\n",
       "</g>\n",
       "<!-- 15 -->\n",
       "<g id=\"node16\" class=\"node\">\n",
       "<title>15</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"566,-184 469,-184 469,-146 566,-146 566,-184\"/>\n",
       "<text text-anchor=\"middle\" x=\"517.5\" y=\"-168.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab4 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"517.5\" y=\"-153.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 13&#45;&gt;15 -->\n",
       "<g id=\"edge15\" class=\"edge\">\n",
       "<title>13&#45;&gt;15</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M474.14,-219.83C480.82,-211.54 488.86,-201.56 496.19,-192.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"499.12,-194.4 502.67,-184.41 493.67,-190.01 499.12,-194.4\"/>\n",
       "</g>\n",
       "<!-- 16 -->\n",
       "<g id=\"node17\" class=\"node\">\n",
       "<title>16</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"498,-109 401,-109 401,-73 498,-73 498,-109\"/>\n",
       "<text text-anchor=\"middle\" x=\"449.5\" y=\"-87.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 15&#45;&gt;16 -->\n",
       "<g id=\"edge16\" class=\"edge\">\n",
       "<title>15&#45;&gt;16</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M500.34,-145.83C492.02,-137.02 481.9,-126.31 472.89,-116.76\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"475.36,-114.28 465.95,-109.42 470.27,-119.09 475.36,-114.28\"/>\n",
       "</g>\n",
       "<!-- 17 -->\n",
       "<g id=\"node18\" class=\"node\">\n",
       "<title>17</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"655,-110 516,-110 516,-72 655,-72 655,-110\"/>\n",
       "<text text-anchor=\"middle\" x=\"585.5\" y=\"-94.8\" font-family=\"Times,serif\" font-size=\"14.00\">ml_experience &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"585.5\" y=\"-79.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 15&#45;&gt;17 -->\n",
       "<g id=\"edge17\" class=\"edge\">\n",
       "<title>15&#45;&gt;17</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M534.66,-145.83C542.65,-137.37 552.3,-127.15 561.04,-117.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"563.79,-120.09 568.11,-110.41 558.7,-115.28 563.79,-120.09\"/>\n",
       "</g>\n",
       "<!-- 18 -->\n",
       "<g id=\"node19\" class=\"node\">\n",
       "<title>18</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"576.5,-36 500.5,-36 500.5,0 576.5,0 576.5,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"538.5\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 17&#45;&gt;18 -->\n",
       "<g id=\"edge18\" class=\"edge\">\n",
       "<title>17&#45;&gt;18</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M573.4,-71.72C567.93,-63.46 561.38,-53.56 555.43,-44.58\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"558.19,-42.41 549.76,-36 552.36,-46.27 558.19,-42.41\"/>\n",
       "</g>\n",
       "<!-- 19 -->\n",
       "<g id=\"node20\" class=\"node\">\n",
       "<title>19</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"670.5,-36 594.5,-36 594.5,0 670.5,0 670.5,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"632.5\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 17&#45;&gt;19 -->\n",
       "<g id=\"edge19\" class=\"edge\">\n",
       "<title>17&#45;&gt;19</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M597.6,-71.72C603.07,-63.46 609.62,-53.56 615.57,-44.58\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"618.64,-46.27 621.24,-36 612.81,-42.41 618.64,-46.27\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.files.Source at 0x13b1792b0>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "display_tree(X_binary.columns, model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "What's the prediction for the new example?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['A+'], dtype=object)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.predict(new_example)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "In summary, given a learned tree and a test example, during prediction time,  \n",
    "- Start at the top of the tree. Ask binary questions at each node and follow the appropriate path in the tree. Once you are at a leaf node, you have the prediction. \n",
    "- Note that the model only considers the features which are in the learned tree and ignores all other features. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### How does `fit` work? \n",
    "\n",
    "- Decision tree is inspired by [20-questions game](https://en.wikipedia.org/wiki/Twenty_questions). \n",
    "- Each node either represents a question or an answer. The terminal nodes (called leaf nodes) represent answers. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_fruit_tree()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### How does `fit` work? \n",
    "\n",
    "- Which features are most useful for classification? \n",
    "- Minimize **impurity** at each question\n",
    "- Common criteria to minimize impurity: [gini index](https://scikit-learn.org/stable/modules/tree.html#classification-criteria), information gain, cross entropy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.48.0 (0)\n",
       " -->\n",
       "<!-- Title: Tree Pages: 1 -->\n",
       "<svg width=\"679pt\" height=\"562pt\"\n",
       " viewBox=\"0.00 0.00 678.50 562.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 558)\">\n",
       "<title>Tree</title>\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-558 674.5,-558 674.5,4 -4,4\"/>\n",
       "<!-- 0 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>0</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"305,-554 208,-554 208,-516 305,-516 305,-554\"/>\n",
       "<text text-anchor=\"middle\" x=\"256.5\" y=\"-538.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"256.5\" y=\"-523.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>1</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"239,-480 142,-480 142,-442 239,-442 239,-480\"/>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-464.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-449.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;1 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>0&#45;&gt;1</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M239.85,-515.83C232.09,-507.37 222.72,-497.15 214.24,-487.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"216.72,-485.42 207.38,-480.41 211.56,-490.15 216.72,-485.42\"/>\n",
       "<text text-anchor=\"middle\" x=\"206.29\" y=\"-501.69\" font-family=\"Times,serif\" font-size=\"14.00\">True</text>\n",
       "</g>\n",
       "<!-- 8 -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>8</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"363,-480 282,-480 282,-442 363,-442 363,-480\"/>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-464.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab2 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-449.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;8 -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>0&#45;&gt;8</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M273.15,-515.83C280.91,-507.37 290.28,-497.15 298.76,-487.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"301.44,-490.15 305.62,-480.41 296.28,-485.42 301.44,-490.15\"/>\n",
       "<text text-anchor=\"middle\" x=\"306.71\" y=\"-501.69\" font-family=\"Times,serif\" font-size=\"14.00\">False</text>\n",
       "</g>\n",
       "<!-- 2 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>2</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"97,-405 0,-405 0,-369 97,-369 97,-405\"/>\n",
       "<text text-anchor=\"middle\" x=\"48.5\" y=\"-383.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1&#45;&gt;2 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>1&#45;&gt;2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M154.67,-441.83C135.43,-432.08 111.58,-419.99 91.37,-409.74\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"92.87,-406.57 82.37,-405.17 89.7,-412.82 92.87,-406.57\"/>\n",
       "</g>\n",
       "<!-- 3 -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>3</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"266,-406 115,-406 115,-368 266,-368 266,-406\"/>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-390.8\" font-family=\"Times,serif\" font-size=\"14.00\">class_attendance &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"190.5\" y=\"-375.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 1&#45;&gt;3 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>1&#45;&gt;3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M190.5,-441.83C190.5,-434.13 190.5,-424.97 190.5,-416.42\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"194,-416.41 190.5,-406.41 187,-416.41 194,-416.41\"/>\n",
       "</g>\n",
       "<!-- 4 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>4</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"192,-332 95,-332 95,-294 192,-294 192,-332\"/>\n",
       "<text text-anchor=\"middle\" x=\"143.5\" y=\"-316.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"143.5\" y=\"-301.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;4 -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>3&#45;&gt;4</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M178.64,-367.83C173.34,-359.71 166.98,-349.96 161.13,-341.01\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"163.92,-338.87 155.52,-332.41 158.06,-342.7 163.92,-338.87\"/>\n",
       "</g>\n",
       "<!-- 7 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>7</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"286.5,-331 210.5,-331 210.5,-295 286.5,-295 286.5,-331\"/>\n",
       "<text text-anchor=\"middle\" x=\"248.5\" y=\"-309.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;7 -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>3&#45;&gt;7</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M205.14,-367.83C212.09,-359.2 220.52,-348.73 228.09,-339.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"230.92,-341.4 234.47,-331.42 225.47,-337.01 230.92,-341.4\"/>\n",
       "</g>\n",
       "<!-- 5 -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>5</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"129.5,-257 53.5,-257 53.5,-221 129.5,-221 129.5,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"91.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 4&#45;&gt;5 -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>4&#45;&gt;5</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M130.38,-293.83C124.2,-285.28 116.74,-274.94 110,-265.62\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"112.77,-263.48 104.08,-257.42 107.1,-267.57 112.77,-263.48\"/>\n",
       "</g>\n",
       "<!-- 6 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>6</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"245,-257 148,-257 148,-221 245,-221 245,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"196.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 4&#45;&gt;6 -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>4&#45;&gt;6</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M156.87,-293.83C163.17,-285.28 170.78,-274.94 177.64,-265.62\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"180.57,-267.55 183.68,-257.42 174.93,-263.4 180.57,-267.55\"/>\n",
       "</g>\n",
       "<!-- 9 -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>9</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"360.5,-405 284.5,-405 284.5,-369 360.5,-369 360.5,-405\"/>\n",
       "<text text-anchor=\"middle\" x=\"322.5\" y=\"-383.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;9 -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>8&#45;&gt;9</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M322.5,-441.83C322.5,-433.89 322.5,-424.41 322.5,-415.63\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"326,-415.42 322.5,-405.42 319,-415.42 326,-415.42\"/>\n",
       "</g>\n",
       "<!-- 10 -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>10</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"466.5,-406 378.5,-406 378.5,-368 466.5,-368 466.5,-406\"/>\n",
       "<text text-anchor=\"middle\" x=\"422.5\" y=\"-390.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"422.5\" y=\"-375.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;10 -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>8&#45;&gt;10</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M347.73,-441.83C360.29,-432.79 375.63,-421.75 389.14,-412.02\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"391.2,-414.85 397.27,-406.16 387.11,-409.17 391.2,-414.85\"/>\n",
       "</g>\n",
       "<!-- 11 -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>11</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"413,-332 316,-332 316,-294 413,-294 413,-332\"/>\n",
       "<text text-anchor=\"middle\" x=\"364.5\" y=\"-316.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab1 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"364.5\" y=\"-301.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;11 -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>10&#45;&gt;11</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M407.86,-367.83C401.18,-359.54 393.14,-349.56 385.81,-340.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"388.33,-338.01 379.33,-332.41 382.88,-342.4 388.33,-338.01\"/>\n",
       "</g>\n",
       "<!-- 20 -->\n",
       "<g id=\"node21\" class=\"node\">\n",
       "<title>20</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"507.5,-331 431.5,-331 431.5,-295 507.5,-295 507.5,-331\"/>\n",
       "<text text-anchor=\"middle\" x=\"469.5\" y=\"-309.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;20 -->\n",
       "<g id=\"edge20\" class=\"edge\">\n",
       "<title>10&#45;&gt;20</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M434.36,-367.83C439.88,-359.37 446.55,-349.15 452.59,-339.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"455.59,-341.7 458.13,-331.42 449.73,-337.88 455.59,-341.7\"/>\n",
       "</g>\n",
       "<!-- 12 -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>12</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"365.5,-257 289.5,-257 289.5,-221 365.5,-221 365.5,-257\"/>\n",
       "<text text-anchor=\"middle\" x=\"327.5\" y=\"-235.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 11&#45;&gt;12 -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>11&#45;&gt;12</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M355.16,-293.83C350.91,-285.55 345.78,-275.57 341.11,-266.47\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"344.13,-264.71 336.45,-257.42 337.91,-267.91 344.13,-264.71\"/>\n",
       "</g>\n",
       "<!-- 13 -->\n",
       "<g id=\"node14\" class=\"node\">\n",
       "<title>13</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"535,-258 384,-258 384,-220 535,-220 535,-258\"/>\n",
       "<text text-anchor=\"middle\" x=\"459.5\" y=\"-242.8\" font-family=\"Times,serif\" font-size=\"14.00\">class_attendance &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"459.5\" y=\"-227.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 11&#45;&gt;13 -->\n",
       "<g id=\"edge13\" class=\"edge\">\n",
       "<title>11&#45;&gt;13</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M388.47,-293.83C400.29,-284.88 414.7,-273.96 427.44,-264.3\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"429.68,-266.99 435.54,-258.16 425.45,-261.41 429.68,-266.99\"/>\n",
       "</g>\n",
       "<!-- 14 -->\n",
       "<g id=\"node15\" class=\"node\">\n",
       "<title>14</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"451,-183 354,-183 354,-147 451,-147 451,-183\"/>\n",
       "<text text-anchor=\"middle\" x=\"402.5\" y=\"-161.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 13&#45;&gt;14 -->\n",
       "<g id=\"edge14\" class=\"edge\">\n",
       "<title>13&#45;&gt;14</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M445.12,-219.83C438.28,-211.2 429.99,-200.73 422.55,-191.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"425.24,-189.09 416.29,-183.42 419.75,-193.43 425.24,-189.09\"/>\n",
       "</g>\n",
       "<!-- 15 -->\n",
       "<g id=\"node16\" class=\"node\">\n",
       "<title>15</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"566,-184 469,-184 469,-146 566,-146 566,-184\"/>\n",
       "<text text-anchor=\"middle\" x=\"517.5\" y=\"-168.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab4 &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"517.5\" y=\"-153.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 13&#45;&gt;15 -->\n",
       "<g id=\"edge15\" class=\"edge\">\n",
       "<title>13&#45;&gt;15</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M474.14,-219.83C480.82,-211.54 488.86,-201.56 496.19,-192.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"499.12,-194.4 502.67,-184.41 493.67,-190.01 499.12,-194.4\"/>\n",
       "</g>\n",
       "<!-- 16 -->\n",
       "<g id=\"node17\" class=\"node\">\n",
       "<title>16</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"498,-109 401,-109 401,-73 498,-73 498,-109\"/>\n",
       "<text text-anchor=\"middle\" x=\"449.5\" y=\"-87.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 15&#45;&gt;16 -->\n",
       "<g id=\"edge16\" class=\"edge\">\n",
       "<title>15&#45;&gt;16</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M500.34,-145.83C492.02,-137.02 481.9,-126.31 472.89,-116.76\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"475.36,-114.28 465.95,-109.42 470.27,-119.09 475.36,-114.28\"/>\n",
       "</g>\n",
       "<!-- 17 -->\n",
       "<g id=\"node18\" class=\"node\">\n",
       "<title>17</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"655,-110 516,-110 516,-72 655,-72 655,-110\"/>\n",
       "<text text-anchor=\"middle\" x=\"585.5\" y=\"-94.8\" font-family=\"Times,serif\" font-size=\"14.00\">ml_experience &lt;= 0.5</text>\n",
       "<text text-anchor=\"middle\" x=\"585.5\" y=\"-79.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 15&#45;&gt;17 -->\n",
       "<g id=\"edge17\" class=\"edge\">\n",
       "<title>15&#45;&gt;17</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M534.66,-145.83C542.65,-137.37 552.3,-127.15 561.04,-117.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"563.79,-120.09 568.11,-110.41 558.7,-115.28 563.79,-120.09\"/>\n",
       "</g>\n",
       "<!-- 18 -->\n",
       "<g id=\"node19\" class=\"node\">\n",
       "<title>18</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"576.5,-36 500.5,-36 500.5,0 576.5,0 576.5,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"538.5\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 17&#45;&gt;18 -->\n",
       "<g id=\"edge18\" class=\"edge\">\n",
       "<title>17&#45;&gt;18</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M573.4,-71.72C567.93,-63.46 561.38,-53.56 555.43,-44.58\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"558.19,-42.41 549.76,-36 552.36,-46.27 558.19,-42.41\"/>\n",
       "</g>\n",
       "<!-- 19 -->\n",
       "<g id=\"node20\" class=\"node\">\n",
       "<title>19</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"670.5,-36 594.5,-36 594.5,0 670.5,0 670.5,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"632.5\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 17&#45;&gt;19 -->\n",
       "<g id=\"edge19\" class=\"edge\">\n",
       "<title>17&#45;&gt;19</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M597.6,-71.72C603.07,-63.46 609.62,-53.56 615.57,-44.58\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"618.64,-46.27 621.24,-36 612.81,-42.41 618.64,-46.27\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.files.Source at 0x13b1c0370>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "model = DecisionTreeClassifier() # Create a decision tree\n",
    "model.fit(X_binary, y) # Fit a decision tree\n",
    "display_tree(X_binary.columns, model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "```{admonition} Warning \n",
    ":class: warning \n",
    "We won't go through **how** it does this - that's CPSC 340. But it's worth noting that it support two types of inputs: \n",
    "    1. Categorical (e.g., Yes/No or more options, as shown in the tree above)\n",
    "    2. Numeric (a number)In the numeric case, the decision tree algorithm also picks the _threshold_. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Decision trees with continuous features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>83</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1\n",
       "0              1                 1    92    93    84    91     92\n",
       "1              1                 0    94    90    80    83     91\n",
       "2              0                 0    78    85    83    80     80\n",
       "3              0                 1    91    94    92    91     89\n",
       "4              0                 1    77    83    90    92     85"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.48.0 (0)\n",
       " -->\n",
       "<!-- Title: Tree Pages: 1 -->\n",
       "<svg width=\"517pt\" height=\"414pt\"\n",
       " viewBox=\"0.00 0.00 517.00 414.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 410)\">\n",
       "<title>Tree</title>\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-410 513,-410 513,4 -4,4\"/>\n",
       "<!-- 0 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>0</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"191.5,-406 94.5,-406 94.5,-368 191.5,-368 191.5,-406\"/>\n",
       "<text text-anchor=\"middle\" x=\"143\" y=\"-390.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"143\" y=\"-375.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>1</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"134.5,-331 37.5,-331 37.5,-295 134.5,-295 134.5,-331\"/>\n",
       "<text text-anchor=\"middle\" x=\"86\" y=\"-309.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;1 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>0&#45;&gt;1</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M128.62,-367.83C121.78,-359.2 113.49,-348.73 106.05,-339.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"108.74,-337.09 99.79,-331.42 103.25,-341.43 108.74,-337.09\"/>\n",
       "<text text-anchor=\"middle\" x=\"96.9\" y=\"-352.55\" font-family=\"Times,serif\" font-size=\"14.00\">True</text>\n",
       "</g>\n",
       "<!-- 2 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>2</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"247.5,-332 152.5,-332 152.5,-294 247.5,-294 247.5,-332\"/>\n",
       "<text text-anchor=\"middle\" x=\"200\" y=\"-316.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"200\" y=\"-301.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;2 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>0&#45;&gt;2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M157.38,-367.83C163.95,-359.54 171.85,-349.56 179.06,-340.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"181.96,-342.43 185.42,-332.41 176.47,-338.08 181.96,-342.43\"/>\n",
       "<text text-anchor=\"middle\" x=\"188.31\" y=\"-353.55\" font-family=\"Times,serif\" font-size=\"14.00\">False</text>\n",
       "</g>\n",
       "<!-- 3 -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>3</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"191.5,-258 94.5,-258 94.5,-220 191.5,-220 191.5,-258\"/>\n",
       "<text text-anchor=\"middle\" x=\"143\" y=\"-242.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab2 &lt;= 88.0</text>\n",
       "<text text-anchor=\"middle\" x=\"143\" y=\"-227.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 2&#45;&gt;3 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>2&#45;&gt;3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M185.62,-293.83C179.05,-285.54 171.15,-275.56 163.94,-266.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"166.53,-264.08 157.58,-258.41 161.04,-268.43 166.53,-264.08\"/>\n",
       "</g>\n",
       "<!-- 6 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>6</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"302,-258 214,-258 214,-220 302,-220 302,-258\"/>\n",
       "<text text-anchor=\"middle\" x=\"258\" y=\"-242.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab4 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"258\" y=\"-227.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 2&#45;&gt;6 -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>2&#45;&gt;6</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M214.64,-293.83C221.32,-285.54 229.36,-275.56 236.69,-266.45\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"239.62,-268.4 243.17,-258.41 234.17,-264.01 239.62,-268.4\"/>\n",
       "</g>\n",
       "<!-- 4 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>4</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"76,-183 0,-183 0,-147 76,-147 76,-183\"/>\n",
       "<text text-anchor=\"middle\" x=\"38\" y=\"-161.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;4 -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>3&#45;&gt;4</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M116.5,-219.83C102.79,-210.43 85.91,-198.86 71.33,-188.85\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"73.27,-185.94 63.04,-183.17 69.31,-191.71 73.27,-185.94\"/>\n",
       "</g>\n",
       "<!-- 5 -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>5</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"191.5,-183 94.5,-183 94.5,-147 191.5,-147 191.5,-183\"/>\n",
       "<text text-anchor=\"middle\" x=\"143\" y=\"-161.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;5 -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>3&#45;&gt;5</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M143,-219.83C143,-211.89 143,-202.41 143,-193.63\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"146.5,-193.42 143,-183.42 139.5,-193.42 146.5,-193.42\"/>\n",
       "</g>\n",
       "<!-- 7 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>7</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"306.5,-183 209.5,-183 209.5,-147 306.5,-147 306.5,-183\"/>\n",
       "<text text-anchor=\"middle\" x=\"258\" y=\"-161.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 6&#45;&gt;7 -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>6&#45;&gt;7</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M258,-219.83C258,-211.89 258,-202.41 258,-193.63\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"261.5,-193.42 258,-183.42 254.5,-193.42 261.5,-193.42\"/>\n",
       "</g>\n",
       "<!-- 8 -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>8</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"413,-184 325,-184 325,-146 413,-146 413,-184\"/>\n",
       "<text text-anchor=\"middle\" x=\"369\" y=\"-168.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab2 &lt;= 94.5</text>\n",
       "<text text-anchor=\"middle\" x=\"369\" y=\"-153.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 6&#45;&gt;8 -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>6&#45;&gt;8</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M286.01,-219.83C300.08,-210.71 317.29,-199.54 332.39,-189.74\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"334.51,-192.54 341,-184.16 330.7,-186.67 334.51,-192.54\"/>\n",
       "</g>\n",
       "<!-- 9 -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>9</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"357,-109 281,-109 281,-73 357,-73 357,-109\"/>\n",
       "<text text-anchor=\"middle\" x=\"319\" y=\"-87.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;9 -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>8&#45;&gt;9</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M356.38,-145.83C350.51,-137.37 343.41,-127.15 336.99,-117.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"339.67,-115.64 331.1,-109.42 333.92,-119.63 339.67,-115.64\"/>\n",
       "</g>\n",
       "<!-- 10 -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>10</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"463,-110 375,-110 375,-72 463,-72 463,-110\"/>\n",
       "<text text-anchor=\"middle\" x=\"419\" y=\"-94.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 95.0</text>\n",
       "<text text-anchor=\"middle\" x=\"419\" y=\"-79.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;10 -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>8&#45;&gt;10</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M381.62,-145.83C387.32,-137.62 394.17,-127.76 400.44,-118.73\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"403.38,-120.62 406.21,-110.41 397.63,-116.63 403.38,-120.62\"/>\n",
       "</g>\n",
       "<!-- 11 -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>11</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"414.5,-36 317.5,-36 317.5,0 414.5,0 414.5,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"366\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;11 -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>10&#45;&gt;11</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M405.35,-71.72C399.06,-63.29 391.49,-53.15 384.68,-44.02\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"387.48,-41.92 378.69,-36 381.87,-46.11 387.48,-41.92\"/>\n",
       "</g>\n",
       "<!-- 12 -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>12</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"509,-36 433,-36 433,0 509,0 509,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"471\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;12 -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>10&#45;&gt;12</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M432.39,-71.72C438.5,-63.37 445.84,-53.35 452.47,-44.3\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"455.46,-46.14 458.55,-36 449.81,-42 455.46,-46.14\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.files.Source at 0x13b2aacd0>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = DecisionTreeClassifier()\n",
    "model.fit(X, y)\n",
    "display_tree(X.columns, model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Decision tree for regression problems\n",
    "\n",
    "- We can also use decision tree algorithm for regression. \n",
    "- Instead of gini, we use [some other criteria](https://scikit-learn.org/stable/modules/tree.html#mathematical-formulation) for splitting. A common one is mean squared error (MSE). (More on this in later videos.)\n",
    "- `scikit-learn` supports regression using decision trees with `DecisionTreeRegressor` \n",
    "    - `fit` and `predict` paradigms similar to classification\n",
    "    - `score` returns somethings called [$R^2$ score](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.r2_score.html#sklearn.metrics.r2_score).     \n",
    "        - The maximum $R^2$ is 1 for perfect predictions. \n",
    "        - It can be negative which is very bad (worse than `DummyRegressor`). \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "      <th>quiz2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "      <td>92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>83</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1  quiz2\n",
       "0              1                 1    92    93    84    91     92     90\n",
       "1              1                 0    94    90    80    83     91     84\n",
       "2              0                 0    78    85    83    80     80     82\n",
       "3              0                 1    91    94    92    91     89     92\n",
       "4              0                 1    77    83    90    92     85     90"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regression_df = pd.read_csv(\"data/quiz2-grade-toy-regression.csv\")\n",
    "regression_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R^2 score on the training data: 0.989\n",
      "\n",
      "\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>ml_experience</th>\n",
       "      <th>class_attendance</th>\n",
       "      <th>lab1</th>\n",
       "      <th>lab2</th>\n",
       "      <th>lab3</th>\n",
       "      <th>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "      <th>quiz2</th>\n",
       "      <th>predicted_quiz2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92</td>\n",
       "      <td>93</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "      <td>90</td>\n",
       "      <td>90.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>94</td>\n",
       "      <td>90</td>\n",
       "      <td>80</td>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "      <td>84</td>\n",
       "      <td>83.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>85</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "      <td>82</td>\n",
       "      <td>83.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>91</td>\n",
       "      <td>94</td>\n",
       "      <td>92</td>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "      <td>92</td>\n",
       "      <td>92.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>83</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "      <td>90</td>\n",
       "      <td>90.333333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ml_experience  class_attendance  lab1  lab2  lab3  lab4  quiz1  quiz2  \\\n",
       "0              1                 1    92    93    84    91     92     90   \n",
       "1              1                 0    94    90    80    83     91     84   \n",
       "2              0                 0    78    85    83    80     80     82   \n",
       "3              0                 1    91    94    92    91     89     92   \n",
       "4              0                 1    77    83    90    92     85     90   \n",
       "\n",
       "   predicted_quiz2  \n",
       "0        90.333333  \n",
       "1        83.000000  \n",
       "2        83.000000  \n",
       "3        92.000000  \n",
       "4        90.333333  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = regression_df.drop([\"quiz2\"], axis=1)\n",
    "y = regression_df[\"quiz2\"]\n",
    "\n",
    "depth = 2\n",
    "reg_model = DecisionTreeRegressor(max_depth=depth)\n",
    "reg_model.fit(X, y); \n",
    "regression_df[\"predicted_quiz2\"] = reg_model.predict(X)\n",
    "print(\"R^2 score on the training data: %0.3f\\n\\n\" % (reg_model.score(X, y)))\n",
    "regression_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### ❓❓ Questions for you to ponder on"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "```{admonition} Exercise 2.5\n",
    "1. Should change in features (i.e., binarizing features above) change `DummyClassifier` predictions? \n",
    "2. When you play 20-questions game, how do you pick the next question to ask?\n",
    "``` "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "```{admonition} Exercise 2.6 True or False \n",
    "1. For the decision tree algorithm to work, the feature values must be numeric.  \n",
    "2. For the decision tree algorithm to work, the target values must be numeric.\n",
    "3. The decision tree algorithm creates balanced decision trees. \n",
    "``` "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Break (5 min)\n",
    "\n",
    "![](img/eva-coffee.png)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br><br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## More terminology [[video](https://youtu.be/KEtsfXn4w2E)]\n",
    "\n",
    "- Parameters and hyperparameters\n",
    "- Decision boundary \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{seealso}\n",
    "Check out [the accompanying video](https://youtu.be/KEtsfXn4w2E) on this material. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Parameters \n",
    "\n",
    "- The decision tree algorithm primarily learns two things: \n",
    "    - the best feature to split on\n",
    "    - the threshold for the feature to split on at each node\n",
    "- These are called **parameters** of the decision tree model.  \n",
    "- When predicting on new examples, we need parameters of the model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "classification_df = pd.read_csv(\"data/quiz2-grade-toy-classification.csv\")\n",
    "X = classification_df.drop(columns=[\"quiz2\"])\n",
    "y = classification_df[\"quiz2\"]\n",
    "model = DecisionTreeClassifier()\n",
    "model.fit(X, y);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.48.0 (0)\n",
       " -->\n",
       "<!-- Title: Tree Pages: 1 -->\n",
       "<svg width=\"570pt\" height=\"491pt\"\n",
       " viewBox=\"0.00 0.00 570.00 491.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 487)\">\n",
       "<title>Tree</title>\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-487 566,-487 566,4 -4,4\"/>\n",
       "<!-- 0 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>0</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"219,-483 107,-483 107,-430 219,-430 219,-483\"/>\n",
       "<text text-anchor=\"middle\" x=\"163\" y=\"-467.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"163\" y=\"-452.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [10, 11]</text>\n",
       "<text text-anchor=\"middle\" x=\"163\" y=\"-437.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>1</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"152,-386.5 54,-386.5 54,-348.5 152,-348.5 152,-386.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"103\" y=\"-371.3\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [0, 6]</text>\n",
       "<text text-anchor=\"middle\" x=\"103\" y=\"-356.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;1 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>0&#45;&gt;1</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M145.32,-429.87C137.74,-418.87 128.89,-406.03 121.24,-394.95\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"123.98,-392.75 115.42,-386.51 118.21,-396.73 123.98,-392.75\"/>\n",
       "<text text-anchor=\"middle\" x=\"110.9\" y=\"-407.4\" font-family=\"Times,serif\" font-size=\"14.00\">True</text>\n",
       "</g>\n",
       "<!-- 2 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>2</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"275.5,-394 170.5,-394 170.5,-341 275.5,-341 275.5,-394\"/>\n",
       "<text text-anchor=\"middle\" x=\"223\" y=\"-378.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"223\" y=\"-363.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [10, 5]</text>\n",
       "<text text-anchor=\"middle\" x=\"223\" y=\"-348.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;2 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>0&#45;&gt;2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M180.68,-429.87C186.55,-421.36 193.17,-411.75 199.41,-402.7\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"202.45,-404.47 205.24,-394.25 196.68,-400.49 202.45,-404.47\"/>\n",
       "<text text-anchor=\"middle\" x=\"209.76\" y=\"-415.14\" font-family=\"Times,serif\" font-size=\"14.00\">False</text>\n",
       "</g>\n",
       "<!-- 3 -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>3</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"214,-305 116,-305 116,-252 214,-252 214,-305\"/>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-289.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab1 &lt;= 97.5</text>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-274.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [1, 3]</text>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-259.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 2&#45;&gt;3 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>2&#45;&gt;3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M205.91,-340.87C200.24,-332.36 193.83,-322.75 187.8,-313.7\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"190.63,-311.63 182.17,-305.25 184.8,-315.51 190.63,-311.63\"/>\n",
       "</g>\n",
       "<!-- 6 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>6</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"330,-305 232,-305 232,-252 330,-252 330,-305\"/>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-289.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab4 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-274.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [9, 2]</text>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-259.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 2&#45;&gt;6 -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>2&#45;&gt;6</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M240.09,-340.87C245.76,-332.36 252.17,-322.75 258.2,-313.7\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"261.2,-315.51 263.83,-305.25 255.37,-311.63 261.2,-315.51\"/>\n",
       "</g>\n",
       "<!-- 4 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>4</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"98,-208.5 0,-208.5 0,-170.5 98,-170.5 98,-208.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"49\" y=\"-193.3\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [0, 3]</text>\n",
       "<text text-anchor=\"middle\" x=\"49\" y=\"-178.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;4 -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>3&#45;&gt;4</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M130.82,-251.87C115.15,-240.11 96.66,-226.25 81.23,-214.67\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"83.11,-211.71 73.01,-208.51 78.91,-217.31 83.11,-211.71\"/>\n",
       "</g>\n",
       "<!-- 5 -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>5</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"214,-208.5 116,-208.5 116,-170.5 214,-170.5 214,-208.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-193.3\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [1, 0]</text>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-178.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;5 -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>3&#45;&gt;5</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M165,-251.87C165,-241.42 165,-229.3 165,-218.61\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"168.5,-218.51 165,-208.51 161.5,-218.51 168.5,-218.51\"/>\n",
       "</g>\n",
       "<!-- 7 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>7</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"330,-208.5 232,-208.5 232,-170.5 330,-170.5 330,-208.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-193.3\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [0, 1]</text>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-178.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 6&#45;&gt;7 -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>6&#45;&gt;7</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M281,-251.87C281,-241.42 281,-229.3 281,-218.61\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"284.5,-218.51 281,-208.51 277.5,-218.51 284.5,-218.51\"/>\n",
       "</g>\n",
       "<!-- 8 -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>8</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"446,-216 348,-216 348,-163 446,-163 446,-216\"/>\n",
       "<text text-anchor=\"middle\" x=\"397\" y=\"-200.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab2 &lt;= 94.5</text>\n",
       "<text text-anchor=\"middle\" x=\"397\" y=\"-185.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [9, 1]</text>\n",
       "<text text-anchor=\"middle\" x=\"397\" y=\"-170.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 6&#45;&gt;8 -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>6&#45;&gt;8</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M315.18,-251.87C327.46,-242.66 341.46,-232.15 354.37,-222.47\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"356.77,-225.05 362.67,-216.25 352.57,-219.45 356.77,-225.05\"/>\n",
       "</g>\n",
       "<!-- 9 -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>9</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"388,-119.5 290,-119.5 290,-81.5 388,-81.5 388,-119.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"339\" y=\"-104.3\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [8, 0]</text>\n",
       "<text text-anchor=\"middle\" x=\"339\" y=\"-89.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;9 -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>8&#45;&gt;9</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M379.91,-162.87C372.58,-151.87 364.02,-139.03 356.63,-127.95\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"359.46,-125.89 351,-119.51 353.64,-129.77 359.46,-125.89\"/>\n",
       "</g>\n",
       "<!-- 10 -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>10</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"504,-127 406,-127 406,-74 504,-74 504,-127\"/>\n",
       "<text text-anchor=\"middle\" x=\"455\" y=\"-111.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab4 &lt;= 96.5</text>\n",
       "<text text-anchor=\"middle\" x=\"455\" y=\"-96.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [1, 1]</text>\n",
       "<text text-anchor=\"middle\" x=\"455\" y=\"-81.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 8&#45;&gt;10 -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>8&#45;&gt;10</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M414.09,-162.87C419.76,-154.36 426.17,-144.75 432.2,-135.7\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"435.2,-137.51 437.83,-127.25 429.37,-133.63 435.2,-137.51\"/>\n",
       "</g>\n",
       "<!-- 11 -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>11</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"446,-38 348,-38 348,0 446,0 446,-38\"/>\n",
       "<text text-anchor=\"middle\" x=\"397\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [0, 1]</text>\n",
       "<text text-anchor=\"middle\" x=\"397\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;11 -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>10&#45;&gt;11</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M436.33,-73.91C429.9,-65.1 422.7,-55.23 416.23,-46.36\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"419.03,-44.25 410.31,-38.24 413.37,-48.38 419.03,-44.25\"/>\n",
       "</g>\n",
       "<!-- 12 -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>12</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"562,-38 464,-38 464,0 562,0 562,-38\"/>\n",
       "<text text-anchor=\"middle\" x=\"513\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [1, 0]</text>\n",
       "<text text-anchor=\"middle\" x=\"513\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 10&#45;&gt;12 -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>10&#45;&gt;12</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M473.67,-73.91C480.1,-65.1 487.3,-55.23 493.77,-46.36\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"496.63,-48.38 499.69,-38.24 490.97,-44.25 496.63,-48.38\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.files.Source at 0x13b2cea90>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_tree(X.columns, model, counts=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "- With the default setting, the nodes are expanded until all leaves are \"pure\". "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "- The decision tree is creating very specific rules, based on just one example from the data. \n",
    "- Is it possible to control the learning in any way? \n",
    "    - Yes! One way to do it is by controlling the **depth** of the tree, which is the length of the longest path from the tree root to a leaf.    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Decision tree with `max_depth=1`\n",
    "\n",
    "**Decision stump**\n",
    ": A decision tree with only one split (depth=1) is called a **decision stump**. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.48.0 (0)\n",
       " -->\n",
       "<!-- Title: Tree Pages: 1 -->\n",
       "<svg width=\"230pt\" height=\"135pt\"\n",
       " viewBox=\"0.00 0.00 229.50 135.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 131)\">\n",
       "<title>Tree</title>\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-131 225.5,-131 225.5,4 -4,4\"/>\n",
       "<!-- 0 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>0</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"165,-127 53,-127 53,-74 165,-74 165,-127\"/>\n",
       "<text text-anchor=\"middle\" x=\"109\" y=\"-111.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"109\" y=\"-96.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [10, 11]</text>\n",
       "<text text-anchor=\"middle\" x=\"109\" y=\"-81.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>1</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"98,-38 0,-38 0,0 98,0 98,-38\"/>\n",
       "<text text-anchor=\"middle\" x=\"49\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [0, 6]</text>\n",
       "<text text-anchor=\"middle\" x=\"49\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;1 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>0&#45;&gt;1</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M89.69,-73.91C83.04,-65.1 75.59,-55.23 68.9,-46.36\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"71.58,-44.11 62.77,-38.24 66,-48.33 71.58,-44.11\"/>\n",
       "<text text-anchor=\"middle\" x=\"59.3\" y=\"-59.3\" font-family=\"Times,serif\" font-size=\"14.00\">True</text>\n",
       "</g>\n",
       "<!-- 2 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>2</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"221.5,-38 116.5,-38 116.5,0 221.5,0 221.5,-38\"/>\n",
       "<text text-anchor=\"middle\" x=\"169\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [10, 5]</text>\n",
       "<text text-anchor=\"middle\" x=\"169\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;2 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>0&#45;&gt;2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M128.31,-73.91C134.96,-65.1 142.41,-55.23 149.1,-46.36\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"152,-48.33 155.23,-38.24 146.42,-44.11 152,-48.33\"/>\n",
       "<text text-anchor=\"middle\" x=\"158.7\" y=\"-59.3\" font-family=\"Times,serif\" font-size=\"14.00\">False</text>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.files.Source at 0x13b1a9340>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = DecisionTreeClassifier(max_depth=1)\n",
    "model.fit(X, y)\n",
    "display_tree(X.columns, model, counts=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "`max_depth` is a **hyperparameter** of `DecisionTreeClassifier`. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Decision tree with `max_depth=3`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.48.0 (0)\n",
       " -->\n",
       "<!-- Title: Tree Pages: 1 -->\n",
       "<svg width=\"454pt\" height=\"313pt\"\n",
       " viewBox=\"0.00 0.00 454.00 313.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 309)\">\n",
       "<title>Tree</title>\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-309 450,-309 450,4 -4,4\"/>\n",
       "<!-- 0 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>0</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"219,-305 107,-305 107,-252 219,-252 219,-305\"/>\n",
       "<text text-anchor=\"middle\" x=\"163\" y=\"-289.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"163\" y=\"-274.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [10, 11]</text>\n",
       "<text text-anchor=\"middle\" x=\"163\" y=\"-259.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 1 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>1</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"152,-208.5 54,-208.5 54,-170.5 152,-170.5 152,-208.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"103\" y=\"-193.3\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [0, 6]</text>\n",
       "<text text-anchor=\"middle\" x=\"103\" y=\"-178.3\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;1 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>0&#45;&gt;1</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M145.32,-251.87C137.74,-240.87 128.89,-228.03 121.24,-216.95\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"123.98,-214.75 115.42,-208.51 118.21,-218.73 123.98,-214.75\"/>\n",
       "<text text-anchor=\"middle\" x=\"110.9\" y=\"-229.4\" font-family=\"Times,serif\" font-size=\"14.00\">True</text>\n",
       "</g>\n",
       "<!-- 2 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>2</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"275.5,-216 170.5,-216 170.5,-163 275.5,-163 275.5,-216\"/>\n",
       "<text text-anchor=\"middle\" x=\"223\" y=\"-200.8\" font-family=\"Times,serif\" font-size=\"14.00\">quiz1 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"223\" y=\"-185.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [10, 5]</text>\n",
       "<text text-anchor=\"middle\" x=\"223\" y=\"-170.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;2 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>0&#45;&gt;2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M180.68,-251.87C186.55,-243.36 193.17,-233.75 199.41,-224.7\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"202.45,-226.47 205.24,-216.25 196.68,-222.49 202.45,-226.47\"/>\n",
       "<text text-anchor=\"middle\" x=\"209.76\" y=\"-237.14\" font-family=\"Times,serif\" font-size=\"14.00\">False</text>\n",
       "</g>\n",
       "<!-- 3 -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>3</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"214,-127 116,-127 116,-74 214,-74 214,-127\"/>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-111.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab3 &lt;= 94.5</text>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-96.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [1, 3]</text>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-81.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 2&#45;&gt;3 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>2&#45;&gt;3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M205.91,-162.87C200.24,-154.36 193.83,-144.75 187.8,-135.7\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"190.63,-133.63 182.17,-127.25 184.8,-137.51 190.63,-133.63\"/>\n",
       "</g>\n",
       "<!-- 6 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>6</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"330,-127 232,-127 232,-74 330,-74 330,-127\"/>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-111.8\" font-family=\"Times,serif\" font-size=\"14.00\">lab4 &lt;= 83.5</text>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-96.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [9, 2]</text>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-81.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 2&#45;&gt;6 -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>2&#45;&gt;6</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M240.09,-162.87C245.76,-154.36 252.17,-144.75 258.2,-135.7\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"261.2,-137.51 263.83,-127.25 255.37,-133.63 261.2,-137.51\"/>\n",
       "</g>\n",
       "<!-- 4 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>4</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"98,-38 0,-38 0,0 98,0 98,-38\"/>\n",
       "<text text-anchor=\"middle\" x=\"49\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [0, 3]</text>\n",
       "<text text-anchor=\"middle\" x=\"49\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;4 -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>3&#45;&gt;4</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M127.66,-73.91C113.62,-64.29 97.72,-53.39 83.91,-43.92\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"85.84,-41.01 75.61,-38.24 81.88,-46.78 85.84,-41.01\"/>\n",
       "</g>\n",
       "<!-- 5 -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>5</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"214,-38 116,-38 116,0 214,0 214,-38\"/>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [1, 0]</text>\n",
       "<text text-anchor=\"middle\" x=\"165\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;5 -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>3&#45;&gt;5</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M165,-73.91C165,-65.74 165,-56.65 165,-48.3\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"168.5,-48.24 165,-38.24 161.5,-48.24 168.5,-48.24\"/>\n",
       "</g>\n",
       "<!-- 7 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>7</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"330,-38 232,-38 232,0 330,0 330,-38\"/>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [0, 1]</text>\n",
       "<text text-anchor=\"middle\" x=\"281\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = not A+</text>\n",
       "</g>\n",
       "<!-- 6&#45;&gt;7 -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>6&#45;&gt;7</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M281,-73.91C281,-65.74 281,-56.65 281,-48.3\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"284.5,-48.24 281,-38.24 277.5,-48.24 284.5,-48.24\"/>\n",
       "</g>\n",
       "<!-- 8 -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>8</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"446,-38 348,-38 348,0 446,0 446,-38\"/>\n",
       "<text text-anchor=\"middle\" x=\"397\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\">counts = [9, 1]</text>\n",
       "<text text-anchor=\"middle\" x=\"397\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\">class = A+</text>\n",
       "</g>\n",
       "<!-- 6&#45;&gt;8 -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>6&#45;&gt;8</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M318.34,-73.91C332.38,-64.29 348.28,-53.39 362.09,-43.92\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"364.12,-46.78 370.39,-38.24 360.16,-41.01 364.12,-46.78\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.files.Source at 0x13b2cee50>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = DecisionTreeClassifier(\n",
    "    max_depth=3\n",
    ")  # Let's try another value for the hyperparameter\n",
    "model.fit(X, y)\n",
    "display_tree(X.columns, model, counts=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Parameters and hyperparameters: Summary \n",
    "**Parameters**\n",
    ": When you call `fit`, a bunch of values get set, like the features to split on and split thresholds. These are called **parameters**. These are learned by the algorithm from the data during training. We need them during prediction time. \n",
    "\n",
    "**Hyperparameters**\n",
    ": Even before calling `fit` on a specific data set, we can set some \"knobs\" that control the learning. These are called **hyperparameters**. These are specified based on: expert knowledge, heuristics, or systematic/automated optimization (more on this in the coming lectures).    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "```{admonition} Attention\n",
    ":class: important\n",
    "In `sklearn` hyperparameters are set in the constructor. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Above we looked at the `max_depth` hyperparameter. Some other commonly used hyperparameters of decision tree are:\n",
    "\n",
    "- `min_samples_split`\n",
    "- `min_samples_leaf`\n",
    "- `max_leaf_nodes`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "```{seealso}\n",
    "See [here](https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html) for other hyperparameters of a tree.\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Decision boundary \n",
    "\n",
    "What do we do with learned models? So far we have been using them to predict the class of a new instance. Another way to think about them is to ask: what sort of test examples will the model classify as positive, and what sort will it classify as negative? "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Example 1: quiz 2 grade prediction \n",
    "\n",
    "For visualization purposes, let's consider a subset of the data with only two features. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "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>lab4</th>\n",
       "      <th>quiz1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>91</td>\n",
       "      <td>92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>83</td>\n",
       "      <td>91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>80</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>91</td>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>92</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   lab4  quiz1\n",
       "0    91     92\n",
       "1    83     91\n",
       "2    80     80\n",
       "3    91     89\n",
       "4    92     85"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_subset = X[[\"lab4\", \"quiz1\"]]\n",
    "X_subset.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "##### Decision boundary for `max_depth=1` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "depth = 1  # decision stump\n",
    "model = DecisionTreeClassifier(max_depth=depth)\n",
    "model.fit(X_subset, y)\n",
    "plot_tree_decision_boundary_and_tree(\n",
    "    model, X_subset, y, x_label=\"lab4\", y_label=\"quiz1\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "We assume geometric view of the data. Here, the red region corresponds to \"not A+\" class and blue region corresponds to \"A+\" class. And there is a line separating the red region and the blue region which is called the **decision boundary** of the model. Different models have different kinds of decision boundaries. \n",
    "In decision tree models, when we are working with only two features, the decision boundary is made up of horizontal and vertical lines. In the example above, the decision boundary is created by asking one question `lab4 <= 84.5`. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "##### Decision boundary for `max_depth=2` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = DecisionTreeClassifier(max_depth=2)\n",
    "model.fit(X_subset, y)\n",
    "plot_tree_decision_boundary_and_tree(\n",
    "    model, X_subset, y, x_label=\"lab4\", y_label=\"quiz1\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The decision boundary, i.e., the model gets a bit more complicated. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "##### Decision boundary for `max_depth=5` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = DecisionTreeClassifier(max_depth=5)\n",
    "model.fit(X_subset, y)\n",
    "plot_tree_decision_boundary_and_tree(\n",
    "    model, X_subset, y, x_label=\"lab4\", y_label=\"quiz1\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The decision boundary, i.e., the model gets even more complicated with `max_depth=5`. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Example 2: Predicting country using the longitude and latitude \n",
    "\n",
    "Imagine that you are given longitude and latitude of some border cities of USA and Canada along with which country they belong to. Using this training data, you are supposed to come up with a classification model to predict whether a given longitude and latitude combination is in the USA or Canada. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>country</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-130.0437</td>\n",
       "      <td>55.9773</td>\n",
       "      <td>USA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-134.4197</td>\n",
       "      <td>58.3019</td>\n",
       "      <td>USA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-123.0780</td>\n",
       "      <td>48.9854</td>\n",
       "      <td>USA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-122.7436</td>\n",
       "      <td>48.9881</td>\n",
       "      <td>USA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-122.2691</td>\n",
       "      <td>48.9951</td>\n",
       "      <td>USA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>204</th>\n",
       "      <td>-72.7218</td>\n",
       "      <td>45.3990</td>\n",
       "      <td>Canada</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205</th>\n",
       "      <td>-66.6458</td>\n",
       "      <td>45.9664</td>\n",
       "      <td>Canada</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>206</th>\n",
       "      <td>-79.2506</td>\n",
       "      <td>42.9931</td>\n",
       "      <td>Canada</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>207</th>\n",
       "      <td>-72.9406</td>\n",
       "      <td>45.6275</td>\n",
       "      <td>Canada</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>208</th>\n",
       "      <td>-79.4608</td>\n",
       "      <td>46.3092</td>\n",
       "      <td>Canada</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>209 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     longitude  latitude country\n",
       "0    -130.0437   55.9773     USA\n",
       "1    -134.4197   58.3019     USA\n",
       "2    -123.0780   48.9854     USA\n",
       "3    -122.7436   48.9881     USA\n",
       "4    -122.2691   48.9951     USA\n",
       "..         ...       ...     ...\n",
       "204   -72.7218   45.3990  Canada\n",
       "205   -66.6458   45.9664  Canada\n",
       "206   -79.2506   42.9931  Canada\n",
       "207   -72.9406   45.6275  Canada\n",
       "208   -79.4608   46.3092  Canada\n",
       "\n",
       "[209 rows x 3 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### US Canada cities data\n",
    "df = pd.read_csv(\"data/canada_usa_cities.csv\")\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df[[\"longitude\", \"latitude\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df[\"country\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEQCAYAAABBQVgLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAABEd0lEQVR4nO2dd5xU5dX4v2d2ZwtbKIqwr11ssJRFQE0W0SAJCBqNEgtIEcUkKEGwxCBY0B++IRGRWBKVJoKI5jWxgUixUKQJKCUia4ehSN1dtu/z++POrLOzszOzu9PnfD+f+Qxz7zP3nvvscM89z2lijEFRFEVRfGGLtACKoihK9KPKQlEURfGLKgtFURTFL6osFEVRFL+oslAURVH8khxpAULBiSeeaM4444xIi6EoihJTbNy48UdjTGtv++JSWZxxxhls2LAh0mIoiqLEFCLybX37dBlKURRF8YsqC0VRFMUvqiwURVEUv6iy8MDhcNDv8kvZu3dvpEVRFEWJGlRZeDBl8qOsW7OKKZMfjbQoiqIoUYMqCzccDgdz5sxm2ZA05syZpdaFoiiKE1UWbkyZ/CjDOifRNSeJoZ2S1LpQFEVxosrCicuquO8i6/N9F6HWhaIoihNVFk5cVkVOljUlOVk2tS4URVGcqLKgrlXhQq0LRVEUC1UW1LUqXKh1oSiKYpHwyqI+q8KFWheKoiiqLOq1KlyodaEoihKnVWcbwvq1a1i1vohpq3yPyy9fHR6B4oiCggKmTpvOvPnzOXbkENktWjF40CDG3fVH2rVrF2nxFEVpAGKMibQMQad79+5GS5RHlkWLFnH9TYNJ7fgrUnP7kNz8JCqP7qds21LKti5h4SvzuOKKKyItpqIobojIRmNMd2/7Et6yUIJPQUEB1980mMyrxpN6cvua7faWOdh7DsF+Zneuv2kwmzeuVwtDUWKEhPdZKMFn6rTplkXhpijcST25Pam5v+TJp/4eZskURWksqiyCjFathXnz55Oa28fnmNSOv+TlefPDJJGiKE1FlUWQKCgo4I7RYzjv7LP4ZNVHnNvuTO4YPYaCgoJIixZ2jh05RHLzk3yOSc5uTeGRQ2GSSFGUpqLKIggsWrSIvG49eGXtN1RWVbFiWAZV1dW8svYb8rr1YNGiRZEWMaxkt2hF5dH9PsdUHjtAVotWYZJIUZSmosqiibg7c22mklvy7HTNSWJ4l2RsporMq8Zz/U2DE8rCGDxoEGXblvocU7b1fW4ePChMEimK0lRUWTQRlzM3qXkbSrYuZUK+NaUT8m2UbFtKcvO2CefMHXfXHynbuoSy3Tu87i/bvYOybe8zdszoMEumKEpjUWXRRFzO3NK1CxjeJblW1drhnZMoWbsg4Zy57dq1Y+Er8yh6azJFH79ExWEHpqqSisMOij5+iaK3JrPwlXkaNqsoMYQqiyZy7MghJCm5llXhwmVdSFJywjlzr7jiCjZvXM+N3XIofn08P0y9juLXx3Njtxw2b1yvCXmKEmNoBncTaXFCa2wnd+SmZut4pl/dHMc7FlfyyvELqd69jSMHfTt9FUVRIomvDG61LJrINVdfTekXH9exKlxMyLdR9sXH/Oaaq8MsmaIoSvBQZdFEbFWlDPdTtXZY5yRsVSVhlkxRFCV4aG2oJrJzxzZWbSjnuQ3lPsfly7YwSaQoihJ81LJoIivXbcIYw65du7hj9Biat2qNzZZE81atuWP0GHbt2oUxhpXrNkVaVEVRlEajDm5FURQFUAd33OGqQ9XihNbYkpJocULrhK1DpShKeFBlEWO46lC9umkfGQMf59S73yBj4OO8umlfQtahUhQlPKiyiCHc61Bl9hyCvWUOYkvC3jKHzJ5DIlKHavny5XTK64YtJQ0RG7aUNDrldWP58uVhk0FRlNCjyiKGiLamQpMmTaJPv/58l3YmObc8zWn3/pucW57mu7Qz6dOvP5MmTQqLHIqihB51cMcQLU5oTcbAx7G3zKl3TMVhB8Wvjw95tvjy5cvp068/bW54zKvyKtu9g32vTmDp4nfp3bt3SGVRFCU4qIM7ToimpkJjxt1L1gVX+rRysroO4K677w25LIqihB5VFjFENDUV2rZ9G5ld+vkck5l3BVu3aTKiosQDqixiiGhqKmQqygOyckyF78x2RVFiA1UWMUQ0NRUSe0pAVo7YU0Iui6IooUeVRQwRTU2FcjvkUrRlsc8xRZsX0TE3N+SyKIoSelRZxBjR0lToT/eMpXDj2z6tnMJN7zDtib+GRR5FUUKLhs4qDWbRokVcf9NgyMnl+DdbyOzSl8wufUnObk3lsQMUbV5E4aZ3eHjCeB588MEmn6+goICp06Yzb/58jh05RHaLVgweNIhxd/1RW7MqShDxFTqrykJpEAUFBeR160HmVeNJPbk9FYcdFH76NsU7PqT6+DFsaZlQVc6cGS9w8803N/l8LsWU2vFXpOb2Ibn5SVQe3U/ZtqWUbV3CwlfmaYtWRQkSqiyUoHHH6DG8umkfmT2H1Dum6OOXuLFbDk9Pn9akc3kqJk/Kdu+g6K3JbN64Xi0MRQkCmpSnBI158+eTmtvH55jUjr/k5Xnzm3yuaCtvoiiJjCoLpUGEM4s8nIpJURTfqLJQGkQ4s8ijqbyJoiQ6qiyUBhHOLPJoKm+iKImOKgulQYy7648Ub1nkM7+i+LPFQckij6byJoqS6KiyUBpOdTX7X3+Ewx/OrpVFfvjD2ex//RGorg7KaaKpvImiJDqqLOKUUPXpnjptOhldB9B26JOYygr2zruX7564lr3z7sVUVtB26JNk5PUPSoRSNJU3UZRER/Ms4hBfiWwln73HLy7rxcpVqxuVDR2JBkwFBQU8+dTfeXnefAqPHCKrRStuHjyIsWNG065dO83wVpQgoUl5CUQgiWz7XnuI1r95gLRTOzY4G9qWlMSpd7+B2JLqHWOqKvlh6nVUVVU2+Xr8oRneihI8NCkvgQgkkS2ra39Kdq1DbEnYW+aQ2XMImVeN5/qbBvtdpmpshFIolsUKCgq4/qbBZF41nsyeQ7C3zGnUNSmK4p+wKgsRuUxEjJfXEY9xLUXkRRH5UUSKRWSpiHQKp6yxSiCJbJmd+1L0+ftUHHbUbAs0G7oxEUqLFi0ir1sPXt20j4yBj3Pq3W+QMfBxXt20j7xuPVi0aFEAV1aXSGV4h8ofpCjRTFiXoUTkMmAF8EdgvduuSmPMBucYAT4CzgTuBQ4DfwZygTxjzA/+zpPIy1CBLhN998S12NIyOXHAONLbWVZnIL6GhtZrCmV9p0j4T3TZS4lnfC1DJYdbGCc7jDGf1LPv10BPoLcxZgWAiKwBvgbuw1I0Sj24lol83UArjx3A1iybk37zAPv/9ShthzyBvWVOQNnQrgil628aTEXuL0nt+EuSs1tT+sM2jn00l7L9X0NVBd0uvJjBgwZx9NhRv0//Fc6n/4YWHjx25BDNw5jh7b7s5X499pY52HsOwX5md66/abAWNlTikmj0Wfwa2ONSFADGmKPAW8DVEZMqRghkmajos/fIaH8pqSe3J7PLryj89G0g8GxozwZM3z9xLQf+NYm003LJGfE0p97z0zLT/PkLQlbfKdwZ3lrYUElkIqUs5olIlYgcFJH5InKa275cYKuX72wDThORzPCIGJ0sX76cM9qdiySnIvY0bPZURGwkpaZz9bXXcd1vrvabyFa0ZQlZF1wJWP6L4h0fWvsakA3drl07np4+jY3r1pCR3Zw2NzxG80tvqeNkNlXlIavvFO4Mby1sqCQy4VYWR4EngNuA3sCjQB9gjYi47iitsPwUnrjuJi29HVhEbheRDSKy4cCBA8GVOsK4HKrpmdlcfnkfvv36KxBBbEmYqgpszbJJP7cny3YV0qdvf07Oacux//w/Di1/sW6G9b8e5cQB42qWqZKzW1N9/Fijs6H9PW3b0puH7Ok/3BneWthQSWTCqiyMMZuMMfcYY94yxnxojJkG9APa8JMvQgBvXnfxc+znjTHdjTHdW7duHVS5I4krkmjBpw5aDZ5Ky753IEnJZHbsTeaJbWl+6rm06jcG2b2Z4ztX0+KSm9n51beUl5VS8tkS9r58T+0M6yFP1Di0wem/SElvdDa0r6ftyqJDNEtN4tiG//g8RmOf/gPJ8H5q6t+YOm062S1PIN0uZLc8odGRS+7LXpVFhzi68H6qimo/12hhQyVeibjPwhjzKbAT6OHcdAjLuvDEZVF4szriEneHatYlwwA4smIGbW58jNRkgR+/gv07KV89h6SSQzQ/qwtH17xKRodepLS7EASyz+/J6fe9yal3vkyry0fWcXwXbXqX3PbnsXnj+kZF8fh62i5du4CkksNUbAvd07+n/+SHqddR/Pp4buyWw1NT/8aYcffw6qZ9JJ/amdRkG8mndm50yO4lPfMp2vwulUWHODbvLoxjOyVrF9S+Hqfi0/BaJd6IuLJw4m5NbMPyW3jSAfjOGFMUNqkijOcSz9HVr5LZpR9Jzdtw/PP3SRJYODCNih+/ZcXQdMq/2Uj6+ZdQVVJI6Xef0axzX4q3f+DzRl298wPeeH1ho6N36nMyVxYdomTrUlYMTSeJavYueIDDH8wKSX0nl//kyMH9VFVVcuTgfsaOGc2YcfeQedV40vIGUPblalYMTads1xrS865scMJeQUEBy5evoOiz9yl87ykqiw6xYmg6JduW1lgXZbt3UPL5Ei7I61yTV5LW/080P709yZfcxsw3lnD2eR2abOEoSiSIuLIQke7AucBa56Y3gZNF5FK3MdnAVc59CYPnEs/xnavJ6tqf0rULaN/KMDwvhUVfVjGiq52uOUkM75xESlUJpd9sovr4MTLz+pOSYg9pIb76nMylaxcwvEsyXXOSGNY5iYyWJ2GqKtk75y6+f+Lamqf/xlo0/nBXtO6yDO+cRMnaBQ2OXJo6bTrpXa6g5eW/o/KbjYzsZq+5tuKPZnD4w9nse+0hLrqwW42Syuw5hMr/rsA4tlO0eCopp3aiZYefN8rCUUulNjof4SfcGdzzROQxEblWRHqLyN3AYmA34Ppf+yawBnhZRG4Ukb7ObQJMCae8kcZziceUlyBJyRz//H2+OlzJsC7JzNlSzgOXpAIwId9G2ZerMWXHsTXLJjm7NSVFhfUu0wTjRu3NyeyyKibkWz+viZfYMYX7aX7RQNoOm0Z2yxM4cnA/T0+fFrJ8BJei9ZRlQr6txhpoSOSS63jV32/CZoP78605n5hvo/S/H1JVUkjr3zzAqjXrapRUZdEhjn/+PiuGNiNJDBnn92qUhROqDPhYZebMmbTv1IV/vDCDo4cPImlZVP5PHvNXFyTkfISLcFsWW7HyKGYB7wF3Af8HXGSM+RHAGFMNXAm8DzwLvAFUAb8wxnwfZnkjiucSjyTbOb5yjmVVdElhzpZKhnVJISfL+jPmZNkY1iWJ1GQbGe0vrXG2elumCdaN2t3J7Iq+Kv1kPsO6JNWSa3iXZErWLghbtJBL0bqsilqyOK2Lhshy7MghJCmZ8v9+wK1d7bWPd0E6afZk0k7JpaK8rMYaLF27gOGdbZZF08VO8eK/NdjC0fpXtZk5cya3/X4U6Z2voO2waZx2z79pe/PfSMpsybEdH2O/4NqEmo9wEu5oqMeNMZ2NMc2NMXZjzKnGmNuNMQ6PcYeMMSOMMa2MMc2MMZcbY7aEU9ZowHOJx9iSKPnvR7WsivvyU2p9Z2J+EjaqaXb+JRRtfjcsXeRcTuaybUtxzB1H8ZbFTMyvXW7E9URftndXWKKFslu0osyxs5ZV0RRZslu0omjF8wjV/MlpVbiY6HY8U1lhlQBxWjQTL7EDMLILVPz4bYMtHE0E/ImCggJ+f+cfaXPDY7S8bHgtxdny0mGcdN1Ejn7yGsntfpYQ8xFuIu6zUOrHc4kno/kJdDgBr1aFi5wsG8O7plK28f+o2L48bF3k2rVrx20jRpDVqg0jujXzLlfnJEpWPBcWBTZ40CBKVvyzllXRFFmuufpqKgo+qWVVuB9vmPN4Kc0yqDy6v45FM2dLJSM8LZIALBxNBPyJqdOmk97Jt+LM7PIrKioqEmI+wo0qiyjGtcSzb8EEDi55jqqje/nqSHUdq8JRWE2/l4vZW2S1M514iZ2yXat54R/PhLVG0c2DbqRiX0GdJ3kXE/JtVOz/ipsH3RhzspQWHsYmpo5V4WKi83hXDuhPyaa3alk0jsLqWr4ldxn8WTiaCPgT8+bPJzOvv88xmZ37UlKwPiHmI9yosohyrrjiCjIymmF2b6FDq2qvVsWUVeWs213FlFXlgPXUOrJHBhvXrg6rrAvnz2Vkj/Q6T94ucrJsjOyezsL5c2NKFofDwbtv/8erVeF+vNu6pdOimZ3yre8xrHNSrb9PvVagHwsn3PWvoplAFWd1yTFMUjJis5GVnsItI26Lex9GOKLDIlV1VmkAQ4cM4eWX5/Lf4mo+3VtOMzscr4Bpay3lkJYMq0dkkD+zmCc/Ka/5Xn5FeJXF+rVrWLX+ONPX+B6XXxl6uYIpy5TJj3JCmuGptRU8tbbC59geJZtJFsPEnpbPxmVVbBvlvaTZhHwbs/5ev4UzeNAgXt20FHvPIfWeM5j1r6KZQCsqS0o6zc79OSmmjOpdHzN/7ixemTsDAarszbj+t7/lkQcnxk1lYPey+RkDH6e5s2z+q5uW8lK3HkErm69tVWMAXz0hipY9y/W2FTx7hZ0x71eR1G0YU6c/EyFJ45OeF3Zl1frNfsfl98ijx0U/g01zebKPZUWMXVwKwJP90ur93pgllSR1H+717xbKfiCxxh2jx7Dg071kXTK03jGHP5hF4aZFNP/5TZSunMHqERn0nFVMkrNYUJWB49V2xJbE5EkPcf/994dJ+tAQ7N+H9uCOA2qeHtx6SJTt/ZJjC++n4E5rucVRWE3HF6vY9sVXtG3bNtIiJySeisVlBfojv0ceK9dt8rrP29++8tgByra+T9m29xOm4VJA/eUXTCD75zdQ+skrjOgME3ulcvbfi0ixwZXnJvPWzkrkrJ9TnZ1D4ca3GXPnH3jyyScjcDXB4Y7RY3h10z4yfVieRR+/xI3dcgLqF6PKIk4oKCjgyaf+zsvz5lN45BAZaXaGdoKn+/0UPjt2aTVywVC1LuIMz799VotW3Dx4EGPHjI57i8KdmTNn8rs7RpPe6VdkdR1QozgLN71D0Zb3aNl7JGV7vqB6xxIKRmcwZVU5K7+rpGvbJF7fUcGyoRn0nF1Kq5GzqTy6l30LJrD0vXfp3bt3pC+tUQS7W6QqizjE4XCQe147to1MquU4VetCCSYFBQVMnTadefPnc+zIIbJbtGLwoEGMu+uPYVdSixYt4pqB12Nv9zMASgrWU11yDElJx1RXQWUFzX9+I6UbXmNEJ5jYK4X2zxQhAr/tYCfDLjzZL41R75TwakUvsvqN5fCKmZxe/g2fbdoY1msJFoG2Uf5h6nVUVVX6PZ4vZaHRUDHKlMmP1oq4cZGTZWNopySmTH40QpIp0czy5cvplNcNsaeSbhfEnkqnvG4sX768zthoKjNSUFDAddffRKtrH+LE/mM4sf8YTh39Mqff9yan3fUqbW94FFtqOkfXLESqypnYK4Upq8o5p5WN37a38/r2ippQ84m9Uind8QFVRYfJzLuCrdu2he06gk04o+VUWcQgDoeDOXNmc99F3vffdxHMmTOLvXv3hlcwJaqZNGkSffr157u0M2nZ3ipo2LL9z/ku7Uz69OvPpEmTasb6KzOSdtntXH3db8lueUJIC/m5QkI7dM4juX1v3wl5ef1Iz2jGLXlW1vyszeXsOlwNQt2yOJ1tHF85m+Ts1piKcq/HjAXC2S1SlUUMUp9V4UKtC8WT5cuX8/Bjk2lzw2Nkdb+6VkHD7O7X0OaGx3j4sck1FoavMiMlBRs4tOQ50jv3I+v6v4TM4nC3bCqNjayuvhPy0s++CFNylIm9Uuu1KlwJrL/rZqd0xweU7d2F2GuXzAlGzkK4quKGs1uk+ixikIaEctYXYaMkFp3yulkWxWW3ULTsWW5I+oBn+iVzx+JKXq36BZmX/6HW+n19jtOKww72zr2bk66bGNJQXs/Ip2//chVZ3a6keMfHVJccw5aeTUaHS8m64MoaGYuWPs31soyJvVK9+irACmWes6Wc4XkplFYaZmyppryimozs5gwbOpRuXbswZtw9lqLM7WPV+Tq6n7JtSynbuiSgyDP3vIfGHqMhBDNaTh3cipLg2FLSyLnlacSeyuEZt7FrVEpNuPXZz1XQasSLVFeU4ph1J9XlpfU6Tg8tewFJTqHlpcPqPVdDQjXdcXemHz18EFtKM9LP+RmVhT9SvnsHmV37k9W1f83Nt+izJRRteY8TB4wjvV13js75A0f2fk9KEnRsbaNbjhUBtW1UZs215j5bxLKhGfSZe5zlQ9P5+axSWt46E1NVQenWJRxe/yYtet5M9oW/qSNfIIowUnkxwYqWUwe3oiQ4pqI8oJLtrvX7+hynxds/JLPzr3yeqzGFDd17zWcMfJzT7vk3J/5mPBU/fkv57u2YqnKKt62g8NO3a7K4XZVmf3xnKhWHHVSfcREpOeeSldGM/x6sZs5nFdzY0V6n7ErXnCSGdrZbZXM6J1O6/jXsLXPIumQYbW54jKOfvEbFYUcdGQOp8BupKsGhbEPgosHKQkROFpGpIrJBRL4SkY7O7XeJSD0uVyUUOBwO+l1+qTqyFb+IPSWgku2u9fv6HKfVJceaVNjQ21r+zUOHMfCGQTW95u0tcyj9ehM//ucvpJ2RR86IZ5x9K/6KJKewd+7dlBRYKweuSrNHPp5L0ebFlB/4hoVvvEVxuaHHBXk8t6ECeeQY8sgx/rHRKr7pKKxmy74qnltfznMbyqne+9N6v+t4hZ++7VV+f4ownqsEN0hZiEgu8DkwBNgDnA64vEOnA2OCKp3ikymTH2XdmlXqyFb8ktshl+Jlz/ks2V687Fk65uYC9TtObenZjQ7VXLRoER27dGXGvxZTWFKOMVBYUs6Cf79L0vm/qHkarzjs4Md3pnLSdRNpeekwr30rfnz7iZqn/8zOfSn5ci3VFaU8MvGBmgS7les2YYxBbDaan9ONW/Kspbcpq8rZvLeK3DbJnNB9ANmDp9eSM7NzX4p3fOj12vxV+I3nKsENtSyeAHYAZwLXYrU6dbEauDhIcil+cIXPLhuSpmGyil8efOB+Kg985bNke9WBr3jwgT8DtTsguvdvTz+rO0VbFvs8l7dQzYKCAq6+9jrKqwypp3eh7ZAnOO2eN2g75Amqy4/XinQq/PRtMrv09bmUk9GpD4dXzASsm6+pKmfZkvd48MEH64xPz8ii7OuNTOyVUlPYccHAdL74sZLiz5dQVXS41vjk7NZUHz/m9dz+chbiuUpwQ5VFT+B/jTFFgKdnfB+gKcNhwhU+2zUnScNkY4xILB+u/mgFt3XzXbL91m7prP5oRc02VwdE9/7t8v2nlHy2uMGhmuPuvZfKauGkgQ/VsRZMeWmtp3FffpHKokMcnn83snsLJV9tpOKww6o0m5zC6aef7vU7zdPt3NI1tcaqGNYlhUVfVpFsgxNSKylZu6D2OY4dwNYs2/v1+clZCGfeQ7hpqLKo9rHvRKCkCbIoAeKZlKdJeLFFJJYP169dw9NrS2rW7729nl5bwrpPapds93ScFh45yH/+9Vodi6PisIOij1+i6K3JLHxlXh3H6jvvLibrgv5erQVJy+TogntrnvB9+UVK1y7A7PsCOfQ1KZRT+OnbFH32Himtz/DqNHY4HBw5dJCJl9hrrApX87Bn+6dxpMRQuGVxLeuiaMtiMtpfWudYgeQshDPvIdw0VFmsA26pZ9/1wKqmiaMEgmdSnibhxQ6RWj50rd/7ewWSl+PN4ih+fTw3dsth88b1XmP6q6uqyOzct9a2yqJDHF14P82ymsOBL2ue8Ovzi1QWHeL45++TJLBiaDOSbFC0dRlFW5bQvNdQXpr7cp3vTJn8KMPzkmtZFa7mYVPXlJNsgzRbVc25y3bvoHDj25jKslqK8NhHszmw8EGKigo5++xzsKWkeS2TUt/ynT9lGgs0KM9CRC4FlgIrgPnADODPQC5wI9DLGLM2BHI2iHjOs9ACgrHN2NGjavpdJFKFYBEbp93771p5G0XLnqX880VUGmHV8HR6vlROq1tncnTt6yA2WvW+tdYxipY9y+m7l9DrNJuzIGApMzaV0+I3D5N2Rh7fP3Et1dVVtb7jnsCalgxrRjSj90vHeXVgOlcvKGGVs2kYGa2wt+9N0ZYltLh0OBU/fkvx9g+pPn6UtIxMSsvKycy9jOyLBv6U57FlMYWfvs3DE8bX8ZXEapXgoCblicgAYBrgfsXfAHcYY8JXWcwH8aws3G82dfYl0M0nFvFU9LGs4L1Vo71yQH8Q4e2336lTofac8zuQM+KZmmzryqJDHJ5xG9eeU01WqvDsgHRGvVPC3D2n0azXCA78axJtbnisZtmqsugQh168lVSpYLtbkl27vxdxwu1zayUUenLH6DG8+Nx0bu1qJ9XZBWnld5VcfEoSf++fzp3vljB7cwW2LlfXygivOOzg6Kv3cfz48VqyuFO2ewf7Xp3A0sWxW+bcnZBkcIvI2cBJwEFjzBdNkC/oxKuyqM+qqNkfwzefRMCboo9FBe+tnEXR58s4vOJFsrr0JTOvf60SFyWfvYcxVZSWV2HKS0hPScLWIofrTj7AWztKa2VYt3v6OKVJmZiyYiQ5lawL+pPZuS9lG//F6Xvep9dpSbW6Do56p5SF/JKKpHQKN/yH6sq6RQHXrFlD714/r2NVFPzR7bzTi2h2xf1kduhZ872ij1+CXR9TcebPaXlZfavvxHyZc3e03Eec4MuqqBkTgzefRCBelg9d5SzSLrudst3/tZZqSo4iSXbSz76YFr2G1KknVbZ7B/sWPkTrax+g6suVlG9dTGU13NwpuVbdJsCyLnafQrNLb+PAG4+TftYFlHyzmZTyY6TbqbEqXDgKqzn7mTJKSCEjLZXCIwfryDx29Cgq1s3ELta9zt2qcHHnuyXM/hxOHPNGjcxFb02mqLiY/3GziLxRcdhRr1UTa/hSFskBfLn+hrdeMMa81JDxSuCsX7uGVeuLmOYnjCC/fLXvAUrY8d1/xNofCwp+6rTpyMmdOLTkOTK79KXtzX+tVatp79y7a2o1uUg9uT1ZF/SnePtHmJ0ruPbcJGwCr2+36ja5M7FXKrOnf83+1x6GqgpKvlxLRuv/4VyK6XWa9/kb1gle+jqLYYNu8Cqz9f+mjLRkePPGdP65sZw3b2pWa8wDl6Qyc1MRhVuXU7H/a8zOD1j4yjz69x8QUJJdLJc5DxS/loWIeIbLur4gXrZhjKm/ZVOYiFfLQolN4mn5MKtFK46XVXLSwAfrXcPf/69HaTvkiVpP4xWHHRyccwc3dkjizR0lDPSoBuvOHYsrmXfwPI7/sJMkm2DKirxaFS4s30Uxyz9azcUXe88LdlkXa7+vqGNVuLjz3RJmba7g1HM68s6b/6Zdu3Y1BRjVsggsdPZMt9clwA/AP4HLgPbO9+eB77GS9hRFcSMe+o+4MrCLio+T0amP7yZEXmoriS0JqSonRSoZ6NFjwpMJ+TbKv93MrBf/ye9HDKFTm2SGuzUv8iQny8aIriksnD+3XvnXr13DM+vK2HagmvGXpHod88AlqRgDT0z535qIpdwOuX4z1os2L6opkxLP+F2GMsZ86/q3iDwFLDDG/MltyBfARyLyF+A+oG5tX0VJYGJ9+dDV+zo195fYku1+mxBldu7L3nn30urykTXbjq96ies7WEpiYAd7rc51nuRk2RjZoxkb165m/do1bD9QyYY9MG2t76WefFP//K1ct4keXTtykX2nH6Vj55EJ9zNgwAAAnpr6V/r060+zcy6u15Iq3PQO0xa/61O2eMCvsvDgcuDpeva9D/yhaeIoSvwRyw2o3HtfV1eWU7j+317X8CuLDlH87hQy+/+pTm2lyqJDlH7xMakdbQzrksL6PVWs+r7K/82/YnVQ5+67r79iw9EKnllf4XPcSc1/6mbXu3dvHp4wnocfm0BW1wFk5l1R01yoaPMiCje9w8MTxsdF2Kw/GqosyoDuWIl5nvQA4t/LoyhRhsPh4Jabb2T2vFeb5PNYvnw5Y8bdy9ZtW0mjnFJSaNG8BUlnXUh1aTEH3pyCpKbX9JOoOOyg8NO3Kd7+IcllR0hPhsOvjyfjF7+vVVupePlzDO+cxGf7LCXhj1B1eNx35Hijvvfggw/Ss2dP7rr7XrbOuhNTUY7YU+iYm8u0OMmvCISGZnA/C4wAJgKvYRUPbINV6mMSMMMYc0cI5GwQ6uBWEomxo0cxZ8bzDL/td42OqJo0aRIPPzaZrAuuJOn4Qaq//AjbOb2oTGtJ8efvY6qrQGzWU/XR/ZjyEsSeQlbXK0k/92KOLbyfVcPTyJ9VTIlJI/2MLrTsfRtH1/6Lqm2L+eqPjXdOK+EjmJ3y7sZSEo8DBUCR830ysNC5X1GUMBGMWlPLly/n4ccm0+aGx8jqfjVlX65mxdB0ynatofmF13LSwIcw1dVQVUl6u+60vvYBbOlZtLnhMVr+4hYqPl/E8M42uuYkMbyLnczsLEq/2cyeGaOo2P4+t+TZffsnuqf7dE4r0UGDlqGMMSXAEBF5FLgIyAEcwFpjzM4QyKcoig9+KlVvY2gnaVS+xphx95J1wZWkntyeomXPMrxLsnXj72x4de0CUi+4BrEl0eb6R0g9uT2Hlr1Q02+isugQJTs+ZOKdVijqxF6pzN58gPTzLqf0280klx/iuQ1VPLfBt58gvzI6nfvKTzSqB7cxZqcxZq4xZorzXRWFooSZYJWq37ZtK5ld+nH8602UbnmXa8+1lqZd7VaPrn29Vnlx934TJStnM7yzrXZP7zw7SXu3Wkl11fagVbutbw60tXB4aGhb1dP8vUIlqKIotQlGqfpJkyZhKitIbn4SJYv+QrIN/rysrOZ4wzsnUfnfFbXKi7v6TdRYFb1q5y1M7JVK5dF9VJccw1T6tiiairYWDh8NtSy+Ab7281IUJcR4WhUuGmJduHwVtvRMirauoKqkiBXDMth2oJrP9lUClnVhqypHkuw133P1m/C0Kly4rIt0uw1JthMqtLVweGmoshjh5XUv8AHwHTCy3m8qihI0fNeaCsy6cPkqMnJ7U/bBs4zoaqdrThIj8uzc9mZpzfGG56VwfOXsmu9ldLiUYxv+49WqcDGxVyqY6jq9l4OJthYOL0GrOisic4FvjTETgnLAJqChs0o8E6xaU666R6Xfb6Pk/Wm1Snaf/fci1tzajM5tkp2lw0s4YeRskjJbWnWeZo9iRGd4dkDdGksuRr1Twuu7Utl/uDAo113rGuOoN0g0EczQWV+8jGVpKIoSQoJVa8pUlJPc/CTKV77IiK72Wr6POtZFZxtFy58DwN4yh9QUO89tqPDZ0/u5DRWUlhSHZHlIWwuHn4ZmcPviJKBuCUlFUYJKsGpNiT2F4wUbqCop4oFLapcKH39JKu2mFyGP/FS2IyNlDYeXzyCza39szdtAkX8X5QnpjQvn9YXLV7FtZO0C1/ddBB1fnMV94yeqdRECGhoN1cvLq4+I3AX8Dfg4JFIqSgwQrjDOles21YSc7tmzh5ZZ6ewZl4l5KJs94zJplZ2Ow+HwG46a2yGXkvefqmVVuHAV1cvIyqZT3gUYY9iy/UuGXHwaxa+P55jjO7LT/D9rfnOkmnWfBDeHIhj+GqXhNHQZ6gNghcdrCTAV2I4WElQSmEiEcTZlOWb0qN9RXXKMB3yU7K4uOcboUb8HoF27djw9fRpHDu6nqqqSoyUVIc2h8EZ9UWAuGptrovinobWhLoM6AQ6lWI7tqPnrqINbCTcuh+uym2z0WVAdFkdrU1u1ukp2P+2lEZCLO98tYV3Feazb9HlQZW8s2lo4tGgPbkUJMe43sXDdrHzdOAORoU2LZuw/WuL3PCc1T290xdbG4KuKbs8Lu7Jq/Wa/xwhV5dp4J2jKQkSqgJ8ZY9Z52dcNWKdtVZVEIxJhnPHUqtWTYFTRVRpHMENnxce+JOouUSlK3BOJMM54aNXqDc3Kjl4CCp0VERs/KQqb87M76cAVwI9BlE1Rop5IhXHGeqvW+ghGFV0lNPi1LETkIaACqwueAVY5P7u/jgEPYvW6aBAislhEjIg85rG9pYi8KCI/ikixiCwVkU4NPb6ihJJIhXG6h8+GKxIp1ASriq4SGgJZhvoAqwveo1jWxUznZ/fXA8BvgDENObmI3AR08bJdgDeBfsBo4DrADqwQkVMacg5FCRXxFsYZiXLf7ufUrOzoxq+yMMZ8aIx5xBjzMPAI8LDzs/vrf40xbxpjqgM9sYi0AJ4ExnnZ/WugJzDEGPOKMWaxc5sNuC/QcyiJQaR6GkS736Ch8xKpPJF1a1bx0IT7m1xFVwkxgZiyoXgBzwPLnP82wGNu+2YAu718Zw5WTofPY3fr1s0oicNdd/7BtExPMmNHjwrrefN75Bnnb9fnK79HXljlctGQedmzZ49pmZVuPr09w7TKTjcOhyPk8rmfMyst2Yzs0cyYh7LrvO7Kzwz73zZRATaYeu6rfkNnRWQm8Kgx5mvnv/3oHnOrPwUlIj2BpUAXY8wXImKA/2ecFWtF5BPgqDGmr8f37gP+AmQZY4rqO76GziYOkUiGiwUaOi+RzhMZ9U4JVdXwz6vqJgjGchhwrNHU0NlfANnOf/d2fvb18ieMHfgn8DdjzBf1DGsFHPay/ZDzvaWX494uIhtEZMOBAwf8iaHECdrTwDsNmZdwOZbdl8U8zzmxVyqv76hkb1HdlexIL+cpFmHP4BaRCVilzHONMSXObZ6WxZdY5tBNHt8dibV8dZox5vv6zqGWRWIQCz0NfGUjh/KcDZkXb5ngobAu3JPtjDF1zjnqnRJmbKqgvMr79zUrO/QELSnPWWU2s559GSLSy8/3T8OKnJoIpIpIC6ejG7fPSVgWRCsvh3BZFN6sDiXBiIXomWgvLhiM9qz+cDgc/KLnxcyZPYtlQ9KYPXsms2bNqnPOib1SycywKuZ6WzNXRRFZGprBvQLoUM++8537fXEWVs+Ll7Fu+K4XwD3Of3cCtgG5Xr7fAfjOl79CSQw8b3KOwmr6vVzM8I7VURM9E4ls5Ibe/MORJzJl8qNs3LCOwbnQNSeJwR3g3BZVWmI8xghmuY9UoB4DsobN1O/neNn5711YORYni8ilNScWyQaucu5T6iFSYaThxvMmN2VVOet2VzFnS2XU3HAi4U9pyM0/HHkiDoeDWbNmkYRhfL5VMGJ8fjK7DlZ49U9oqGz0EkgG9xki0ltEejs3dXd9dnsNAO4GvvN1LGPMEWPMB54v5+5vnZ+LsBTCGuBlEblRRPo6twkwpZHXmhBEYtkj3HizKuZsKWfZ0AzmbKmICusiEtnIDb35hyNPZMrkRzmnRRXD81JqLYsN65LClFXlITmnEhoCCZ19CHiI2kUC3S0M4/xcCdxhjHmhwUJ4OLid21phdd+7Bmvpag0wzhizxd/xEtXBnShhpJ4O2bGLrV7RT/ZLY+ziUkTAJKVEtKdBuJzG/s4JljK95T8lzL4mnb98Qo0MoS737XA4aH/OWUhVKdtHZdbpudFuehEllcE9p9I0mlSiXEROB87AUgjLgTuwuuK5UwbsNMYcIgpIVGURiVj5SOB5k0tLhq/+mFkT+eO6CUXqhtPUpkSNpb6bf0oSpCdDSSWUV4VvXsaOHsXKN16k5ynCk/3S6u6P499orBLMfhaXAhuj3cGciMoiFsJIQ0EknuAbI1PNvjDLFilr05dVUTMmQX6jsUTQQmeNVScqqhVFouEKS8y/sCsDz5eoDiMNNuEI+wyWTC7CLVukkhZrfBVdUqK2dpbSMBqclOd0Nv8eOA/Ll+COMca0C5JsjSaRLIuxo0cx4/nnsAnckGuvVS4h3p/coukJPhCZasaESbZIWZuu856TXca63f5ri6p/InoI5jJUf+AtrLpOvwQWA82AfOBb4GNjzC1NlriJJIqycJn6VRWlfDQ8gz5zj7NtVAZtM6NnSSZURGtb0WjqER2pJbpoUphKwwimslgDrAfGYjU96m6M+VREzgXeA/5kjFkYBJmbRKIoi7GjR/HBv17gZycLzw5Ir4kEmtr3J4MvXq0LvSH5JlJOdoguhak0jGAqi8PA9ViWRSXwM2PMOue+YcA9xpiId7NLBGXhblXsvPOnSKCOzxUnhHWhNyTfROMSnRL9+FIWAfXgdqMaqDTGGBE5AJwGrHPu2wNE3F+RKLgciBfm2Gs5tW/ITeb0aUV1irHFWi9mfySiAgiU+vqCuwh1f3AlPmlouY8vsHIuADYAd4lIjoi0xsrg/iZ4oin14SqhsOtgBRN6pdbaN7FXKqn2pDrF2F77z7sJUQZEib8Ofkp00FDLYh7Q3vnvh7CWo35wfq4CBgVJLsUHLqui5yl1wxJzsmzc3NHGwxP+zD9enFXrO64yILr8EN+sX7uGVeuLmLbK97hIWZv6W4xNmtTPQkROAa4A0oGlxhjPzO6IEM8+i0CTnc57toSdX/9A27ZtE6YMiBL96G8xuglaUp4nxpgfjDEvGGOmR4uiiHcCTXZyWReu72g3OSUa0N9iDFNfc27XC8upXRXgq9Lf8cLx6tatm4lX8nvkmWZ2DPh/nfE/J5o9e/aYllnpZs+4TGMeyjZ7xmWaVtnpxuFwNPjce/bsMX1792rUdxOBaJmfaJHDk2D+FpXQgNWh1Ot9NRCfxSRqV5xVIogrCiiwpLRiHp7wZy9d02jUerGuNfsmWuYnWuTwxHsHv8b9FpXwE/Ye3OEgnn0WLgJJSrv93Upe+bycnaPSmpyYpWvNvomW+YmkHL76jUcySVAJnJD5LJTIsX7tGqatKkIeOVbva86nxxnaKTgtM3Wt2TfRMj+RlMNX461wtG9VQotaFnFKMGsnJWr58/rwfIKOlvmJpBy+LJporeOl1EUtiwQkmIlZ3teaE/dp0PMJOlrmJ9Ry+Eqm82XRRHuSoBIg9Xm+Y/kVz9FQgZLfIy+giKn8Hnk+j+MZweJ6RSKSJRqifFzz8entGaZVdrrZvHlzyOcnkOsOx9/prjv/YFqmJ5mxo0f5PLfnOYP1W1RCDz6iodSyiFNWrtsUkGL1V2Mpmtaafa2Jh1MG9yfokcNvDvn8BHLdof47uepNLRuSVqd5kz+LJli/RSWyqM9CqZdoWmuOhmgjz/nY7Kjk5zOPU/DH0LUNDeS6w/F3qq+/u0Y5xRfqs1AaRTStNUdDtJHnfMzZUsmtXe0hnZ9ArjvUfyfPVrHurWGjyfJUQotaFkq9REvPiGiINvL2BN1zZjGrvq/y883Gz0+g1x3qv1N9HfeKz7uO119bGBWWpxIcgtb8KFZQZRFfRKo9qD8ZQi1LNFy3r2Wm858r4ZauaUz7pfe+GaCNlmINVRZKzBINa+KR8N1Ew3WDbyV55lNFfHOk2u8xErVbYSyiPgulXqK9EU00rIlHwncTDdft6avwZPWIZrTKTq/TaEujnOITtSwSnNtHDOOVeXMZdPMw/jljlv8vhJFoicYKt+8mHNftq46Ti0Dqj+kyU3yhy1CKVxwOB+eddSrJVFElSXzx1Q9R5YhM1JtVOK577OhRzJnxPMNv+129x4iWAAclfPhSFhHPtg7FSzO4A2PkLUNNZgrm09szTFYK5vYRw4N6/D179pjL8i8yv+h5caMyiBM18zfU1+2Zha79JBQXaAa34onD4WDBK/O4taudrjlJjOhqZ8Erc4Pqu5gy+VE2bljHpg1rG7XGnqiZv6G+7mjIWVFiD1UWCcpDD9yPqa7iT/mpAPwpPxVTVcVDD/w5KMd3OBzMmjWLJAzLhzZj9uxZUetETyR8Jdgpii9UWSQg7laFez2fYFoXNb3C81LompPE4A5Gn2CjgGipkKvEHqosEhBPq8JFsKwLl1Wx62AF9+WnADA+PzkurYtoDz12p75QWLUulEBQZZFgeLMqXORk2bilq535L8/ms88+a/Q5aqyKLim1nmDjybpwOBz8oufF5F/YlU9Wr4yJ64qG3A0ldtHQ2QTDyqt4iZ131l8ptd30Is477zw2bf1vg4/vcDhof85ZSFUp20dl1sk+zn2hiu07Y79W0NjRo5jxwnPYgKvOTebdb+1RXQMpWnJWlOhGM7iVGpYsfochnX1XSh3Zzc6XO79o1LKEN6vC/djxYF385LyHFcMyeOfLSq47T6L6uqKpgrASm6hlkWAEmmh1Rgsbvxny+wYlffmyKmrGFFaT+3wl27/8OmafYMeOHsXKN16g5yk2nuyXxuh3Syivgtd3JUftk7km2CmBoBncil+CUQbcuom+SM9ThCf7pdU77o53S0m9+LaYzLr2phAdhdXkPlvEwE7NyLx4eExel6KAb2WRHG5hlOjEe0iltT3Qm9/6tWvYvr+CDbth2tpyn2MvrPyoUXIGUtMolFjLbJX0PMXDed/JTnlVBXPmzOK+8ROj0rpQlKagPgslaCGVK9dtorjce7bxXXf+gbvyMzEPZXNXfib5PXs1StZI9uH+KSS4siYk2MX4S1J5bXtF1PsuFKWxqLJQQh5SGaysYddxlg1Ji0hegMuqqNd538mOOK0LzVlQ4g1VFgmOv54FwUjYClbWcCRrGvmyKlyodaHEM6osEpxQh1QGa4kr0jWNfIUEu8jJsnFjRztzPj3Ouk9Wh0UuRQkX6uBOcNavXcOq9UVMW+V7XH55425+vpe4AnegB8MB3xQa4rzX8FMlHtHQWSVkBCtrOFr6UStKvKMZ3EpECNYSl9Y0UpTIo5aFEjKCkTWsNY0UJXyoZaFEhGB0fNOaRpEllkqwK6FFHdxKVBNqB3xDiXQGebhxT4LUMiaJTVgtCxHpKyLLRWSviJSJyA8islBEOniMaykiL4rIjyJSLCJLRaRTOGVVooNo68MdyQzycBPpJEglugj3MlQrYCNwJ/Ar4M9ALvCJiJwOICICvAn0A0YD1wF2YIWInBJmeRWlhkjdPCO1FBTJJEgl+girsjDGvGKMudcY87ox5kNjzFzgWiALGOgc9mugJzDEOX6xc5sNuC+c8iqKO4HePIN9c4+ENRPpJEgl+ogGB/dB53uF8/3XwB5jzArXAGPMUeAt4Oowy6YoQMNunsG8uUfKmglWiRYlfoiIshCRJBFJEZFzgH8Ce4EFzt25wFYvX9sGnCYimWESU1EA64bd6+JuDO1k83vzDPbNPRJLQcEq0aLEF5GyLNYCZcBOoDPQ2xiz37mvFXDYy3cOOd9bejugiNwuIhtEZMOBAweCLa+SwDz0wP3s3uPgTxdLre3ebp7BvLlHailIkyAVb0RKWQwBLgYGAceA90XkDOc+AbxlCoqXbTUYY543xnQ3xnRv3bp1MGVVEhiHw8GCV+YxslvdvuWeN89g39wjsRQUjirESmwSEWVhjNlhjFlrjHkFuBzIBO537j6EZV144rIovFkdihISHnrgfkx1Fffnp3rd737zDObNPVJLQZoEqdRHVJT7EJENwBFjTB8RmQn8yhhziseY2cAvjDGn+zuelvtQgoHD4eC8s05lRBcb0/ql1ztu7NJqjp9/Ha8tXBi0YodjR4+CTXN5sk/dm/bYpdXIBUNDkiQXjBItSuziq9xHxJWFiLQBCoB5xpjficg1wBvAZcaYD51jsoGvgfnGmNH+jqnKQgkGY0eP4t8v/5NvjlT7HXtGzolcc1ZpUG7uWg9LiRRRoyxE5A3gU+AzLF/FucBYoC1woTFmp4jYgJXAqcC9WMtOf8ZyhHcxxnzv7zyqLJSm0pCy6MG+ufuyKmrGhNC6UBIXX8oi3LWhPgGuB+4GUoDvgQ+Ax40x3wAYY6pF5Ergb8CzQBqwBmsJyq+iUJRg0JCmTYGt8wfeqCna6mEpCkTBMlQoUMtCaQoNtRR0nV+JF7REuaI0gIZGBEVbscOmomXJFW9oiXJF8SDRl4G0LLniDV2GUhSlBtcS3LKbbPRZUK0RVwmGLkMpihIQWpZcqQ9VFoqiAFqWXPGNKgtFUQAtS674RpWFoihallzxiyoLRVG0LLniF1UWipLgaFlyJRBUWShKgqNlyZVA0KQ8RUlwEj0JUQkMVRaKkuDEShkSJbLoMpSiKIriF1UWiqI0GC02mHioslAUpcG4FxtUEgNVFoqiNAhXqO2yIWkaUptAqLJQFKVBaLHBxESVhaIoAaPFBhMXVRaKogSMFhtMXFRZKIoSEFpsMLFRZaEoSkBoscHERpWFoih+0WKDiioLRVH8osUGFa0NpSiKX7TYoKLKQlEUv2ixQUWXoRRFURS/qLJQFEVR/KLKQlEURfGLGGMiLUPQEZEDwLeRliPKOBH4MdJCxAg6V4Gh8xQ4sTJXpxtjWnvbEZfKQqmLiGwwxnSPtByxgM5VYOg8BU48zJUuQymKoih+UWWhKIqi+EWVReLwfKQFiCF0rgJD5ylwYn6u1GehKIqi+EUtC0VRFMUvqiwURVEUv6iyiCNEZJyIvCUiDhExIvKwlzE5IvK4iGwQkaMickBElolIr3qOeY2IbBKRUhH5VkQmiEhSyC8mxAQyV25jR4rIf0WkTES+EJHf1zMuLufKExE5QUSeEpGvRKRERL4WkadFpE58fqBzF8+IyMkiMlNE9jrn4WsRedzLuKieK1UW8cVI4CTg3z7GdANuAP4DDASGA6XAByJypftAEekL/AtYD1wBPAVMACYHWe5IEMhcISIjgX9izUM/4DXgWRH5g8e4eJ6rGkREgDeBQcBfsa71r8BNwJvO/a6xAc1dPCMiZwDrgHOBPwK/Ah4GKj3GRf9cGWP0FScvwOZ8TwYM8LCXMS2AZI9tycAXwEce2zcBH3psexAoB9pG+nrDMFfJwH5gjsf2mVjZuPZEmCuPazrXOV+3e2z/vXP7eQ2du3h+AYuxlEW91xsrc6WWRRxhjKkOYMwRY0ylx7ZKYDNwsmubiJwK5AEvexxiLmDHeqKMWQKZK+BnQGu8z8EJQE+I/7nyIMX5fsxj+xHnu+ueEtDcxTMi0g7oC/zdGFPhY2hMzJUqCwURScH6we5w25zrfN/qPtYY8zVwHOgQHukiitc5ALY53zv4Ghenc7UN+AiYKCLdRSRTRC7EsqIWGWNcv6FA5y6eyXe+l4jI+05fxGEReUlETnAbFxNzpcpCAWsN9RTgL27bWjnfD3sZf9htfzxT3xwc8tifMHNlrPWR/ljLluuBQmAt8BVwndvQQOcunvkf5/tMYCeWhfknYADwnoi47r8xMVeqLKIUEenjjNLx9/qgiecZBNwPPGqM+dh9l/PdW9ameNkWMUI4V77mINBxUTVXnjRy7l4ALsbyU1zqfO8OvO52Awx07mKGRsyVay4+MMbcYYxZbox5HhiFFWjS13Vo53tUz5W2VY1eVgPtAxh3vLEnEJGrgNnADGPMQx67fT3VtHDbHw2Eaq7c58Dhtr2Vx/5YmitPGjR3IjIAK/KpjzFmmXPfRyLyFbAEuAor0i7QuYslGvo7O+h8f99j/xLne1dgETEyV6osohRjzHHgv6E6vohcjhWe9wbwOy9DXOulucAat++dATQDtodKtoYSwrlynwP3/8SuNeTtXsZF9Vx50oi56+R8X++xfZ3zvT2Wsgh07mKGRsyVaw7qsxiqPcZF9VzpMlQCIiI/w/oPvQy42VtkkDHmO2ALMNhj181ABdYTUbyzBit00dscHAJWQcLN1V7n+4Ue2y9yvu92vgc0d3HOJ1jz1c9ju+uzS+HGxFypZRFHiEh34Ax+egjoICIDnf9+1xhzXETOB97B+nH+FejmlkeFMeYTt0OOB94WkX8Cr2CZzROAp4wxe4lhApkrY0yFiEzESo7aDSwFegMjgNHGmHK3Q8btXHnwf8D/A14SkUexnrTPBx4CvseyVGng3MUlxphKEbkfmC0i/8Cau7Ox5u8DYLlzXGzMVaQTPfQVvBeW/8HU8zrDOWa4jzHGyzGvxXpqLgO+wwqRTIr0tYZjrtzG/g4rmqUM+BIYVc8x43KuvFznqcAM4Gus7P+vsZzeJ3sZG9DcxfMLGIIVFluGtcz0dyAz1uZKS5QriqIoflGfhaIoiuIXVRaKoiiKX1RZKIqiKH5RZaEoiqL4RZWFoiiK4hdVFoqiKIpfVFkocY2IPCwiEY8PF5EP3IvxiUieU7agVxQVP21iFaUxaAa3ooSHUR6f87Cynl8mSgrFKYovVFkoShgwxkRFMThFaSy6DKUkFCKSLSJPi8geZ+eyL0RkrLgVyBKRy5xLOb92jv1RRA6IyMsi0sLjeK1F5BUROebsgjbL+T0jIpe5jatZhhKR4cAs564v3fognOF8GecY9/Nc5uWYSSLymIg4ROS48xy5eEFEuojIm04ZS0RklYhc0viZVBINVRZKwuBszPMOcAvwBFbvhcXAVKzibp48hVUrahAwCasT3FMeY/4PqwPan4EbsarM/t2PKO8Ajzn//VuslrY/o3Z56kB4GKuA4TzgGqw+CW96DhKRC7B6MbQCRjqv4yCwVES6NfCcSoKiy1BKItEf6AncYoyZ7dy2REQygLtFZKox5ke38R8ZY0a7jTsPuE1EhhtjjIj8ynm8G4wxC53j3hORN4HT6hPCGHNARAqcHzcbY3a59rlXAPaFiLQExgLPG2PucZOxCvhfj+F/xSps2Ns4K5iKyHtYxe0mYikaRfGJWhZKItELq+HMKx7bXwZSsJ7u3XnH4/PnQCrQxvn5YqAKZ1luN15vsqT+6QRkAAs9ti9w/yAi6VitT18DqkUkWUSSsVp5LsWaE0Xxi1oWSiLRCjhkjCnz2L7Xbb87nlFKru+lOd9zgMPGmAqPcfuaJGVg5NRzLs/PrYAkLAtiorcDiYjNeGmApSjuqLJQEolDQCsRSTG1G8q0db4f9PIdXziAliJi91AYber7QgCUOt9TPLaf4OXcrnNtc9vuee4jWNbUM8BL3k6oikIJBF2GUhKJD7F+87/12D4YKMdqg9kQPsF6av+Nx3bP43vDZaWke2zf59zX0WP7AI/PnwHFwPUe2290/2CMKQY+BroAnxpjNni+ApBVUdSyUBKKRcBK4B8i0hrribw/cBvwuIdz2y/GmCUishJ4XkROBHYBA7FuzGA90deHK+/iDhGZgxVF9ZkxplxEXgVuFZGdwBdYiuIyj3MfEZEngQdEpBArEqoHcKuXc40DPsJyvs/AskpOBC7A6uR3f0OuW0lM1LJQEgbncssAYA7wJywH9gCsm+kDjTzstVjht3/Bcjan8ZNv4KgPWbZghb5ehaXA1gP/49w9Bisk92HgVecxR9c5iLV/MlbbzjeBXzmP53muT7EUyUFgOpZieQrLSf5RQFepJDzaVlVRgoyIPIPV67yVF2e6osQkugylKE3AmWndHGtJKwXoB/we+KsqCiWeUGWhKE2jGLgLaIeVg/E1Vlb1XyMok6IEHV2GUhRFUfyiDm5FURTFL6osFEVRFL+oslAURVH8ospCURRF8YsqC0VRFMUv/x+wwoC7hlnJzgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.discrete_scatter(X.iloc[:, 0], X.iloc[:, 1], y)\n",
    "plt.xlabel(\"longitude\")\n",
    "plt.ylabel(\"latitude\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "##### Real boundary between Canada and USA\n",
    "\n",
    "In real life we know what's the boundary between USA and Canada. \n",
    "\n",
    "![](img/canada-us-border.jpg)\n",
    "\n",
    "<!-- <img src=\"img/canada-us-border.jpg\" height=\"500\" width=\"500\">  -->\n",
    "\n",
    "[Source](https://sovereignlimits.com/blog/u-s-canada-border-history-disputes)\n",
    "\n",
    "Here we want to pretend that we do not know this boundary and we want to infer this boundary based on the limited training examples given to us. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = DecisionTreeClassifier(max_depth=1)\n",
    "model.fit(X, y)\n",
    "plot_tree_decision_boundary_and_tree(\n",
    "    model,\n",
    "    X,\n",
    "    y,\n",
    "    height=6,\n",
    "    width=16,\n",
    "    eps=10,\n",
    "    x_label=\"longitude\",\n",
    "    y_label=\"latitude\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = DecisionTreeClassifier(max_depth=2)\n",
    "model.fit(X, y)\n",
    "plot_tree_decision_boundary_and_tree(\n",
    "    model,\n",
    "    X,\n",
    "    y,\n",
    "    height=6,\n",
    "    width=16,\n",
    "    eps=10,\n",
    "    x_label=\"longitude\",\n",
    "    y_label=\"latitude\",\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Practice exercises \n",
    "\n",
    "- If you want more practice, check out module 2 in [this online course](https://ml-learn.mds.ubc.ca/en/module2). All the sections **without** video or notes symbol are exercises.  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{admonition} Attention\n",
    ":class: important\n",
    "If all of you are working on the exercises, especially coding exercises, at the same time, you might have to wait for the real-time feedback for a long time or you might even get an error. There is no solution for this other than waiting for a while and trying it again. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```{note}\n",
    "Some background on [the online course](https://ml-learn.mds.ubc.ca/en/) above: This course is designed by Hayley Boyce, Mike Gelbart, and myself. It'll be a great resource at the beginning of this class, as it give you a chance to practice what we learn and the framework will provide you real-time feedback. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br><br><br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Final comments and summary\n",
    "\n",
    "What did we learn today? \n",
    "\n",
    "- There is a lot of terminology and jargon used in ML. Some of the basic \n",
    "terminology includes:\n",
    "    - Features, target, examples, training\n",
    "    - Supervised vs. Unsupervised machine learning     \n",
    "    - Classification and regression    \n",
    "    - Accuracy and error    \n",
    "    - Parameters and hyperparameters\n",
    "    - Decision boundary "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Baselines and steps to train a supervised machine learning model \n",
    "    - Baselines serve as reference points in ML workflow. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Decision trees    \n",
    "    - are models that make predictions by sequentially looking at features and checking whether they are above/below a threshold\n",
    "    - learn a hierarchy of if/else questions, similar to questions you might ask in a 20-questions game.       \n",
    "    - learn axis-aligned decision boundaries (vertical and horizontal lines with 2 features)    \n",
    "    - One way to control the complexity of decision tree models is by using the depth hyperparameter (`max_depth` in `sklearn`). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![](img/eva-logging-off.png)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Python [conda env:cpsc330]",
   "language": "python",
   "name": "conda-env-cpsc330-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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}