01_sans_machine_learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e5f48c1a-8154-4ce8-9734-8eb4501b501c",
   "metadata": {},
   "source": [
    "### Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a048dd28-c228-47a0-91d0-79795c8cf6f9",
   "metadata": {},
   "source": [
    "Before jumping into machine learning, let's try to solve a problem manually."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f484a6f-4572-4220-b200-147d8d45ef50",
   "metadata": {},
   "source": [
    "You have started working as a researcher in a botanical garden and you noticed that botanists have been struggling to tell the difference between two types of Iris Setosa and Iris Versicolor plants because they look very similar.\n",
    "\n",
    "Luckily a friend of yours has taken some measurements of both types of plants.\n",
    "\n",
    "### Exploration of the problem\n",
    "\n",
    "Let's print out your friend's dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "eedf8749-5ba2-4699-aa35-a68197f70ee7",
   "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>Petal_Length</th>\n",
       "      <th>Petal_Width</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.6</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.4</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.3</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.7</td>\n",
       "      <td>0.5</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.4</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.2</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.6</td>\n",
       "      <td>0.6</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>4.5</td>\n",
       "      <td>1.5</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>4.5</td>\n",
       "      <td>1.3</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>3.9</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>4.7</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>4.1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>4.0</td>\n",
       "      <td>1.3</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>4.4</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>3.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>5.1</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>4.4</td>\n",
       "      <td>1.3</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>4.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>4.2</td>\n",
       "      <td>1.2</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>4.1</td>\n",
       "      <td>1.3</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Petal_Length  Petal_Width            Class\n",
       "0            1.4          0.2      Iris-setosa\n",
       "1            1.5          0.2      Iris-setosa\n",
       "2            1.5          0.2      Iris-setosa\n",
       "3            1.6          0.2      Iris-setosa\n",
       "4            1.5          0.4      Iris-setosa\n",
       "5            1.5          0.3      Iris-setosa\n",
       "6            1.7          0.5      Iris-setosa\n",
       "7            1.5          0.2      Iris-setosa\n",
       "8            1.5          0.4      Iris-setosa\n",
       "9            1.2          0.2      Iris-setosa\n",
       "10           1.5          0.2      Iris-setosa\n",
       "11           1.6          0.6      Iris-setosa\n",
       "12           1.4          0.2      Iris-setosa\n",
       "13           4.5          1.5  Iris-versicolor\n",
       "14           4.5          1.3  Iris-versicolor\n",
       "15           3.9          1.4  Iris-versicolor\n",
       "16           4.7          1.4  Iris-versicolor\n",
       "17           4.1          1.0  Iris-versicolor\n",
       "18           4.0          1.3  Iris-versicolor\n",
       "19           4.4          1.4  Iris-versicolor\n",
       "20           3.5          1.0  Iris-versicolor\n",
       "21           5.1          1.6  Iris-versicolor\n",
       "22           4.4          1.3  Iris-versicolor\n",
       "23           4.6          1.4  Iris-versicolor\n",
       "24           4.2          1.2  Iris-versicolor\n",
       "25           4.1          1.3  Iris-versicolor"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use pandas (https://pandas.pydata.org/) a fast, powerful, flexible and easy to use open source data analysis and manipulation tool.\n",
    "# We only use very basic pandas functionality in this course so you don't need to worry too much about what it is doing.\n",
    "# At the end of this course, we provide a list of resources for learning some of the third party tools used in this course.\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "# load the dataset\n",
    "iris = pd.read_csv(\"data/iris_subset.csv\")\n",
    "\n",
    "# print out the dataset\n",
    "iris"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d4caf7a-938f-419a-98ba-f8a8f17ee63c",
   "metadata": {},
   "source": [
    "Your goal was to come up with an API that allows users to enter the petal length and petal width and it will predict the Class of Iris.\n",
    "\n",
    "Let's plot the dataset from your friend to look for any interesting patterns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "faaaf8ab-22ca-4d66-9b70-90a7ef1f9a92",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# use seaborn (https://seaborn.pydata.org/) a library for drawing attractive and informative statistical graphics\n",
    "# We only use very basic seaborn functionality in this course so you don't need to worry too much about what it is doing.\n",
    "# At the end of this course, we provide a list of resources for learning some of the third party tools used in this course.\n",
    "\n",
    "import seaborn as sns\n",
    "\n",
    "# increase the plot size\n",
    "sns.set( rc={'figure.figsize':(10, 8)} )\n",
    "\n",
    "# use the seaborn scatterplot function to print out the data set\n",
    "plot = sns.scatterplot (\n",
    "        x = iris['Petal_Length'], \n",
    "        y = iris['Petal_Width'], \n",
    "        hue = iris['Class'], \n",
    "        s=100\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d75c876-4bde-46f6-9936-5ef15f2dfeff",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Build a prediction engine"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac597d48-0a53-45ce-976b-b8194923b037",
   "metadata": {},
   "source": [
    "We are really excited by our analysis and think we can easily build an Iris predictor.\n",
    "\n",
    "One way to decide the class of Iris flower is to draw a verical line separating the data points.\n",
    "\n",
    "**Exercise 01:** where would you draw a line?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a33af9e2-968c-4ba9-80a7-57ae2d03d667",
   "metadata": {},
   "outputs": [],
   "source": [
    "YOUR_PETAL_LENGTH_DECIDER = 0 ### enter a non-zero float value ####"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "92161ad9-9953-4b9c-b2df-dd2e5859656d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7f013384b100>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the data again\n",
    "plot = sns.scatterplot(x=iris['Petal_Length'], y=iris['Petal_Width'], hue=iris['Class'], s=100)\n",
    "\n",
    "# now add a verical line to the plot\n",
    "plot.axvline(YOUR_PETAL_LENGTH_DECIDER)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1874526-6a43-4094-ae7f-b5109b3a064f",
   "metadata": {},
   "source": [
    "We can now implement a simple function that uses YOUR_PETAL_LENGTH_DECIDER to determine the class of Iris"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bf56de65-46d8-4b47-b46d-1a08c318a6a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def iris_predictor( petal_length ):\n",
    "    if petal_length < YOUR_PETAL_LENGTH_DECIDER:\n",
    "        return \"Iris-setosa\"\n",
    "    else:\n",
    "        return \"Iris-versicolor\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb4bd28f-f824-4295-9c5d-c9fb3a728b5b",
   "metadata": {},
   "source": [
    "### Test your prediction engine\n",
    "\n",
    "**Exercise 02:** Test your predictor\n",
    "\n",
    "Try some different values of petal length and observe the prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c6483a6c-c4ab-4101-a7cc-7cc9c4c76e0d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Iris-versicolor'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_predictor( petal_length=1.5 ) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "dc92d0b9-d705-4b51-a7e4-5b15519a5789",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Iris-versicolor'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_predictor( petal_length=3.5 )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d3c5732-3afa-48f0-bf1b-4b3838acd54e",
   "metadata": {},
   "source": [
    "While this test is useful to ensure we haven't made any big mistakes in our logic, we should verify the predictor with new data.\n",
    "\n",
    "While you have been busy building our prediction engine, your friend has been collecting more data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3f15054e-c21c-47df-9662-6a8c36cabca7",
   "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>Petal_Length</th>\n",
       "      <th>Petal_Width</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.1</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.3</td>\n",
       "      <td>0.4</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.7</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.9</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.1</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.3</td>\n",
       "      <td>0.3</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.9</td>\n",
       "      <td>0.4</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>4.9</td>\n",
       "      <td>1.5</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>4.7</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>3.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>3.6</td>\n",
       "      <td>1.3</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>4.5</td>\n",
       "      <td>1.5</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>4.9</td>\n",
       "      <td>1.5</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>4.8</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>3.8</td>\n",
       "      <td>1.1</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>4.5</td>\n",
       "      <td>1.5</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>4.1</td>\n",
       "      <td>1.3</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>4.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>4.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>Iris-versicolor</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Petal_Length  Petal_Width            Class\n",
       "0            1.4          0.2      Iris-setosa\n",
       "1            1.4          0.2      Iris-setosa\n",
       "2            1.4          0.2      Iris-setosa\n",
       "3            1.4          0.1      Iris-setosa\n",
       "4            1.3          0.4      Iris-setosa\n",
       "5            1.7          0.2      Iris-setosa\n",
       "6            1.9          0.2      Iris-setosa\n",
       "7            1.4          0.2      Iris-setosa\n",
       "8            1.5          0.1      Iris-setosa\n",
       "9            1.3          0.2      Iris-setosa\n",
       "10           1.3          0.3      Iris-setosa\n",
       "11           1.9          0.4      Iris-setosa\n",
       "12           1.5          0.2      Iris-setosa\n",
       "13           4.9          1.5  Iris-versicolor\n",
       "14           4.7          1.6  Iris-versicolor\n",
       "15           3.5          1.0  Iris-versicolor\n",
       "16           3.6          1.3  Iris-versicolor\n",
       "17           4.5          1.5  Iris-versicolor\n",
       "18           4.9          1.5  Iris-versicolor\n",
       "19           4.8          1.4  Iris-versicolor\n",
       "20           3.8          1.1  Iris-versicolor\n",
       "21           4.5          1.5  Iris-versicolor\n",
       "22           4.1          1.3  Iris-versicolor\n",
       "23           4.0          1.2  Iris-versicolor\n",
       "24           4.2          1.3  Iris-versicolor"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# read the test dataset\n",
    "test_data = pd.read_csv(\"data/iris_test.csv\")\n",
    "\n",
    "# print the data\n",
    "test_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "7250ff33-e010-412b-b7c3-96969fa27d92",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     Iris-versicolor\n",
       "1     Iris-versicolor\n",
       "2     Iris-versicolor\n",
       "3     Iris-versicolor\n",
       "4     Iris-versicolor\n",
       "5     Iris-versicolor\n",
       "6     Iris-versicolor\n",
       "7     Iris-versicolor\n",
       "8     Iris-versicolor\n",
       "9     Iris-versicolor\n",
       "10    Iris-versicolor\n",
       "11    Iris-versicolor\n",
       "12    Iris-versicolor\n",
       "13    Iris-versicolor\n",
       "14    Iris-versicolor\n",
       "15    Iris-versicolor\n",
       "16    Iris-versicolor\n",
       "17    Iris-versicolor\n",
       "18    Iris-versicolor\n",
       "19    Iris-versicolor\n",
       "20    Iris-versicolor\n",
       "21    Iris-versicolor\n",
       "22    Iris-versicolor\n",
       "23    Iris-versicolor\n",
       "24    Iris-versicolor\n",
       "Name: Petal_Length, dtype: object"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create a new column 'predicted_class' by applying our iris_predictor function to each value of Petal_Length\n",
    "predicted_class = test_data['Petal_Length'].apply( lambda x: iris_predictor(x) )\n",
    "\n",
    "# view the new column\n",
    "predicted_class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2cbacc3d-a25c-4062-9c91-1fb50e023ff2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     False\n",
       "1     False\n",
       "2     False\n",
       "3     False\n",
       "4     False\n",
       "5     False\n",
       "6     False\n",
       "7     False\n",
       "8     False\n",
       "9     False\n",
       "10    False\n",
       "11    False\n",
       "12    False\n",
       "13     True\n",
       "14     True\n",
       "15     True\n",
       "16     True\n",
       "17     True\n",
       "18     True\n",
       "19     True\n",
       "20     True\n",
       "21     True\n",
       "22     True\n",
       "23     True\n",
       "24     True\n",
       "dtype: bool"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# compare the predicted column to our new column\n",
    "predicted_class == test_data['Class']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1488a201-1e35-41eb-9cde-b62edbb12d9c",
   "metadata": {},
   "source": [
    "Was your engine able to correctly predict all of the new data?  If not, try using a value of `2.7` for `YOUR_PETAL_LENGTH_DECIDER`.\n",
    "\n",
    "What happens if you use a value of `3.9` for `YOUR_PETAL_LENGTH_DECIDER` and rerun the notebook.  How many of the test data records were predicted right?\n",
    "\n",
    "Let's call it a day with this fantasic win in the bag :)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2511d0d7-b1bb-4139-b5c5-38be7c5ef963",
   "metadata": {},
   "source": [
    "### Summary\n",
    "\n",
    "In this notebook, you manually created an algorithm (without machine learning) that was able to predict Iris flower classes based on the petal length.\n",
    "\n",
    "**Questions:**\n",
    "\n",
    "- What happens when we have more data?  I.e. how do we re-train the predictor?\n",
    "- How would we build a predictor that has more features (e.g. Sepal Length and Sepal Width)?\n",
    "- What type of prediction did we do? (regression, classification, clustering)\n",
    "- What approach did we use? (supervised, semi-supervised, unsupervised and reinforcement)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca57aa0a-b6b5-47a9-b4f6-2e50bd9c88fc",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "### Navigation\n",
    "\n",
    "[Previous](./00-README-FIRST.ipynb) | [Home](./00-README-FIRST.ipynb) | [Next](./02_manual_method_issues.ipynb) notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "572714fa-c4a7-4116-b7bc-db71e3a386af",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}