DefineAndSolveMLProblem.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lab 8: Define and Solve an ML Problem of Your Choosing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os \n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this lab assignment, you will follow the machine learning life cycle and implement a model to solve a machine learning problem of your choosing. You will select a data set and choose a predictive problem that the data set supports.  You will then inspect the data with your problem in mind and begin to formulate a  project plan. You will then implement the machine learning project plan. \n",
    "\n",
    "You will complete the following tasks:\n",
    "\n",
    "1. Build Your DataFrame\n",
    "2. Define Your ML Problem\n",
    "3. Perform exploratory data analysis to understand your data.\n",
    "4. Define Your Project Plan\n",
    "5. Implement Your Project Plan:\n",
    "    * Prepare your data for your model.\n",
    "    * Fit your model to the training data and evaluate your model.\n",
    "    * Improve your model's performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 1: Build Your DataFrame\n",
    "\n",
    "You will have the option to choose one of four data sets that you have worked with in this program:\n",
    "\n",
    "* The \"census\" data set that contains Census information from 1994: `censusData.csv`\n",
    "* Airbnb NYC \"listings\" data set: `airbnbListingsData.csv`\n",
    "* World Happiness Report (WHR) data set: `WHR2018Chapter2OnlineData.csv`\n",
    "* Book Review data set: `bookReviewsData.csv`\n",
    "\n",
    "Note that these are variations of the data sets that you have worked with in this program. For example, some do not include some of the preprocessing necessary for specific models. \n",
    "\n",
    "#### Load a Data Set and Save it as a Pandas DataFrame\n",
    "\n",
    "The code cell below contains filenames (path + filename) for each of the four data sets available to you.\n",
    "\n",
    "<b>Task:</b> In the code cell below, use the same method you have been using to load the data using `pd.read_csv()` and save it to DataFrame `df`. \n",
    "\n",
    "You can load each file as a new DataFrame to inspect the data before choosing your data set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>education-num</th>\n",
       "      <th>marital-status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>sex_selfID</th>\n",
       "      <th>capital-gain</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>hours-per-week</th>\n",
       "      <th>native-country</th>\n",
       "      <th>income_binary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>39.0</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>77516</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>2174</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>50.0</td>\n",
       "      <td>Self-emp-not-inc</td>\n",
       "      <td>83311</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38.0</td>\n",
       "      <td>Private</td>\n",
       "      <td>215646</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Handlers-cleaners</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>53.0</td>\n",
       "      <td>Private</td>\n",
       "      <td>234721</td>\n",
       "      <td>11th</td>\n",
       "      <td>7</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Handlers-cleaners</td>\n",
       "      <td>Husband</td>\n",
       "      <td>Black</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>28.0</td>\n",
       "      <td>Private</td>\n",
       "      <td>338409</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Prof-specialty</td>\n",
       "      <td>Wife</td>\n",
       "      <td>Black</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>Cuba</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    age         workclass  fnlwgt  education  education-num  \\\n",
       "0  39.0         State-gov   77516  Bachelors             13   \n",
       "1  50.0  Self-emp-not-inc   83311  Bachelors             13   \n",
       "2  38.0           Private  215646    HS-grad              9   \n",
       "3  53.0           Private  234721       11th              7   \n",
       "4  28.0           Private  338409  Bachelors             13   \n",
       "\n",
       "       marital-status         occupation   relationship   race  sex_selfID  \\\n",
       "0       Never-married       Adm-clerical  Not-in-family  White  Non-Female   \n",
       "1  Married-civ-spouse    Exec-managerial        Husband  White  Non-Female   \n",
       "2            Divorced  Handlers-cleaners  Not-in-family  White  Non-Female   \n",
       "3  Married-civ-spouse  Handlers-cleaners        Husband  Black  Non-Female   \n",
       "4  Married-civ-spouse     Prof-specialty           Wife  Black      Female   \n",
       "\n",
       "   capital-gain  capital-loss  hours-per-week native-country income_binary  \n",
       "0          2174             0            40.0  United-States         <=50K  \n",
       "1             0             0            13.0  United-States         <=50K  \n",
       "2             0             0            40.0  United-States         <=50K  \n",
       "3             0             0            40.0  United-States         <=50K  \n",
       "4             0             0            40.0           Cuba         <=50K  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# File names of the four data sets\n",
    "adultDataSet_filename = os.path.join(os.getcwd(), \"data\", \"censusData.csv\")\n",
    "airbnbDataSet_filename = os.path.join(os.getcwd(), \"data\", \"airbnbListingsData.csv\")\n",
    "WHRDataSet_filename = os.path.join(os.getcwd(), \"data\", \"WHR2018Chapter2OnlineData.csv\")\n",
    "bookReviewDataSet_filename = os.path.join(os.getcwd(), \"data\", \"bookReviewsData.csv\")\n",
    "\n",
    "\n",
    "df = pd.read_csv(adultDataSet_filename, header=0)# YOUR CODE HERE\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['State-gov', 'Self-emp-not-inc', 'Private', 'Federal-gov',\n",
       "       'Local-gov', nan, 'Self-emp-inc', 'Without-pay', 'Never-worked'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['occupation'].unique()\n",
    "df['workclass'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 2: Define Your ML Problem\n",
    "\n",
    "Next you will formulate your ML Problem. In the markdown cell below, answer the following questions:\n",
    "\n",
    "1. List the data set you have chosen.\n",
    "2. What will you be predicting? What is the label?\n",
    "3. Is this a supervised or unsupervised learning problem? Is this a clustering, classification or regression problem? Is it a binary classificaiton or multi-class classifiction problem?\n",
    "4. What are your features? (note: this list may change after your explore your data)\n",
    "5. Explain why this is an important problem. In other words, how would a company create value with a model that predicts this label?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the census dataset, I wish to predict whether an individual's salary is above or below 50K, using the occupations, work class, and hours per week as features. Initially, I hoped to use word embeddings to process the categorical feature values, but realized that was probably unnecessary, considering there are never more than 15 different categories, and one-hot encoding would be easy to implement and evaluate. Thus, the actual content of the categories will not be taken into consideration as it might have using word embeddings, but it shouldn't be necessary in this situation.\n",
    "\n",
    "This is a supervised learning binary classificiation problem. Not necessarily sure if this will maximize shareholder value, but it can provide insight into how much different careers earn."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 3: Understand Your Data\n",
    "\n",
    "The next step is to perform exploratory data analysis. Inspect and analyze your data set with your machine learning problem in mind. Consider the following as you inspect your data:\n",
    "\n",
    "1. What data preparation techniques would you like to use? These data preparation techniques may include:\n",
    "\n",
    "    * addressing missingness, such as replacing missing values with means\n",
    "    * finding and replacing outliers\n",
    "    * renaming features and labels\n",
    "    * finding and replacing outliers\n",
    "    * performing feature engineering techniques such as one-hot encoding on categorical features\n",
    "    * selecting appropriate features and removing irrelevant features\n",
    "    * performing specific data cleaning and preprocessing techniques for an NLP problem\n",
    "    * addressing class imbalance in your data sample to promote fair AI\n",
    "    \n",
    "\n",
    "2. What machine learning model (or models) you would like to use that is suitable for your predictive problem and data?\n",
    "    * Are there other data preparation techniques that you will need to apply to build a balanced modeling data set for your problem and model? For example, will you need to scale your data?\n",
    " \n",
    " \n",
    "3. How will you evaluate and improve the model's performance?\n",
    "    * Are there specific evaluation metrics and methods that are appropriate for your model?\n",
    "    \n",
    "\n",
    "Think of the different techniques you have used to inspect and analyze your data in this course. These include using Pandas to apply data filters, using the Pandas `describe()` method to get insight into key statistics for each column, using the Pandas `dtypes` property to inspect the data type of each column, and using Matplotlib and Seaborn to detect outliers and visualize relationships between features and labels. If you are working on a classification problem, use techniques you have learned to determine if there is class imbalance.\n",
    "\n",
    "<b>Task</b>: Use the techniques you have learned in this course to inspect and analyze your data. You can import additional packages that you have used in this course that you will need to perform this task.\n",
    "\n",
    "<b>Note</b>: You can add code cells if needed by going to the <b>Insert</b> menu and clicking on <b>Insert Cell Below</b> in the drop-drown menu."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education-num</th>\n",
       "      <th>capital-gain</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>hours-per-week</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>32399.000000</td>\n",
       "      <td>3.256100e+04</td>\n",
       "      <td>32561.000000</td>\n",
       "      <td>32561.000000</td>\n",
       "      <td>32561.000000</td>\n",
       "      <td>32236.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>38.589216</td>\n",
       "      <td>1.897784e+05</td>\n",
       "      <td>10.080679</td>\n",
       "      <td>615.907773</td>\n",
       "      <td>87.303830</td>\n",
       "      <td>40.450428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>13.647862</td>\n",
       "      <td>1.055500e+05</td>\n",
       "      <td>2.572720</td>\n",
       "      <td>2420.191974</td>\n",
       "      <td>402.960219</td>\n",
       "      <td>12.353748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>17.000000</td>\n",
       "      <td>1.228500e+04</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>28.000000</td>\n",
       "      <td>1.178270e+05</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>40.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>37.000000</td>\n",
       "      <td>1.783560e+05</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>40.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>48.000000</td>\n",
       "      <td>2.370510e+05</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>45.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>90.000000</td>\n",
       "      <td>1.484705e+06</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>14084.000000</td>\n",
       "      <td>4356.000000</td>\n",
       "      <td>99.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                age        fnlwgt  education-num  capital-gain  capital-loss  \\\n",
       "count  32399.000000  3.256100e+04   32561.000000  32561.000000  32561.000000   \n",
       "mean      38.589216  1.897784e+05      10.080679    615.907773     87.303830   \n",
       "std       13.647862  1.055500e+05       2.572720   2420.191974    402.960219   \n",
       "min       17.000000  1.228500e+04       1.000000      0.000000      0.000000   \n",
       "25%       28.000000  1.178270e+05       9.000000      0.000000      0.000000   \n",
       "50%       37.000000  1.783560e+05      10.000000      0.000000      0.000000   \n",
       "75%       48.000000  2.370510e+05      12.000000      0.000000      0.000000   \n",
       "max       90.000000  1.484705e+06      16.000000  14084.000000   4356.000000   \n",
       "\n",
       "       hours-per-week  \n",
       "count    32236.000000  \n",
       "mean        40.450428  \n",
       "std         12.353748  \n",
       "min          1.000000  \n",
       "25%         40.000000  \n",
       "50%         40.000000  \n",
       "75%         45.000000  \n",
       "max         99.000000  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# YOUR CODE HERE\n",
    "df.describe()\n",
    "#according to the internet, capital gain/loss are not counted as part of income by the census \n",
    "#bureau so we won't use that as an income feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        age     workclass  fnlwgt     education  education-num  \\\n",
      "5361   18.0  Never-worked  206359          10th              6   \n",
      "10845  23.0  Never-worked  188535       7th-8th              4   \n",
      "14772  17.0  Never-worked  237272          10th              6   \n",
      "20337  18.0  Never-worked  157131          11th              7   \n",
      "23232  20.0  Never-worked  462294  Some-college             10   \n",
      "32304  30.0  Never-worked  176673       HS-grad              9   \n",
      "32314  18.0  Never-worked  153663  Some-college             10   \n",
      "\n",
      "           marital-status  occupation   relationship   race  sex_selfID  \\\n",
      "5361        Never-married  Unemployed      Own-child  White  Non-Female   \n",
      "10845            Divorced  Unemployed  Not-in-family  White  Non-Female   \n",
      "14772       Never-married  Unemployed      Own-child  White  Non-Female   \n",
      "20337       Never-married  Unemployed      Own-child  White      Female   \n",
      "23232       Never-married  Unemployed      Own-child  Black  Non-Female   \n",
      "32304  Married-civ-spouse  Unemployed           Wife  Black      Female   \n",
      "32314       Never-married  Unemployed      Own-child  White  Non-Female   \n",
      "\n",
      "       capital-gain  capital-loss  hours-per-week native-country income_binary  \n",
      "5361              0             0            40.0  United-States         <=50K  \n",
      "10845             0             0            35.0  United-States         <=50K  \n",
      "14772             0             0            30.0  United-States         <=50K  \n",
      "20337             0             0             NaN  United-States         <=50K  \n",
      "23232             0             0            40.0  United-States         <=50K  \n",
      "32304             0             0            40.0  United-States         <=50K  \n",
      "32314             0             0             4.0  United-States         <=50K  \n"
     ]
    }
   ],
   "source": [
    "df['occupation'].isnull().sum()\n",
    "\n",
    "df.loc[df['occupation'].isnull() & (df['workclass'] == 'Never-worked'), 'occupation'] = 'Unemployed'\n",
    "\n",
    "print(df[df['workclass'] == 'Never-worked'])\n",
    "#out of all the null values for occupation, the ones that have 'Never-worked' in the workclass\n",
    "#are change to unemployed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.dropna(subset=['occupation'])\n",
    "df['workclass'].isnull().sum()\n",
    "#no values in occupation or workclass are null"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.dropna(subset=['hours-per-week'])\n",
    "df['hours-per-week'].isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 4: Define Your Project Plan\n",
    "\n",
    "Now that you understand your data, in the markdown cell below, define your plan to implement the remaining phases of the machine learning life cycle (data preparation, modeling, evaluation) to solve your ML problem. Answer the following questions:\n",
    "\n",
    "* Do you have a new feature list? If so, what are the features that you chose to keep and remove after inspecting the data? \n",
    "* Explain different data preparation techniques that you will use to prepare your data for modeling.\n",
    "* What is your model (or models)?\n",
    "* Describe your plan to train your model, analyze its performance and then improve the model. That is, describe your model building, validation and selection plan to produce a model that generalizes well to new data. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "SOOO I have already removed null values from the features I plan to use (occupation, workclass, hours-per-week) and now will use one-hot encoding to turn the two categorical values into some kind of numerical representation. then do the checking for outliers and all that stuff, idk if i have to standerdize it i will figure that out too, only for the hours worked i suppose since the others are all technically binary. \n",
    "\n",
    "The model i would like to implement is decision trees! but I am not yet sure if random forest enseamble method or just standard decision tree is the best decision in this situation. I shall consult the oracle (chatgpt).\n",
    "\n",
    "In terms of training, just do the usual, pick some very basic parameters and then adjust to improve model performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 5: Implement Your Project Plan\n",
    "\n",
    "<b>Task:</b> In the code cell below, import additional packages that you have used in this course that you will need to implement your project plan."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOUR CODE HERE\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>Task:</b> Use the rest of this notebook to carry out your project plan. \n",
    "\n",
    "You will:\n",
    "\n",
    "1. Prepare your data for your model.\n",
    "2. Fit your model to the training data and evaluate your model.\n",
    "3. Improve your model's performance by performing model selection and/or feature selection techniques to find best model for your problem.\n",
    "\n",
    "Add code cells below and populate the notebook with commentary, code, analyses, results, and figures as you see fit. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "so first lets transform the output data into the proper binany calsification label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['<=50K', '>50K'], dtype=object)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# YOUR CODE HERE\n",
    "df['income_binary'].unique()\n",
    "\n",
    "# Verify the changes\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['<=50K' '>50K']\n",
      "[1 0]\n"
     ]
    }
   ],
   "source": [
    "df.rename(columns={'income_binary': 'income <=50k'}, inplace=True)\n",
    "\n",
    "# Check for unique values before mapping\n",
    "print(df['income <=50k'].unique())\n",
    "\n",
    "# Adjust the values for binary classification to 0 and 1\n",
    "df['income <=50k'] = df['income <=50k'].map({'<=50K': 1, '>50K': 0})\n",
    "\n",
    "print(df['income <=50k'].unique())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>education-num</th>\n",
       "      <th>marital-status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>sex_selfID</th>\n",
       "      <th>capital-gain</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>hours-per-week</th>\n",
       "      <th>native-country</th>\n",
       "      <th>income &lt;=50k</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>39.0</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>77516</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>2174</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>United-States</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>50.0</td>\n",
       "      <td>Self-emp-not-inc</td>\n",
       "      <td>83311</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>United-States</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38.0</td>\n",
       "      <td>Private</td>\n",
       "      <td>215646</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Handlers-cleaners</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>United-States</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>53.0</td>\n",
       "      <td>Private</td>\n",
       "      <td>234721</td>\n",
       "      <td>11th</td>\n",
       "      <td>7</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Handlers-cleaners</td>\n",
       "      <td>Husband</td>\n",
       "      <td>Black</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>United-States</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>28.0</td>\n",
       "      <td>Private</td>\n",
       "      <td>338409</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Prof-specialty</td>\n",
       "      <td>Wife</td>\n",
       "      <td>Black</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>Cuba</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    age         workclass  fnlwgt  education  education-num  \\\n",
       "0  39.0         State-gov   77516  Bachelors             13   \n",
       "1  50.0  Self-emp-not-inc   83311  Bachelors             13   \n",
       "2  38.0           Private  215646    HS-grad              9   \n",
       "3  53.0           Private  234721       11th              7   \n",
       "4  28.0           Private  338409  Bachelors             13   \n",
       "\n",
       "       marital-status         occupation   relationship   race  sex_selfID  \\\n",
       "0       Never-married       Adm-clerical  Not-in-family  White  Non-Female   \n",
       "1  Married-civ-spouse    Exec-managerial        Husband  White  Non-Female   \n",
       "2            Divorced  Handlers-cleaners  Not-in-family  White  Non-Female   \n",
       "3  Married-civ-spouse  Handlers-cleaners        Husband  Black  Non-Female   \n",
       "4  Married-civ-spouse     Prof-specialty           Wife  Black      Female   \n",
       "\n",
       "   capital-gain  capital-loss  hours-per-week native-country  income <=50k  \n",
       "0          2174             0            40.0  United-States             1  \n",
       "1             0             0            13.0  United-States             1  \n",
       "2             0             0            40.0  United-States             1  \n",
       "3             0             0            40.0  United-States             1  \n",
       "4             0             0            40.0           Cuba             1  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_encoded = pd.get_dummies(df, columns=['occupation', 'workclass'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>education-num</th>\n",
       "      <th>marital-status</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>sex_selfID</th>\n",
       "      <th>capital-gain</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>...</th>\n",
       "      <th>occupation_Transport-moving</th>\n",
       "      <th>occupation_Unemployed</th>\n",
       "      <th>workclass_Federal-gov</th>\n",
       "      <th>workclass_Local-gov</th>\n",
       "      <th>workclass_Never-worked</th>\n",
       "      <th>workclass_Private</th>\n",
       "      <th>workclass_Self-emp-inc</th>\n",
       "      <th>workclass_Self-emp-not-inc</th>\n",
       "      <th>workclass_State-gov</th>\n",
       "      <th>workclass_Without-pay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>39.0</td>\n",
       "      <td>77516</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>2174</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>50.0</td>\n",
       "      <td>83311</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38.0</td>\n",
       "      <td>215646</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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>53.0</td>\n",
       "      <td>234721</td>\n",
       "      <td>11th</td>\n",
       "      <td>7</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Husband</td>\n",
       "      <td>Black</td>\n",
       "      <td>Non-Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>28.0</td>\n",
       "      <td>338409</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Wife</td>\n",
       "      <td>Black</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    age  fnlwgt  education  education-num      marital-status   relationship  \\\n",
       "0  39.0   77516  Bachelors             13       Never-married  Not-in-family   \n",
       "1  50.0   83311  Bachelors             13  Married-civ-spouse        Husband   \n",
       "2  38.0  215646    HS-grad              9            Divorced  Not-in-family   \n",
       "3  53.0  234721       11th              7  Married-civ-spouse        Husband   \n",
       "4  28.0  338409  Bachelors             13  Married-civ-spouse           Wife   \n",
       "\n",
       "    race  sex_selfID  capital-gain  capital-loss  ...  \\\n",
       "0  White  Non-Female          2174             0  ...   \n",
       "1  White  Non-Female             0             0  ...   \n",
       "2  White  Non-Female             0             0  ...   \n",
       "3  Black  Non-Female             0             0  ...   \n",
       "4  Black      Female             0             0  ...   \n",
       "\n",
       "   occupation_Transport-moving occupation_Unemployed  workclass_Federal-gov  \\\n",
       "0                            0                     0                      0   \n",
       "1                            0                     0                      0   \n",
       "2                            0                     0                      0   \n",
       "3                            0                     0                      0   \n",
       "4                            0                     0                      0   \n",
       "\n",
       "   workclass_Local-gov  workclass_Never-worked  workclass_Private  \\\n",
       "0                    0                       0                  0   \n",
       "1                    0                       0                  0   \n",
       "2                    0                       0                  1   \n",
       "3                    0                       0                  1   \n",
       "4                    0                       0                  1   \n",
       "\n",
       "   workclass_Self-emp-inc  workclass_Self-emp-not-inc  workclass_State-gov  \\\n",
       "0                       0                           0                    1   \n",
       "1                       0                           1                    0   \n",
       "2                       0                           0                    0   \n",
       "3                       0                           0                    0   \n",
       "4                       0                           0                    0   \n",
       "\n",
       "   workclass_Without-pay  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  \n",
       "\n",
       "[5 rows x 36 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_encoded.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14, 8))\n",
    "sns.countplot(data=df, y='occupation', order=df['occupation'].value_counts().index)\n",
    "plt.title('Count of Each Occupation')\n",
    "plt.xlabel('Count')\n",
    "plt.ylabel('Occupation')\n",
    "plt.show()\n",
    "#wow awesome graph!!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Time to split the data wahoo!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df_encoded['income <=50k']\n",
    "\n",
    "irrelevant_columns = [\n",
    "    'age', 'education', 'education-num', 'marital-status', 'fnlwgt',\n",
    "   'relationship', 'race', 'sex_selfID', 'capital-gain',\n",
    "    'capital-loss', 'native-country', 'income <=50k'\n",
    "]\n",
    "X = df_encoded.drop(columns=irrelevant_columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>hours-per-week</th>\n",
       "      <th>occupation_Adm-clerical</th>\n",
       "      <th>occupation_Armed-Forces</th>\n",
       "      <th>occupation_Craft-repair</th>\n",
       "      <th>occupation_Exec-managerial</th>\n",
       "      <th>occupation_Farming-fishing</th>\n",
       "      <th>occupation_Handlers-cleaners</th>\n",
       "      <th>occupation_Machine-op-inspct</th>\n",
       "      <th>occupation_Other-service</th>\n",
       "      <th>occupation_Priv-house-serv</th>\n",
       "      <th>...</th>\n",
       "      <th>occupation_Transport-moving</th>\n",
       "      <th>occupation_Unemployed</th>\n",
       "      <th>workclass_Federal-gov</th>\n",
       "      <th>workclass_Local-gov</th>\n",
       "      <th>workclass_Never-worked</th>\n",
       "      <th>workclass_Private</th>\n",
       "      <th>workclass_Self-emp-inc</th>\n",
       "      <th>workclass_Self-emp-not-inc</th>\n",
       "      <th>workclass_State-gov</th>\n",
       "      <th>workclass_Without-pay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>40.0</td>\n",
       "      <td>1</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>13.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>40.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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>40.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>40.0</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32556</th>\n",
       "      <td>38.0</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32557</th>\n",
       "      <td>40.0</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",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32558</th>\n",
       "      <td>40.0</td>\n",
       "      <td>1</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32559</th>\n",
       "      <td>20.0</td>\n",
       "      <td>1</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32560</th>\n",
       "      <td>40.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>30424 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       hours-per-week  occupation_Adm-clerical  occupation_Armed-Forces  \\\n",
       "0                40.0                        1                        0   \n",
       "1                13.0                        0                        0   \n",
       "2                40.0                        0                        0   \n",
       "3                40.0                        0                        0   \n",
       "4                40.0                        0                        0   \n",
       "...               ...                      ...                      ...   \n",
       "32556            38.0                        0                        0   \n",
       "32557            40.0                        0                        0   \n",
       "32558            40.0                        1                        0   \n",
       "32559            20.0                        1                        0   \n",
       "32560            40.0                        0                        0   \n",
       "\n",
       "       occupation_Craft-repair  occupation_Exec-managerial  \\\n",
       "0                            0                           0   \n",
       "1                            0                           1   \n",
       "2                            0                           0   \n",
       "3                            0                           0   \n",
       "4                            0                           0   \n",
       "...                        ...                         ...   \n",
       "32556                        0                           0   \n",
       "32557                        0                           0   \n",
       "32558                        0                           0   \n",
       "32559                        0                           0   \n",
       "32560                        0                           1   \n",
       "\n",
       "       occupation_Farming-fishing  occupation_Handlers-cleaners  \\\n",
       "0                               0                             0   \n",
       "1                               0                             0   \n",
       "2                               0                             1   \n",
       "3                               0                             1   \n",
       "4                               0                             0   \n",
       "...                           ...                           ...   \n",
       "32556                           0                             0   \n",
       "32557                           0                             0   \n",
       "32558                           0                             0   \n",
       "32559                           0                             0   \n",
       "32560                           0                             0   \n",
       "\n",
       "       occupation_Machine-op-inspct  occupation_Other-service  \\\n",
       "0                                 0                         0   \n",
       "1                                 0                         0   \n",
       "2                                 0                         0   \n",
       "3                                 0                         0   \n",
       "4                                 0                         0   \n",
       "...                             ...                       ...   \n",
       "32556                             0                         0   \n",
       "32557                             1                         0   \n",
       "32558                             0                         0   \n",
       "32559                             0                         0   \n",
       "32560                             0                         0   \n",
       "\n",
       "       occupation_Priv-house-serv  ...  occupation_Transport-moving  \\\n",
       "0                               0  ...                            0   \n",
       "1                               0  ...                            0   \n",
       "2                               0  ...                            0   \n",
       "3                               0  ...                            0   \n",
       "4                               0  ...                            0   \n",
       "...                           ...  ...                          ...   \n",
       "32556                           0  ...                            0   \n",
       "32557                           0  ...                            0   \n",
       "32558                           0  ...                            0   \n",
       "32559                           0  ...                            0   \n",
       "32560                           0  ...                            0   \n",
       "\n",
       "       occupation_Unemployed  workclass_Federal-gov  workclass_Local-gov  \\\n",
       "0                          0                      0                    0   \n",
       "1                          0                      0                    0   \n",
       "2                          0                      0                    0   \n",
       "3                          0                      0                    0   \n",
       "4                          0                      0                    0   \n",
       "...                      ...                    ...                  ...   \n",
       "32556                      0                      0                    0   \n",
       "32557                      0                      0                    0   \n",
       "32558                      0                      0                    0   \n",
       "32559                      0                      0                    0   \n",
       "32560                      0                      0                    0   \n",
       "\n",
       "       workclass_Never-worked  workclass_Private  workclass_Self-emp-inc  \\\n",
       "0                           0                  0                       0   \n",
       "1                           0                  0                       0   \n",
       "2                           0                  1                       0   \n",
       "3                           0                  1                       0   \n",
       "4                           0                  1                       0   \n",
       "...                       ...                ...                     ...   \n",
       "32556                       0                  1                       0   \n",
       "32557                       0                  1                       0   \n",
       "32558                       0                  1                       0   \n",
       "32559                       0                  1                       0   \n",
       "32560                       0                  0                       1   \n",
       "\n",
       "       workclass_Self-emp-not-inc  workclass_State-gov  workclass_Without-pay  \n",
       "0                               0                    1                      0  \n",
       "1                               1                    0                      0  \n",
       "2                               0                    0                      0  \n",
       "3                               0                    0                      0  \n",
       "4                               0                    0                      0  \n",
       "...                           ...                  ...                    ...  \n",
       "32556                           0                    0                      0  \n",
       "32557                           0                    0                      0  \n",
       "32558                           0                    0                      0  \n",
       "32559                           0                    0                      0  \n",
       "32560                           0                    0                      0  \n",
       "\n",
       "[30424 rows x 24 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X\n",
    "#trying to think if there is a way to incorporate the fnlwgt stat into this? gonna remove for now"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=123)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7606, 24)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape\n",
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_depth': [4, 8, 16], 'min_samples_leaf': [25, 50, 100]}"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the parameter grid is recycled from a previous assignment (might adjust as needed)\n",
    "#Note these are the same values as those we used above\n",
    "hyperparams_depth = [2**n for n in range(2,5)]\n",
    "\n",
    "# Create a range of hyperparameter values for 'min_samples_leaf'. \n",
    "hyperparams_leaf = [25*2**n for n in range(0,3)]\n",
    "\n",
    "# Create parameter grid.\n",
    "param_grid={'max_depth':hyperparams_depth, 'min_samples_leaf':hyperparams_leaf}\n",
    "param_grid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = DecisionTreeClassifier()\n",
    "grid = GridSearchCV(model, param_grid, cv=5)\n",
    "grid_search = grid.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimal hyperparameters: {'max_depth': 8, 'min_samples_leaf': 100}\n",
      "Accuracy score: 0.7762291114579929\n"
     ]
    }
   ],
   "source": [
    "print('Optimal hyperparameters: {0}'.format(grid_search.best_params_))\n",
    "\n",
    "# print best accuracy score resulting from this configuration\n",
    "print('Accuracy score: {0}'.format(grid_search.best_score_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7713647120694189"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = DecisionTreeClassifier(criterion = 'entropy', max_depth = 8, min_samples_leaf = 100)\n",
    "\n",
    "    # 2. Fit the model to the training data below\n",
    "    # YOUR CODE HERE\n",
    "model.fit(X_train, y_train)\n",
    "    \n",
    "    # 3. Make predictions on the test data below and assign the result to the variable 'class_label_predictions'\n",
    "    # YOUR CODE HERE\n",
    "class_label_predictions = model.predict(X_test)\n",
    "\n",
    "    # 4. Compute the accuracy here and save the result to the variable 'acc_score'\n",
    "    # YOUR CODE HERE\n",
    "acc_score = accuracy_score(y_test, class_label_predictions)\n",
    "acc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top five features: ['hours-per-week', 'occupation_Exec-managerial', 'occupation_Prof-specialty', 'occupation_Other-service', 'workclass_Private']\n"
     ]
    }
   ],
   "source": [
    "#now lets check feature importance even though we dont have many features\n",
    "feature_imp = model.feature_importances_\n",
    "\n",
    "#2. Create a Pandas DataFrame with a list of all features and their scores. \n",
    "# Save the result to the variable 'df_features'\n",
    "\n",
    "# YOUR CODE HERE\n",
    "df_features = pd.DataFrame(data={'name': X_train.columns.values,'imp': feature_imp})\n",
    "\n",
    "#3. Sort df_features in descending order and\n",
    "# save the result to the variable 'df_sorted'\n",
    "\n",
    "# YOUR CODE HERE\n",
    "df_sorted = df_features.sort_values(by='imp', ascending=False)\n",
    "\n",
    "#4. Obtain the top 5 sorted feature names and save the result to list 'top_five' \n",
    "top_five = df_sorted.iloc[:5]['name'].tolist()\n",
    "\n",
    "# YOUR CODE HERE\n",
    "print('Top five features: {0}'.format(top_five))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "alright so i tried to perform feature selection by determining which of the 3 features we had were most useful NOT REALIZING that we totally did one-hot encoding, so this isn't really a useful evaluation metric. Now based on the fact that i really did only want to use like 3 features to make these predictions, it is probably best to leave things here as is."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgMAAAGwCAYAAAA0bWYRAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAAA+IklEQVR4nO3de1hVddr/8c/msDkIG8UEJPCUeWAyTSulmSzLkYwpTXuaSo1Mm1+FlZqnntI8VPZYeZpMmyzRRiubmZpRKzNNzERLjDJTyiMaguYBBOW09/r9Qezao7vYbmAj6/26rnVdw1rftfa9enjcN/f9/a5lMQzDEAAAMC0/XwcAAAB8i2QAAACTIxkAAMDkSAYAADA5kgEAAEyOZAAAAJMjGQAAwOQCfB2ANxwOh3JzcxUeHi6LxeLrcAAAHjIMQ6dOnVJsbKz8/Grv79OSkhKVlZV5fR2r1arg4OAaiKh+uaCTgdzcXMXHx/s6DACAlw4ePKi4uLhauXZJSYlatwxT3hG719eKiYnRvn37GlxCcEEnA+Hh4ZKk6yL+rACL1cfRALWjtHMbX4cA1JqKilJt/uw557/ntaGsrEx5R+w6kNlKtvDzrz4UnnKoZbf9KisrIxmoT6paAwEWK8kAGix7QMP6Rwc4l7po9YaFWxQWfv6f41DDbUdf0MkAAADVZTccsnvxNh674ai5YOoZkgEAgCk4ZMih888GvDm3vmNpIQAAJkdlAABgCg455E2h37uz6zeSAQCAKdgNQ3bj/Ev93pxb39EmAADA5KgMAABMgQmE7pEMAABMwSFDdpKBc6JNAACAyVEZAACYAm0C90gGAACmwGoC92gTAABgclQGAACm4Php8+b8hopkAABgCnYvVxN4c259RzIAADAFuyEv31pYc7HUN8wZAACgFkyePFkWi8Vl69Chg/N4SUmJUlNT1bRpU4WFhWngwIHKz893uUZOTo6Sk5MVGhqqqKgojR07VhUVFS5j1q9fr65duyooKEht27ZVWlqax7GSDAAATMFRA5unfve73+nw4cPObePGjc5jo0aN0ooVK/TOO+8oPT1dubm5GjBggPO43W5XcnKyysrKtGnTJi1evFhpaWmaNGmSc8y+ffuUnJysXr16KSsrSyNHjtTw4cO1evVqj+KkTQAAMAWHLLLL4tX5klRYWOiyPygoSEFBQec8JyAgQDExMWftLygo0GuvvaZly5bphhtukCQtWrRIHTt21ObNm9WjRw999NFH+vbbb/Xxxx8rOjpaXbp00bRp0zR+/HhNnjxZVqtVCxYsUOvWrfXiiy9Kkjp27KiNGzdq1qxZSkpKqva9URkAAMAD8fHxioiIcG7Tp093O/b7779XbGys2rRpo0GDBiknJ0eSlJmZqfLycvXu3ds5tkOHDmrRooUyMjIkSRkZGerUqZOio6OdY5KSklRYWKgdO3Y4x/zyGlVjqq5RXVQGAACm4DAqN2/Ol6SDBw/KZrM597urCnTv3l1paWlq3769Dh8+rClTpujaa6/VN998o7y8PFmtVjVu3NjlnOjoaOXl5UmS8vLyXBKBquNVx35tTGFhoc6cOaOQkJBq3RvJAADAFOxetgmqzrXZbC7JgDt9+/Z1/u/LL79c3bt3V8uWLbV8+fJqf0nXFdoEAADUgcaNG6tdu3bavXu3YmJiVFZWppMnT7qMyc/Pd84xiImJOWt1QdXPvzXGZrN5lHCQDAAATKGqMuDN5o2ioiLt2bNHzZs3V7du3RQYGKi1a9c6j2dnZysnJ0eJiYmSpMTERG3fvl1HjhxxjlmzZo1sNpsSEhKcY355jaoxVdeoLpIBAIApOAyL15snxowZo/T0dO3fv1+bNm3SbbfdJn9/f911112KiIjQsGHDNHr0aH3yySfKzMzU0KFDlZiYqB49ekiS+vTpo4SEBA0ZMkRfffWVVq9erSeffFKpqanOeQoPPPCA9u7dq3HjxmnXrl16+eWXtXz5co0aNcqjWJkzAABALTh06JDuuusuHTt2TM2aNdMf/vAHbd68Wc2aNZMkzZo1S35+fho4cKBKS0uVlJSkl19+2Xm+v7+/Vq5cqQcffFCJiYlq1KiRUlJSNHXqVOeY1q1ba9WqVRo1apTmzJmjuLg4LVy40KNlhZJkMYwL952MhYWFioiI0I2NhyjAYvV1OECtKO3a1tchALWmoqJEG9OnqKCgoFqT8s5H1XdF+jcXKyz8/AviRaccuu6yH2o1Vl+hMgAAMAW7/GT3ojtur8FY6huSAQCAKRjn0ff/7/MbKiYQAgBgclQGAACmUFMPHWqISAYAAKZgN/xkN7yYM3DBTrf/bbQJAAAwOSoDAABTcMgihxd/AzvUcEsDJAMAAFNgzoB7tAkAADA5KgMAAFPwfgIhbQIAAC5olXMGzr/U78259R1tAgAATI7KAADAFBxevpuA1QQAAFzgmDPgHskAAMAUHPLjOQNuMGcAAACTozIAADAFu2GR3YvXEHtzbn1HMgAAMAW7lxMI7bQJAABAQ0VlAABgCg7DTw4vVhM4WE0AAMCFjTaBe7QJAAAwOSoDAABTcMi7FQGOmgul3iEZAACYgvcPHWq4xfSGe2cAAKBaqAwAAEzB+3cTNNy/n0kGAACm4JBFDnkzZ4AnEAIAcEGjMuBew70zAABQLVQGAACm4P1Dhxru388kAwAAU3AYFjm8ec5AA35rYcNNcwAAQLVQGQAAmILDyzZBQ37oEMkAAMAUvH9rYcNNBhrunQEAgGqhMgAAMAW7LLJ78eAgb86t70gGAACmQJvAvYZ7ZwAAoFqoDAAATMEu70r99poLpd4hGQAAmAJtAvdIBgAApsCLitxruHcGAACqhcoAAMAUDFnk8GLOgMHSQgAALmy0CdxruHcGAACqhcoAAMAUeIWxeyQDAABTsHv51kJvzq3vGu6dAQCAaqEyAAAwBdoE7pEMAABMwSE/ObwoiHtzbn3XcO8MAABUC5UBAIAp2A2L7F6U+r05t74jGQAAmAJzBtwjGQAAmILh5VsLDZ5ACAAAGioqAwAAU7DLIrsXLxvy5tz6jmQAAGAKDsO7vr/DqMFg6hnaBAAAmByVAZMblLpfg1JzXPYd3Bui//enqyRJTS4q07Axe9XlmhMKDbXr0P5Qvf1KvD5b0+ysawUEOjTr7S91SYdijRjQVXt3hdXJPQC/1Kl9nv6cvF2Xtv5RFzU5o0mzbtRnmS2dx+8ZsE29euxTs8hiVdj99N2+pnr9nW7atSfK5TrduxzUkP5fqk2LEyor99fXO2M0aXZvlzFJ136v2/t+o7iYQhWfCdSGz1tp7uJr6uQ+4TmHlxMIvTm3viMZgPZ/H6onhl3u/Nle8XMZ7bHpu9Qo3K6pqb9T4YlAXZ98RBNm7tSjd4Ro707XL/thY/bq+BGrLulQXGexA/8tJKhce3Ii9cGGSzV15Lqzjh86HKG/Lu6hw0fCZbXadXvfHfq/8at1z2O3q+BUiCTp2qv2a/SwjXpt+ZXK+ra5/P0cahV/wuU6t/f9Rv/T9xu98uZV2rmnmUKCKhTdrKhO7hHnxyGLHF70/b05t76rF2nOvHnz1KpVKwUHB6t79+76/PPPfR2SqdjtFp340ercCk8GOo91vKJQK5bG6rvtNuUdCtFbr7RU8akAXZpwyuUaV157XFdcc0ILn29T1+EDLj7/Ol6L/tFNn21tdc7j6zIu0bYdF+vwUZsO/NBE85derbDQcrVpUfll7+fnUOqQzfrbm1dr5boOOpQXoQO5TZS+5eff7bDQUg29PVPPvdJT6zIu0eEjNu09GKmMbS3q4haBGufzysDbb7+t0aNHa8GCBerevbtmz56tpKQkZWdnKyoq6rcvAK9d3OKM3li/WWWlftr1VbjSZrXW0cPBkqSdX9rUs+9Rfb4hUsWFAbr2pqOyWh36+ovGzvMbNy3TI1O+07SHf6fSM/4+ugvAcwH+diX3ylZRsVV7DkRKki5tdUzNIk/LYUgLnn5PkY3PaM+BSL3y5tXaf6iJJKlbpx/kZ5EualKs1//vnwoNKdeO76O0YOnVOnqc9lh9xRMI3fN5ZWDmzJm6//77NXToUCUkJGjBggUKDQ3V66+/7uvQTCH7a5tmPtFeE/9ymeZNbavoi0v1/BtfKSS0QpI0fXSC/AMMLc/I0L+zNurhyd9r2iMJOpwT8tMVDI1+Nlvvv91c3+8I992NAB7o0SVHKxcu0QeLFuv2m3Zo3P8lqbCoMgGOjaqseqUM+FJL/91FT7zwR50qDtLMJ95XeKNSSVLzqFOy+Bm6+9av9fLfu2vKnBsU3qhUMyasVoC/3Wf3hV9XNWfAm62h8umdlZWVKTMzU717/zwpx8/PT71791ZGRsZZ40tLS1VYWOiywTtbP43UxtXNtP+7MG37LFJPPXCZGoVX6NqbjkqShjyyX2G2Cj1+Xyc9escVendxnB6fuVOtLq2cF3Dr4FyFhNq1/FXKo7hwZO1srr880V+PTPmTvvj6Yk0c8Yka285IkiyWyvVjS//dWZ9+0Urf779Iz//tWhmGRdd13ydJ8rNIgQEOvbSkh7Zuj9POPVF6Zt71ujimUF0SDvvsvoDz5dNk4Mcff5Tdbld0dLTL/ujoaOXl5Z01fvr06YqIiHBu8fHxdRWqaRSfCtAP+0MU27JEMfFndOugXM16sp2+2txE+7LDtOzllvp+R7j+dHeuJKlz95Pq0KVQ/876VCu+3qDXPqyc7zFn+TaNfnaXL28FcKukNFC5+Tbt3BOlFxZeK7vDT32v+06SdPxkqCTpwA+NnePLK/x1+EiYoppWThA8drKyMnYg9+cxBadCVHgqSFFNmUBbXzlkcb6f4Lw2LyYQPvfcc7JYLBo5cqRzX0lJiVJTU9W0aVOFhYVp4MCBys/PdzkvJydHycnJCg0NVVRUlMaOHauKigqXMevXr1fXrl0VFBSktm3bKi0tzeP4Lqiax+OPP66CggLndvDgQV+H1OAEh9rVvEWJjh+1KjjYIUkyHK7/D+CwW5x/PS149hKNuK2bRgyo3CY9cJkk6bnHOmrxnNZ1GzxwnvwshgIDK8v73+1vqrIyf8U3/7ny6O/vUEyzIuX/WDkfYMd3lX/AxDcvcI4Jb1QqW3ipcwzqH+On1QTnuxnnmQx88cUXeuWVV3T55Ze77B81apRWrFihd955R+np6crNzdWAAQOcx+12u5KTk1VWVqZNmzZp8eLFSktL06RJk5xj9u3bp+TkZPXq1UtZWVkaOXKkhg8frtWrV3sUo08nEF500UXy9/c/KxPKz89XTEzMWeODgoIUFBRUV+GZwrCxe7Xlk0gdyQ1W06hSDR5xQA67RetXNausEhwI1sOTv9PC59uo8GSgEm88piuuOaHJD1V+6VdNNKxy5nTlBMLDB0N0LJ//W6HuBQeV6+Lon7/IY5qd0iUtjulUcZAKi4I0qN9X2pTZQsdOhioivET9/rhTFzU5rfQtlcnr6TNWrVjXXikDt+nIsUbKPxamPydvlyTnmEN5EfpsawulDt6sma//XqfPWDX8jq06mBuhrJ3N6/6mUS2+eGthUVGRBg0apFdffVVPP/20c39BQYFee+01LVu2TDfccIMkadGiRerYsaM2b96sHj166KOPPtK3336rjz/+WNHR0erSpYumTZum8ePHa/LkybJarVqwYIFat26tF198UZLUsWNHbdy4UbNmzVJSUlK14/RpMmC1WtWtWzetXbtW/fv3lyQ5HA6tXbtWI0aM8GVopnFRdKnGv7BLtsblKjgeqB3bIjTqri4qPGGVJD31QCcNHbVPT83boZBQu3JzQjTz8fbauiHSx5ED59a+zY+a+cQHzp8fGlzZulq9oa1mLbpG8c0LNPnRdbKFl6iwKEjZe5tp5NM368APTZznvPLm1bLb/fT4g+myWu3atbuZHnu2r4pO/5zgPvdKTz00aIueHbNGhsOir3bFaMKMPrLbL6iCK87Df89X+7U/VFNTU5WcnKzevXu7JAOZmZkqLy93mTPXoUMHtWjRQhkZGerRo4cyMjLUqVMnl1Z6UlKSHnzwQe3YsUNXXHGFMjIyXK5RNeaX7Yjq8PnSwtGjRyslJUVXXnmlrr76as2ePVvFxcUaOnSor0Mzhf8b0/FXj+ceCNEzIxOqfb0jucG6OaGnt2EB5+2rnc114+D73B6fPOfG37yG3e6nV968Wq+8ebXbMafPWPXCwmv1wsJrzytO1L2aegLhf89Xe+qppzR58uSzxr/11lvatm2bvvjii7OO5eXlyWq1qnHjxi77fzlnLi8v75xz6qqO/dqYwsJCnTlzRiEhIaoOnycDf/7zn3X06FFNmjRJeXl56tKliz788MOzbg4AAG/UVJvg4MGDstlszv3nqgocPHhQjz76qNasWaPg4OCzjtc39aKeNWLECB04cEClpaXasmWLunfv7uuQAAA4J5vN5rKdKxnIzMzUkSNH1LVrVwUEBCggIEDp6emaO3euAgICFB0drbKyMp08edLlvF/OmYuJiTnnnLqqY782xmazVbsqINWTZAAAgNrmzUoCT99rcOONN2r79u3KyspybldeeaUGDRrk/N+BgYFau3at85zs7Gzl5OQoMTFRkpSYmKjt27fryJEjzjFr1qyRzWZTQkKCc8wvr1E1puoa1eXzNgEAAHWhLlcThIeH67LLLnPZ16hRIzVt2tS5f9iwYRo9erQiIyNls9n08MMPKzExUT169JAk9enTRwkJCRoyZIhmzJihvLw8Pfnkk0pNTXVWIx544AG99NJLGjdunO677z6tW7dOy5cv16pVqzy6N5IBAAB8YNasWfLz89PAgQNVWlqqpKQkvfzyy87j/v7+WrlypR588EElJiaqUaNGSklJ0dSpU51jWrdurVWrVmnUqFGaM2eO4uLitHDhQo+WFUqSxTAMo8burI4VFhYqIiJCNzYeogCL1dfhALWitGtbX4cA1JqKihJtTJ+igoICl0l5Nanqu6Lvh/crsNH5f1eUF5fpg5terdVYfYXKAADAFHzx0KELBRMIAQAwOSoDAABToDLgHskAAMAUDMmrNw9esBPsqoFkAABgClQG3GPOAAAAJkdlAABgClQG3CMZAACYAsmAe7QJAAAwOSoDAABToDLgHskAAMAUDMMiw4svdG/Ore9oEwAAYHJUBgAApuCQxauHDnlzbn1HMgAAMAXmDLhHmwAAAJOjMgAAMAUmELpHMgAAMAXaBO6RDAAATIHKgHvMGQAAwOSoDAAATMHwsk3QkCsDJAMAAFMwJBmGd+c3VLQJAAAwOSoDAABTcMgiC08gPCeSAQCAKbCawD3aBAAAmByVAQCAKTgMiyw8dOicSAYAAKZgGF6uJmjAywloEwAAYHJUBgAApsAEQvdIBgAApkAy4B7JAADAFJhA6B5zBgAAMDkqAwAAU2A1gXskAwAAU6hMBryZM1CDwdQztAkAADA5KgMAAFNgNYF7JAMAAFMwftq8Ob+hok0AAIDJURkAAJgCbQL3SAYAAOZAn8AtkgEAgDl4WRlQA64MMGcAAACTozIAADAFnkDoHskAAMAUmEDoHm0CAABMjsoAAMAcDIt3kwAbcGWAZAAAYArMGXCPNgEAACZHZQAAYA48dMgtkgEAgCmwmsC9aiUD//nPf6p9wVtvvfW8gwEAAHWvWslA//79q3Uxi8Uiu93uTTwAANSeBlzq90a1kgGHw1HbcQAAUKtoE7jn1WqCkpKSmooDAIDaZdTA1kB5nAzY7XZNmzZNF198scLCwrR3715J0sSJE/Xaa6/VeIAAAKB2eZwMPPPMM0pLS9OMGTNktVqd+y+77DItXLiwRoMDAKDmWGpga5g8TgaWLFmiv/3tbxo0aJD8/f2d+zt37qxdu3bVaHAAANQY2gRueZwM/PDDD2rbtu1Z+x0Oh8rLy2skKAAAUHc8TgYSEhL06aefnrX/H//4h6644ooaCQoAgBpHZcAtj59AOGnSJKWkpOiHH36Qw+HQv/71L2VnZ2vJkiVauXJlbcQIAID3eGuhWx5XBvr166cVK1bo448/VqNGjTRp0iTt3LlTK1as0B//+MfaiBEAANSi83o3wbXXXqs1a9bUdCwAANQaXmHs3nm/qGjr1q3auXOnpMp5BN26dauxoAAAqHG8tdAtj5OBQ4cO6a677tJnn32mxo0bS5JOnjypa665Rm+99Zbi4uJqOkYAAFCLPJ4zMHz4cJWXl2vnzp06fvy4jh8/rp07d8rhcGj48OG1ESMAAN6rmkDozeaB+fPn6/LLL5fNZpPNZlNiYqI++OAD5/GSkhKlpqaqadOmCgsL08CBA5Wfn+9yjZycHCUnJys0NFRRUVEaO3asKioqXMasX79eXbt2VVBQkNq2bau0tDSP/9N4nAykp6dr/vz5at++vXNf+/bt9de//lUbNmzwOAAAAOqCxfB+80RcXJyee+45ZWZmauvWrbrhhhvUr18/7dixQ5I0atQorVixQu+8847S09OVm5urAQMGOM+32+1KTk5WWVmZNm3apMWLFystLU2TJk1yjtm3b5+Sk5PVq1cvZWVlaeTIkRo+fLhWr17tUawetwni4+PP+XAhu92u2NhYTy8HAEDdqOM5A7fccovLz88884zmz5+vzZs3Ky4uTq+99pqWLVumG264QZK0aNEidezYUZs3b1aPHj300Ucf6dtvv9XHH3+s6OhodenSRdOmTdP48eM1efJkWa1WLViwQK1bt9aLL74oSerYsaM2btyoWbNmKSkpqdqxelwZeP755/Xwww9r69atzn1bt27Vo48+qhdeeMHTywEAcEEpLCx02UpLS3/zHLvdrrfeekvFxcVKTExUZmamysvL1bt3b+eYDh06qEWLFsrIyJAkZWRkqFOnToqOjnaOSUpKUmFhobO6kJGR4XKNqjFV16iualUGmjRpIovl515JcXGxunfvroCAytMrKioUEBCg++67T/379/coAAAA6kQNPXQoPj7eZfdTTz2lyZMnn/OU7du3KzExUSUlJQoLC9O7776rhIQEZWVlyWq1OifiV4mOjlZeXp4kKS8vzyURqDpedezXxhQWFurMmTMKCQmp1q1VKxmYPXt2tS4GAEC9VUNtgoMHD8pmszl3BwUFuT2lffv2ysrKUkFBgf7xj38oJSVF6enpXgRRO6qVDKSkpNR2HAAAXBCqVgdUh9Vqdb7cr1u3bvriiy80Z84c/fnPf1ZZWZlOnjzpUh3Iz89XTEyMJCkmJkaff/65y/WqVhv8csx/r0DIz8+XzWardlVAOo85A79UUlJyVu8EAIB6qR68qMjhcKi0tFTdunVTYGCg1q5d6zyWnZ2tnJwcJSYmSpISExO1fft2HTlyxDlmzZo1stlsSkhIcI755TWqxlRdo7o8Xk1QXFys8ePHa/ny5Tp27NhZx+12u6eXBACg9tXxaoLHH39cffv2VYsWLXTq1CktW7ZM69ev1+rVqxUREaFhw4Zp9OjRioyMlM1m08MPP6zExET16NFDktSnTx8lJCRoyJAhmjFjhvLy8vTkk08qNTXV2Zp44IEH9NJLL2ncuHG67777tG7dOi1fvlyrVq3yKFaPKwPjxo3TunXrNH/+fAUFBWnhwoWaMmWKYmNjtWTJEk8vBwBAg3TkyBHdc889at++vW688UZ98cUXWr16tfOlfrNmzdKf/vQnDRw4UD179lRMTIz+9a9/Oc/39/fXypUr5e/vr8TERA0ePFj33HOPpk6d6hzTunVrrVq1SmvWrFHnzp314osvauHChR4tK5Qki2F49uqFFi1aaMmSJbr++utls9m0bds2tW3bVm+88YbefPNNvf/++x4F4I3CwkJFREToxsZDFGCx1tnnAnWptGtbX4cA1JqKihJtTJ+igoKCavfhPVX1XRH//NPyCwk+7+s4zpTo4NgnazVWX/G4MnD8+HG1adNGUuUkiuPHj0uS/vCHP/AEQgBAvVXXTyC8kHicDLRp00b79u2TVPmAhOXLl0uSVqxYcdZ6SQAAUP95nAwMHTpUX331lSRpwoQJmjdvnoKDgzVq1CiNHTu2xgMEAKBG1IPVBPWVx6sJRo0a5fzfvXv31q5du5SZmam2bdvq8ssvr9HgAABA7fM4GfhvLVu2VMuWLWsiFgAAao1F3vX9vXiQcb1XrWRg7ty51b7gI488ct7BAACAuletZGDWrFnVupjFYvFJMmA/WSiLJbDOPxeoC2v//pqvQwBqTeEph5q0q6MPq6EXFTVE1UoGqlYPAABwwarjJxBeSLx6NwEAALjweT2BEACACwKVAbdIBgAApuDtUwR5AiEAAGiwqAwAAMyBNoFb51UZ+PTTTzV48GAlJibqhx9+kCS98cYb2rhxY40GBwBAjeFxxG55nAz885//VFJSkkJCQvTll1+qtLRUklRQUKBnn322xgMEAAC1y+Nk4Omnn9aCBQv06quvKjDw5wf9/P73v9e2bdtqNDgAAGoKrzB2z+M5A9nZ2erZs+dZ+yMiInTy5MmaiAkAgJrHEwjd8rgyEBMTo927d5+1f+PGjWrTpk2NBAUAQI1jzoBbHicD999/vx599FFt2bJFFotFubm5Wrp0qcaMGaMHH3ywNmIEAAC1yOM2wYQJE+RwOHTjjTfq9OnT6tmzp4KCgjRmzBg9/PDDtREjAABe46FD7nmcDFgsFj3xxBMaO3asdu/eraKiIiUkJCgsLKw24gMAoGbwnAG3zvuhQ1arVQkJCTUZCwAA8AGPk4FevXrJYnE/o3LdunVeBQQAQK3wdnkglYGfdenSxeXn8vJyZWVl6ZtvvlFKSkpNxQUAQM2iTeCWx8nArFmzzrl/8uTJKioq8jogAABQt2rsrYWDBw/W66+/XlOXAwCgZvGcAbdq7K2FGRkZCg4OrqnLAQBQo1ha6J7HycCAAQNcfjYMQ4cPH9bWrVs1ceLEGgsMAADUDY+TgYiICJef/fz81L59e02dOlV9+vSpscAAAEDd8CgZsNvtGjp0qDp16qQmTZrUVkwAANQ8VhO45dEEQn9/f/Xp04e3EwIALji8wtg9j1cTXHbZZdq7d29txAIAAHzA42Tg6aef1pgxY7Ry5UodPnxYhYWFLhsAAPUWywrPqdpzBqZOnarHHntMN998syTp1ltvdXkssWEYslgsstvtNR8lAADeYs6AW9VOBqZMmaIHHnhAn3zySW3GAwAA6li1kwHDqEyJrrvuuloLBgCA2sJDh9zzaGnhr72tEACAeo02gVseJQPt2rX7zYTg+PHjXgUEAADqlkfJwJQpU856AiEAABcC2gTueZQM3HnnnYqKiqqtWAAAqD20Cdyq9nMGmC8AAEDD5PFqAgAALkhUBtyqdjLgcDhqMw4AAGoVcwbc8/gVxgAAXJCoDLjl8bsJAABAw0JlAABgDlQG3CIZAACYAnMG3KNNAACAyVEZAACYA20Ct0gGAACmQJvAPdoEAACYHJUBAIA50CZwi2QAAGAOJANu0SYAAMDkqAwAAEzB8tPmzfkNFckAAMAcaBO4RTIAADAFlha6x5wBAABMjsoAAMAcaBO4RTIAADCPBvyF7g3aBAAAmByVAQCAKTCB0D2SAQCAOTBnwC3aBAAAmByVAQCAKdAmcI/KAADAHIwa2Dwwffp0XXXVVQoPD1dUVJT69++v7OxslzElJSVKTU1V06ZNFRYWpoEDByo/P99lTE5OjpKTkxUaGqqoqCiNHTtWFRUVLmPWr1+vrl27KigoSG3btlVaWppHsZIMAABQC9LT05WamqrNmzdrzZo1Ki8vV58+fVRcXOwcM2rUKK1YsULvvPOO0tPTlZubqwEDBjiP2+12JScnq6ysTJs2bdLixYuVlpamSZMmOcfs27dPycnJ6tWrl7KysjRy5EgNHz5cq1evrnasFsMwLtjCR2FhoSIiInS9+inAEujrcIBasTo3y9chALWm8JRDTdrtVUFBgWw2W+18xk/fFZff96z8rcHnfR17WYm+fv1/dfDgQZdYg4KCFBQU9JvnHz16VFFRUUpPT1fPnj1VUFCgZs2aadmyZbr99tslSbt27VLHjh2VkZGhHj166IMPPtCf/vQn5ebmKjo6WpK0YMECjR8/XkePHpXVatX48eO1atUqffPNN87PuvPOO3Xy5El9+OGH1bo3KgMAAHOooTZBfHy8IiIinNv06dOr9fEFBQWSpMjISElSZmamysvL1bt3b+eYDh06qEWLFsrIyJAkZWRkqFOnTs5EQJKSkpJUWFioHTt2OMf88hpVY6quUR1MIAQAmEMNLS08V2XgtzgcDo0cOVK///3vddlll0mS8vLyZLVa1bhxY5ex0dHRysvLc475ZSJQdbzq2K+NKSws1JkzZxQSEvKb8ZEMAADgAZvN5nFLIzU1Vd988402btxYS1F5hzYBAMAUqpYWerOdjxEjRmjlypX65JNPFBcX59wfExOjsrIynTx50mV8fn6+YmJinGP+e3VB1c+/NcZms1WrKiCRDAAAzKKOlxYahqERI0bo3Xff1bp169S6dWuX4926dVNgYKDWrl3r3Jedna2cnBwlJiZKkhITE7V9+3YdOXLEOWbNmjWy2WxKSEhwjvnlNarGVF2jOmgTAABQC1JTU7Vs2TL9+9//Vnh4uLPHHxERoZCQEEVERGjYsGEaPXq0IiMjZbPZ9PDDDysxMVE9evSQJPXp00cJCQkaMmSIZsyYoby8PD355JNKTU11zlV44IEH9NJLL2ncuHG67777tG7dOi1fvlyrVq2qdqwkAwAAU7AYhixerKb39Nz58+dLkq6//nqX/YsWLdK9994rSZo1a5b8/Pw0cOBAlZaWKikpSS+//LJzrL+/v1auXKkHH3xQiYmJatSokVJSUjR16lTnmNatW2vVqlUaNWqU5syZo7i4OC1cuFBJSUnVjpVkAABgDnX8oqLqPMYnODhY8+bN07x589yOadmypd5///1fvc7111+vL7/80rMAf4E5AwAAmByVAQCAKfCiIvdIBgAA5lDHbYILCW0CAABMjsoAAMAUaBO4RzIAADAH2gRukQwAAEyByoB7zBkAAMDkqAwAAMyBNoFbJAMAANNoyKV+b9AmAADA5KgMAADMwTAqN2/Ob6BIBgAApsBqAvdoEwAAYHJUBgAA5sBqArdIBgAApmBxVG7enN9Q0SYAAMDkqAyY3J/u+VHJ9xxTdHyZJOlAdrCWzorW1k9sio4r05LPd57zvKf/0lKfrmys8CYVmvBSjlp3PKPwJnYVHAtQxmqbFk1vrtNF/nV5K4Ak6Y0XYvT3mTEu++IuKdFrn+5S4Ql/vfFCjLalh+tIrlURkRW65qYCpYw7rEa2n//sy84K0evPxur7r0NlsRhq3+W0hj2Zq0t+VyJJOrg7SHMnxCnnu2AVn/JX0+hy9brthAaPzlNAYJ3eLjxBm8AtkgGTO3o4UK8/21w/7AuSxSL98X+Oa/Ki/Urt004Hdwfpzs4JLuNvHnxMtz94VF+sC5ckGQ4pY7VNaf8Xo4JjAYptXaoRz/6g8MaH9FxqS1/cEqCW7c/oubf3OH/296/8V/x4fqCO5Qfq/km5atGuREcOWTV3QpyO5Qdq4qv7JUlniv30xKBL1OOPBRrx7CHZ7Ra98UKMnrj7Ev196w4FBEoBgYZ6335CbTudVliEXXt3hGj22Hg5HBbd9/hhX9wyqoHVBO75NBnYsGGDnn/+eWVmZurw4cN699131b9/f1+GZDpb1kS4/Jz2f831p3uOqUO3Yh34Llgnjrr+mXNN3wJtWNFYJacr/+ovKgjQyiUXOY8f+cGqFYub6n8ePFr7wQNu+PtLkVEVZ+1v1aFEkxbud/4c26pM944/rBkPt5S9QvIPqPyr/9SJAN0zNk9RF5dLkgaPztMDN3ZQ/iGrLm5dpuYty9S85XHndaLjyvV1xgl9s6VRrd8bvMBzBtzy6ZyB4uJide7cWfPmzfNlGPiJn5+h6/qdUFCoQzu3nv2PWttOp9X2shKtfjPS7TUio8v1+74F+jqDfxThOz/ss+quK36nlB4d9VxqCx055L52X1zor9Awh/x/+tMo7pJS2ZpUaPWbTVVeZlHpGYs+fLOpWlxaopif2mnn+rytn9h0eWJRbdwOUOt8Whno27ev+vbtW+3xpaWlKi0tdf5cWFhYG2GZTqsOZzR7xW5Zgxw6U+ynqcNaKef74LPG3XTXcR34LkjfniNRmPDyASUmFSg4xFDGRzbNGhNfF6EDZ+nQtVhjZp9R3CWlOn4kUH9/MUaP3XapXvlkl0LDXKeDFxzz17LZMeo7+EfnvtAwh57/525Nvq+1ls2OliTFti7Vs2/ucSYMVUbecql2fxOi8lI/3Tz4R90zNq/W7w/njzaBexfUaoLp06crIiLCucXH84VTEw7tCdJDf2ynR5Iv1colF2nMnBy1uLTEZYw12KFet51wWxV45alYjUhqp6fubaXYlqX6f0/l1kXowFmuuuGUet5SoDYJJbry+lN6+u97VVTorw3/aewyrviUnybe00Yt2pVoyGM/f4mXnrFo5mPx+t1VxZq98jvN/Pf3atWhRBOHtFHpGYvLNf53wX7NW52tCfP26/O1Nv1jflRd3CLOl1EDWwN1QSUDjz/+uAoKCpzbwYMHfR1Sg1BR7qfc/UHavT1Ui6Y3175vQ9R/uGvP/9rkkwoKMfTxO+dOBk4cDdTB3cHa/FGE5oyP0y33HlNkVHldhA/8qrAIu+LalCp3f5Bz3+kiPz1x9yUKaeTQU6/tc1kB8Mm7TZR/0KrHZuWofZcz6tjttCbMO6C8HKsyVrvOsYm6uFwt25Wq120ndd//HtbfX4yR3V5XdwbUnAtqNUFQUJCCgoJ+eyC8YrFIgVbXFDjpruPa/JFNBcd/+1fG8tMfT/99DcAXzhT7KfeAVTcOrExOi09VJgKBVkNT0vbKGuz6e1p6xk9+fj//HkuV82ksFsnxKw+dcTikigqLDIckVtXWS7QJ3LugkgHUvKGPH9YX68J19AerQsLs6nXbSV1+TZGeuLuNc0xsq1J16lGsiYNbn3X+VTcUqkmzCmVnhaik2F8t25do+MRcffN5qPIPWevyVgBJ0t+mxKpHnwJFxZXrWF6A3nihufz9pOtvO6HiU37637suUekZP4376z6dLvLX6Z/m/EU0rZC/v3RFz1N69elYvfS/cep331E5HBYtfylK/gFS599XDl73rybyDzDUuuMZBVoNffdVZVXtultP8JyB+ozVBG6RDJhc44sqNHZujiKjKnT6lL/27QzWE3e30bYN4c4xSXce14+HA5WZHn7W+WUlfuo76Jj+3+QSBVoNHc0N1GcfROjtl6Lr8jYApx8PB2r6Q6106oS/IppWOHv/jZva9dWmMO3aVjkBdug1rs/QWLzlW8XEl6nFpaWakrZXS2fGaOQt7WTxM9T2sjN6ZukeNY2uXK7o529o+bwo/bA3SIYhRcWV6dahP2rA/SypxYXJYhi+S3WKioq0e/duSdIVV1yhmTNnqlevXoqMjFSLFi1+8/zCwkJFREToevVTgIV0HA3T6twsX4cA1JrCUw41abdXBQUFstlstfMZP31XJPadqoDAs1dKVVdFeYkyPphUq7H6ik8rA1u3blWvXr2cP48ePVqSlJKSorS0NB9FBQBokHgcsVs+TQauv/56+bAwAQAAxJwBAIBJsJrAPZIBAIA5OIzKzZvzGyiSAQCAOTBnwK0L6gmEAACg5lEZAACYgkVezhmosUjqH5IBAIA58ARCt2gTAABgclQGAACmwNJC90gGAADmwGoCt2gTAABgclQGAACmYDEMWbyYBOjNufUdyQAAwBwcP23enN9A0SYAAMDkqAwAAEyBNoF7JAMAAHNgNYFbJAMAAHPgCYRuMWcAAACTozIAADAFnkDoHskAAMAcaBO4RZsAAACTozIAADAFi6Ny8+b8hopkAABgDrQJ3KJNAACAyVEZAACYAw8dcotkAABgCjyO2D3aBAAAmByVAQCAOTCB0C2SAQCAORiSvFke2HBzAZIBAIA5MGfAPeYMAABgclQGAADmYMjLOQM1Fkm9QzIAADAHJhC6RZsAAACTozIAADAHhySLl+c3UFQGAACmULWawJvNExs2bNAtt9yi2NhYWSwWvffeey7HDcPQpEmT1Lx5c4WEhKh37976/vvvXcYcP35cgwYNks1mU+PGjTVs2DAVFRW5jPn666917bXXKjg4WPHx8ZoxY4bH/21IBgAAqAXFxcXq3Lmz5s2bd87jM2bM0Ny5c7VgwQJt2bJFjRo1UlJSkkpKSpxjBg0apB07dmjNmjVauXKlNmzYoL/85S/O44WFherTp49atmypzMxMPf/885o8ebL+9re/eRQrbQIAgDnU8QTCvn37qm/fvm4uZWj27Nl68skn1a9fP0nSkiVLFB0drffee0933nmndu7cqQ8//FBffPGFrrzySknSX//6V91888164YUXFBsbq6VLl6qsrEyvv/66rFarfve73ykrK0szZ850SRp+C5UBAIA5VCUD3myq/Gv8l1tpaanHoezbt095eXnq3bu3c19ERIS6d++ujIwMSVJGRoYaN27sTAQkqXfv3vLz89OWLVucY3r27Cmr1eock5SUpOzsbJ04caLa8ZAMAADggfj4eEVERDi36dOne3yNvLw8SVJ0dLTL/ujoaOexvLw8RUVFuRwPCAhQZGSky5hzXeOXn1EdtAkAAOZQQ22CgwcPymazOXcHBQV5G5nPURkAAJiDowY2STabzWU7n2QgJiZGkpSfn++yPz8/33ksJiZGR44ccTleUVGh48ePu4w51zV++RnVQTIAADCFul5a+Gtat26tmJgYrV271rmvsLBQW7ZsUWJioiQpMTFRJ0+eVGZmpnPMunXr5HA41L17d+eYDRs2qLy83DlmzZo1at++vZo0aVLteEgGAACoBUVFRcrKylJWVpakykmDWVlZysnJkcVi0ciRI/X000/rP//5j7Zv36577rlHsbGx6t+/vySpY8eOuummm3T//ffr888/12effaYRI0bozjvvVGxsrCTp7rvvltVq1bBhw7Rjxw69/fbbmjNnjkaPHu1RrMwZAACYQx0vLdy6dat69erl/LnqCzolJUVpaWkaN26ciouL9Ze//EUnT57UH/7wB3344YcKDg52nrN06VKNGDFCN954o/z8/DRw4EDNnTvXeTwiIkIfffSRUlNT1a1bN1100UWaNGmSR8sKJcliGBfumxcKCwsVERGh69VPAZZAX4cD1IrVuVm+DgGoNYWnHGrSbq8KCgpcJuXV6Gf89F3R+5KRCvA//8l+FfZSfbxndq3G6iu0CQAAMDnaBAAAc+AVxm6RDAAATMLLZEANNxmgTQAAgMlRGQAAmANtArdIBgAA5uAw5FWp39FwkwHaBAAAmByVAQCAORiOys2b8xsokgEAgDkwZ8AtkgEAgDkwZ8At5gwAAGByVAYAAOZAm8AtkgEAgDkY8jIZqLFI6h3aBAAAmByVAQCAOdAmcItkAABgDg6HJC+eFeBouM8ZoE0AAIDJURkAAJgDbQK3SAYAAOZAMuAWbQIAAEyOygAAwBx4HLFbJAMAAFMwDIcML9486M259R3JAADAHAzDu7/umTMAAAAaKioDAABzMLycM9CAKwMkAwAAc3A4JIsXff8GPGeANgEAACZHZQAAYA60CdwiGQAAmILhcMjwok3QkJcW0iYAAMDkqAwAAMyBNoFbJAMAAHNwGJKFZOBcaBMAAGByVAYAAOZgGJK8ec5Aw60MkAwAAEzBcBgyvGgTGCQDAABc4AyHvKsMsLQQAAA0UFQGAACmQJvAPZIBAIA50CZw64JOBqqytAqVe/UcCaA+KzzVcP8BAgqLKn+/6+Kvbm+/KypUXnPB1DMXdDJw6tQpSdJGve/jSIDa06SdryMAat+pU6cUERFRK9e2Wq2KiYnRxjzvvytiYmJktVprIKr6xWJcwE0Qh8Oh3NxchYeHy2Kx+DocUygsLFR8fLwOHjwom83m63CAGsXvd90zDEOnTp1SbGys/Pxqb057SUmJysrKvL6O1WpVcHBwDURUv1zQlQE/Pz/FxcX5OgxTstls/GOJBovf77pVWxWBXwoODm6QX+I1haWFAACYHMkAAAAmRzIAjwQFBempp55SUFCQr0MBahy/3zCrC3oCIQAA8B6VAQAATI5kAAAAkyMZAADA5EgGAAAwOZIBVNu8efPUqlUrBQcHq3v37vr88899HRJQIzZs2KBbbrlFsbGxslgseu+993wdElCnSAZQLW+//bZGjx6tp556Stu2bVPnzp2VlJSkI0eO+Do0wGvFxcXq3Lmz5s2b5+tQAJ9gaSGqpXv37rrqqqv00ksvSap8L0R8fLwefvhhTZgwwcfRATXHYrHo3XffVf/+/X0dClBnqAzgN5WVlSkzM1O9e/d27vPz81Pv3r2VkZHhw8gAADWBZAC/6ccff5Tdbld0dLTL/ujoaOXl5fkoKgBATSEZAADA5EgG8Jsuuugi+fv7Kz8/32V/fn6+YmJifBQVAKCmkAzgN1mtVnXr1k1r16517nM4HFq7dq0SExN9GBkAoCYE+DoAXBhGjx6tlJQUXXnllbr66qs1e/ZsFRcXa+jQob4ODfBaUVGRdu/e7fx53759ysrKUmRkpFq0aOHDyIC6wdJCVNtLL72k559/Xnl5eerSpYvmzp2r7t27+zoswGvr169Xr169ztqfkpKitLS0ug8IqGMkAwAAmBxzBgAAMDmSAQAATI5kAAAAkyMZAADA5EgGAAAwOZIBAABMjmQAAACTIxkAAMDkSAYAL917773q37+/8+frr79eI0eOrPM41q9fL4vFopMnT7odY7FY9N5771X7mpMnT1aXLl28imv//v2yWCzKysry6joAag/JABqke++9VxaLRRaLRVarVW3bttXUqVNVUVFR65/9r3/9S9OmTavW2Op8gQNAbeNFRWiwbrrpJi1atEilpaV6//33lZqaqsDAQD3++ONnjS0rK5PVaq2Rz42MjKyR6wBAXaEygAYrKChIMTExatmypR588EH17t1b//nPfyT9XNp/5plnFBsbq/bt20uSDh48qDvuuEONGzdWZGSk+vXrp/379zuvabfbNXr0aDVu3FhNmzbVuHHj9N+v9/jvNkFpaanGjx+v+Ph4BQUFqW3btnrttde0f/9+58txmjRpIovFonvvvVdS5Suip0+frtatWyskJESdO3fWP/7xD5fPef/999WuXTuFhISoV69eLnFW1/jx49WuXTuFhoaqTZs2mjhxosrLy88a98orryg+Pl6hoaG64447VFBQ4HJ84cKF6tixo4KDg9WhQwe9/PLLHscCwHdIBmAaISEhKisrc/68du1aZWdna82aNVq5cqXKy8uVlJSk8PBwffrpp/rss88UFhamm266yXneiy++qLS0NL3++uvauHGjjh8/rnffffdXP/eee+7Rm2++qblz52rnzp165ZVXFBYWpvj4eP3zn/+UJGVnZ+vw4cOaM2eOJGn69OlasmSJFixYoB07dmjUqFEaPHiw0tPTJVUmLQMGDNAtt9yirKwsDR8+XBMmTPD4v0l4eLjS0tL07bffas6cOXr11Vc1a9YslzG7d+/W8uXLtWLFCn344Yf68ssv9dBDDzmPL126VJMmTdIzzzyjnTt36tlnn9XEiRO1ePFij+MB4CMG0AClpKQY/fr1MwzDMBwOh7FmzRojKCjIGDNmjPN4dHS0UVpa6jznjTfeMNq3b284HA7nvtLSUiMkJMRYvXq1YRiG0bx5c2PGjBnO4+Xl5UZcXJzzswzDMK677jrj0UcfNQzDMLKzsw1Jxpo1a84Z5yeffGJIMk6cOOHcV1JSYoSGhhqbNm1yGTts2DDjrrvuMgzDMB5//HEjISHB5fj48ePPutZ/k2S8++67bo8///zzRrdu3Zw/P/XUU4a/v79x6NAh574PPvjA8PPzMw4fPmwYhmFccsklxrJly1yuM23aNCMxMdEwDMPYt2+fIcn48ssv3X4uAN9izgAarJUrVyosLEzl5eVyOBy6++67NXnyZOfxTp06ucwT+Oqrr7R7926Fh4e7XKekpER79uxRQUGBDh8+rO7duzuPBQQE6MorrzyrVVAlKytL/v7+uu6666od9+7du3X69Gn98Y9/dNlfVlamK664QpK0c+dOlzgkKTExsdqfUeXtt9/W3LlztWfPHhUVFamiokI2m81lTIsWLXTxxRe7fI7D4VB2drbCw8O1Z88eDRs2TPfff79zTEVFhSIiIjyOB4BvkAygwerVq5fmz58vq9Wq2NhYBQS4/ro3atTI5eeioiJ169ZNS5cuPetazZo1O68YQkJCPD6nqKhIkrRq1SqXL2Gpch5ETcnIyNCgQYM0ZcoUJSUlKSIiQm+99ZZefPFFj2N99dVXz0pO/P39ayxWALWLZAANVqNGjdS2bdtqj+/atavefvttRUVFnfXXcZXmzZtry5Yt6tmzp6TKv4AzMzPVtWvXc47v1KmTHA6H0tPT1bt377OOV1Um7Ha7c19CQoKCgoKUk5PjtqLQsWNH52TIKps3b/7tm/yFTZs2qWXLlnriiSec+w4cOHDWuJycHOXm5io2Ntb5OX5+fmrfvr2io6MVGxurvXv3atCgQR59PoD6gwmEwE8GDRqkiy66SP369dOnn36qffv2af369XrkkUd06NAhSdKjjz6q5557Tu+995527dqlhx566FefEdCqVSulpKTovvvu03vvvee85vLlyyVJLVu2lMVi0cqVK3X06FEVFRUpPDxcY8aM0ahRo7R48WLt2bNH27Zt01//+lfnpLwHHnhA33//vcaOHavs7GwtW7ZMaWlpHt3vpZdeqpycHL311lvas2eP5s6de87JkMHBwUpJSdFXX32lTz/9VI888ojuuOMOxcTESJKmTJmi6dOna+7cufruu++0fft2LVq0SDNnzvQoHgC+QzIA/CQ0NFQbNmxQixYtNGDAAHXs2FHDhg1TSUmJs1Lw2GOPaciQIUpJSVFiYqLCw8N12223/ep158+fr9tvv10PPfSQOnTooPvvv1/FxcWSpIsvvlhTpkzRhAkTFB0drREjRkiSpk2bpokTJ2r69Onq2LGjbrrpJq1atUqtW7eWVNnH/+c//6n33ntPnTt31oIFC/Tss896dL+33nqrRo0apREjRqhLly7atGmTJk6ceNa4tm3basCAAbr55pvVp08fXX755S5LB4cPH66FCxdq0aJF6tSpk6677jqlpaU5YwVQ/1kMdzOfAACAKVAZAADA5EgGAAAwOZIBAABMjmQAAACTIxkAAMDkSAYAADA5kgEAAEyOZAAAAJMjGQAAwORIBgAAMDmSAQAATO7/A6MMve8cJHJIAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#found this cool visual for cm in scikit learn!\n",
    "cm = confusion_matrix(y_test, class_label_predictions)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm)\n",
    "disp.plot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "looks like we had some difficulty labeling the >50k, or it may just seem that way since there are far more examples of <=50k"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.19"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}