ML_Classification.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ML_Classification.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyONHyb0MjFLpGbmOVxbtJyt",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/snoop2head/ml_classification_tutorial/blob/main/ML_Classification.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PIcE_R7JibYP",
        "outputId": "477f8978-e531-47a0-c3aa-0c224f2fdad5"
      },
      "source": [
        "# clone repository from the github\n",
        "!git clone https://github.com/snoop2head/ml_classification_tutorial"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cloning into 'ml_classification_tutorial'...\n",
            "remote: Enumerating objects: 23, done.\u001b[K\n",
            "remote: Counting objects: 100% (23/23), done.\u001b[K\n",
            "remote: Compressing objects: 100% (17/17), done.\u001b[K\n",
            "remote: Total 23 (delta 6), reused 23 (delta 6), pack-reused 0\u001b[K\n",
            "Unpacking objects: 100% (23/23), done.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H7uC2O4Spf1Z",
        "outputId": "2dafc70d-872e-44d1-fb8d-476a7730d130"
      },
      "source": [
        "# update repository when change is made\n",
        "!git pull"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Already up to date.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3pUbPcPlihbW",
        "outputId": "629b77a2-7504-4387-ed4c-975c39a50dab"
      },
      "source": [
        "# change directory to the clone repository\n",
        "%cd ml_classification_tutorial/"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/ml_classification_tutorial/ml_classification_tutorial\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8xUrEXiCtEtQ"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oTIeHeIzjM3n"
      },
      "source": [
        "### Description\n",
        "Pitch, Roll, Yaw는 고개의 회전 축을 세 개로 쪼개놓은 것이다.![68747470733a2f2f7777772e7265736561726368676174652e6e65742f70726f66696c652f5473616e675f496e675f52656e2f7075626c69636174696f6e2f3237393239313932382f6669677572652f666967312f41533a32393235333331383534363232373240313434363735363735343338382f4f72.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SogFxiJivKor"
      },
      "source": [
        "# read csv data from the cloned data\n",
        "df_bad = pd.read_csv(\"./data/df_bad.csv\", index_col=0) # set first column as index\n",
        "df_good = pd.read_csv(\"./data/df_good.csv\", index_col=0) # set first column as index"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3hjuQRJwiq3h"
      },
      "source": [
        "# designate labels for the dataset\n",
        "df_good[\"label\"] = 0\n",
        "df_bad[\"label\"] = 1"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "id": "4cl3UV2MisOw",
        "outputId": "95742445-ea0f-4121-cb33-1c35e663916d"
      },
      "source": [
        "df_bad.head(3)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Width</th>\n",
              "      <th>Height</th>\n",
              "      <th>Eye2Box</th>\n",
              "      <th>Roll</th>\n",
              "      <th>Yaw</th>\n",
              "      <th>Pitch</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>94</td>\n",
              "      <td>124</td>\n",
              "      <td>0.846603</td>\n",
              "      <td>-2.589851</td>\n",
              "      <td>-2.215973</td>\n",
              "      <td>0.846031</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>96</td>\n",
              "      <td>123</td>\n",
              "      <td>0.846122</td>\n",
              "      <td>-2.837860</td>\n",
              "      <td>-0.927490</td>\n",
              "      <td>0.826792</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>92</td>\n",
              "      <td>121</td>\n",
              "      <td>0.817107</td>\n",
              "      <td>-2.599693</td>\n",
              "      <td>-2.786713</td>\n",
              "      <td>0.877005</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Width  Height   Eye2Box      Roll       Yaw     Pitch  label\n",
              "1     94     124  0.846603 -2.589851 -2.215973  0.846031      1\n",
              "2     96     123  0.846122 -2.837860 -0.927490  0.826792      1\n",
              "3     92     121  0.817107 -2.599693 -2.786713  0.877005      1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "id": "gPOWHeb4itSa",
        "outputId": "fa608063-599c-4e8a-ec40-a8baf31f2b23"
      },
      "source": [
        "df_good.head(3)"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Width</th>\n",
              "      <th>Height</th>\n",
              "      <th>Eye2Box</th>\n",
              "      <th>Roll</th>\n",
              "      <th>Yaw</th>\n",
              "      <th>Pitch</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>89</td>\n",
              "      <td>120</td>\n",
              "      <td>0.989072</td>\n",
              "      <td>-0.577896</td>\n",
              "      <td>-1.580688</td>\n",
              "      <td>0.638711</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>105</td>\n",
              "      <td>119</td>\n",
              "      <td>0.941441</td>\n",
              "      <td>-2.540039</td>\n",
              "      <td>-0.858757</td>\n",
              "      <td>0.614961</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>89</td>\n",
              "      <td>120</td>\n",
              "      <td>0.968118</td>\n",
              "      <td>-0.905693</td>\n",
              "      <td>-1.525284</td>\n",
              "      <td>0.666339</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Width  Height   Eye2Box      Roll       Yaw     Pitch  label\n",
              "1     89     120  0.989072 -0.577896 -1.580688  0.638711      0\n",
              "2    105     119  0.941441 -2.540039 -0.858757  0.614961      0\n",
              "3     89     120  0.968118 -0.905693 -1.525284  0.666339      0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 166
        },
        "id": "KR_ItWUzoAeH",
        "outputId": "791d8d21-5821-4b6e-d001-471354438c12"
      },
      "source": [
        "# 좋은 자세 데이터값과 나쁜 자세 데이터값을 한 개의 데이터프레임으로 합친다\n",
        "df = pd.concat([df_bad, df_good])\n",
        "df.sample(4)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Width</th>\n",
              "      <th>Height</th>\n",
              "      <th>Eye2Box</th>\n",
              "      <th>Roll</th>\n",
              "      <th>Yaw</th>\n",
              "      <th>Pitch</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>189</th>\n",
              "      <td>234</td>\n",
              "      <td>219</td>\n",
              "      <td>0.824488</td>\n",
              "      <td>-12.875633</td>\n",
              "      <td>2.324890</td>\n",
              "      <td>0.743652</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>177</th>\n",
              "      <td>280</td>\n",
              "      <td>273</td>\n",
              "      <td>0.871183</td>\n",
              "      <td>-10.697147</td>\n",
              "      <td>-0.517517</td>\n",
              "      <td>1.323010</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>112</th>\n",
              "      <td>85</td>\n",
              "      <td>105</td>\n",
              "      <td>0.828753</td>\n",
              "      <td>-0.574722</td>\n",
              "      <td>-2.930893</td>\n",
              "      <td>0.857552</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>94</td>\n",
              "      <td>124</td>\n",
              "      <td>0.865523</td>\n",
              "      <td>-1.642029</td>\n",
              "      <td>-3.273911</td>\n",
              "      <td>0.862973</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     Width  Height   Eye2Box       Roll       Yaw     Pitch  label\n",
              "189    234     219  0.824488 -12.875633  2.324890  0.743652      0\n",
              "177    280     273  0.871183 -10.697147 -0.517517  1.323010      1\n",
              "112     85     105  0.828753  -0.574722 -2.930893  0.857552      0\n",
              "33      94     124  0.865523  -1.642029 -3.273911  0.862973      1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "58WiAgVEoYNx",
        "outputId": "ff522a95-17c6-4b8c-a379-9e9997bbf360"
      },
      "source": [
        "# Width, Height은 얼굴 크기\n",
        "# Eye2Box는 눈 사이의 거리\n",
        "# Roll, Yaw, Pitch는 고개의 회전\n",
        "# label은 \n",
        "list(df.columns)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['Width', 'Height', 'Eye2Box', 'Roll', 'Yaw', 'Pitch', 'label']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 254
        },
        "id": "utgNhnFtiuPY",
        "outputId": "645875cc-e95e-45d7-dd07-6f69a60b1598"
      },
      "source": [
        "# 얼굴의 크기(Width, Height), Pitch가 자세(label)과 유의미한 상관관계가 있음을 보여준다.\n",
        "df.corr()"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Width</th>\n",
              "      <th>Height</th>\n",
              "      <th>Eye2Box</th>\n",
              "      <th>Roll</th>\n",
              "      <th>Yaw</th>\n",
              "      <th>Pitch</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Width</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.844207</td>\n",
              "      <td>-0.004413</td>\n",
              "      <td>-0.544154</td>\n",
              "      <td>0.025867</td>\n",
              "      <td>0.413702</td>\n",
              "      <td>0.254726</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Height</th>\n",
              "      <td>0.844207</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.007628</td>\n",
              "      <td>-0.301882</td>\n",
              "      <td>0.091593</td>\n",
              "      <td>0.437010</td>\n",
              "      <td>0.387298</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Eye2Box</th>\n",
              "      <td>-0.004413</td>\n",
              "      <td>-0.007628</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.003241</td>\n",
              "      <td>-0.092577</td>\n",
              "      <td>0.014275</td>\n",
              "      <td>-0.030104</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Roll</th>\n",
              "      <td>-0.544154</td>\n",
              "      <td>-0.301882</td>\n",
              "      <td>0.003241</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.011580</td>\n",
              "      <td>-0.292431</td>\n",
              "      <td>-0.135549</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Yaw</th>\n",
              "      <td>0.025867</td>\n",
              "      <td>0.091593</td>\n",
              "      <td>-0.092577</td>\n",
              "      <td>0.011580</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.049483</td>\n",
              "      <td>0.183638</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Pitch</th>\n",
              "      <td>0.413702</td>\n",
              "      <td>0.437010</td>\n",
              "      <td>0.014275</td>\n",
              "      <td>-0.292431</td>\n",
              "      <td>0.049483</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.393617</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>label</th>\n",
              "      <td>0.254726</td>\n",
              "      <td>0.387298</td>\n",
              "      <td>-0.030104</td>\n",
              "      <td>-0.135549</td>\n",
              "      <td>0.183638</td>\n",
              "      <td>0.393617</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            Width    Height   Eye2Box      Roll       Yaw     Pitch     label\n",
              "Width    1.000000  0.844207 -0.004413 -0.544154  0.025867  0.413702  0.254726\n",
              "Height   0.844207  1.000000 -0.007628 -0.301882  0.091593  0.437010  0.387298\n",
              "Eye2Box -0.004413 -0.007628  1.000000  0.003241 -0.092577  0.014275 -0.030104\n",
              "Roll    -0.544154 -0.301882  0.003241  1.000000  0.011580 -0.292431 -0.135549\n",
              "Yaw      0.025867  0.091593 -0.092577  0.011580  1.000000  0.049483  0.183638\n",
              "Pitch    0.413702  0.437010  0.014275 -0.292431  0.049483  1.000000  0.393617\n",
              "label    0.254726  0.387298 -0.030104 -0.135549  0.183638  0.393617  1.000000"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A2S5xQX5oids"
      },
      "source": [
        "### Random Forest"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mVCiZuJloBly"
      },
      "source": [
        "y = df[\"label\"] # y는 대개 우리가 라벨, 지표들이다.\n",
        "X = df.drop(['label'], axis = 1) # x는 분류에 필요한 데이터들이다."
      ],
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FmgCB5PVi7rj",
        "outputId": "06dca006-0e37-450f-8918-06fb2f3531ed"
      },
      "source": [
        "y.head()"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    1\n",
              "2    1\n",
              "3    1\n",
              "4    1\n",
              "5    1\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PABqhayNi9an",
        "outputId": "34db4c4b-35ec-45f4-e259-169334e4e988"
      },
      "source": [
        "X.columns"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['Width', 'Height', 'Eye2Box', 'Roll', 'Yaw', 'Pitch'], dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q-dg6oNZi9sD"
      },
      "source": [
        "# training dataset, test dataset을 분리함\n",
        "from sklearn.model_selection import train_test_split\n",
        "X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.2)"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yRySTe61i97D",
        "outputId": "b82e8cad-70ab-41b5-b86b-1da647ae5663"
      },
      "source": [
        "# Random Forest classifier을 사용해서 분류 모델을 생성한다.\n",
        "# Classifier 안에 들어가는 arguments는 나중에 다시 설명할게요.\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "model = RandomForestClassifier(n_estimators=100)\n",
        "model.fit(X_train, y_train)"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight=None,\n",
              "                       criterion='gini', max_depth=None, max_features='auto',\n",
              "                       max_leaf_nodes=None, max_samples=None,\n",
              "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "                       min_samples_leaf=1, min_samples_split=2,\n",
              "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
              "                       n_jobs=None, oob_score=False, random_state=None,\n",
              "                       verbose=0, warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tvPgVu4Ei-Wv",
        "outputId": "be978ff0-bf6c-4d74-bcc2-924771a51818"
      },
      "source": [
        "accuracy = model.score(X_test, y_test)\n",
        "accuracy"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9141104294478528"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l5kaMUb6pL3D"
      },
      "source": [
        "### Saving and Loading Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i4g94sGppP8c"
      },
      "source": [
        ""
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "P89cH4eZi-mW",
        "outputId": "19b3ff7c-79f0-4fdc-ead5-7ffc547be79a"
      },
      "source": [
        "# save model file to the path\n",
        "from joblib import dump, load\n",
        "accuracy = str(accuracy)\n",
        "file_name = f\"./models/model{accuracy[:5]}\"\n",
        "MODEL_PATH = f\"{file_name}.joblib\"\n",
        "dump(model, MODEL_PATH) "
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['./models/model0.914.joblib']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fQJlQFdni-1b"
      },
      "source": [
        "# load model from the path\n",
        "model_loaded = load(MODEL_PATH) "
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xjk0ofX1i_w9"
      },
      "source": [
        "# predict from created model (earlier)\n",
        "y_predicted = model.predict(X_test)"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K39kFVQ_qNb8"
      },
      "source": [
        "# predict from loaded model\n",
        "y_predicted_loaded = model_loaded.predict(X_test)"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5DiLlVUsqT0Y",
        "outputId": "1a9864a5-bf33-43a8-84e7-06a188ffc8ee"
      },
      "source": [
        "# check whether loaded model yields same result as created model\n",
        "y_predicted == y_predicted_loaded"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qkTOIT6wqcZM"
      },
      "source": [
        "### Using Confusion Matrix to check accuracy"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eom1MouZqZtJ",
        "outputId": "e5fefa17-e15f-4987-f45c-17c2326c45dd"
      },
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "cm = confusion_matrix(y_test, y_predicted)\n",
        "cm"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[77,  9],\n",
              "       [ 5, 72]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303
        },
        "id": "aRXDDQvRqak_",
        "outputId": "54d9d124-76de-40e7-a95a-7a7beba35cf2"
      },
      "source": [
        "%matplotlib inline\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sn\n",
        "\n",
        "# font size setting\n",
        "SMALL_SIZE = 14\n",
        "MEDIUM_SIZE = 14.5\n",
        "LARGE_SIZE = 15.5\n",
        "\n",
        "plt.rc('font', size=SMALL_SIZE)          # controls default text sizes\n",
        "plt.rc('axes', titlesize=SMALL_SIZE)     # fontsize of the axes title\n",
        "plt.rc('axes', labelsize=MEDIUM_SIZE)    # fontsize of the x and y labels\n",
        "plt.rc('xtick', labelsize=SMALL_SIZE)    # fontsize of the tick labels\n",
        "plt.rc('ytick', labelsize=SMALL_SIZE)    # fontsize of the tick labels\n",
        "plt.rc('legend', fontsize=SMALL_SIZE)    # legend fontsize\n",
        "plt.rc('figure', titlesize=LARGE_SIZE)  # fontsize of the figure title\n",
        "\n",
        "# figure description\n",
        "title_text = f\"Confusion Matrix for Classification\"\n",
        "plt.title(label=title_text) # title\n",
        "plt.xlabel('Predicted') # y axis label\n",
        "plt.ylabel('Truth') # x axis label\n",
        "\n",
        "# heatmap annotation for color\n",
        "sn.heatmap(cm, annot=True)\n"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fe854a04358>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1cBoIT8asSXP"
      },
      "source": [
        "### Hyperparameter Tuning\n",
        "[Random Forest Hyperparameter Tuning](https://towardsdatascience.com/hyperparameter-tuning-the-random-forest-in-python-using-scikit-learn-28d2aa77dd74)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Lu4KbGngsSJQ",
        "outputId": "8f060c82-71bc-4895-c801-5484c0ec96d5"
      },
      "source": [
        "from sklearn.ensemble import RandomForestRegressor\n",
        "rf = RandomForestRegressor(random_state = 42)\n",
        "from pprint import pprint\n",
        "\n",
        "# Look at parameters used by our current forest\n",
        "print('Parameters currently in use:\\n')\n",
        "pprint(rf.get_params())"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Parameters currently in use:\n",
            "\n",
            "{'bootstrap': True,\n",
            " 'ccp_alpha': 0.0,\n",
            " 'criterion': 'mse',\n",
            " 'max_depth': None,\n",
            " 'max_features': 'auto',\n",
            " 'max_leaf_nodes': None,\n",
            " 'max_samples': None,\n",
            " 'min_impurity_decrease': 0.0,\n",
            " 'min_impurity_split': None,\n",
            " 'min_samples_leaf': 1,\n",
            " 'min_samples_split': 2,\n",
            " 'min_weight_fraction_leaf': 0.0,\n",
            " 'n_estimators': 100,\n",
            " 'n_jobs': None,\n",
            " 'oob_score': False,\n",
            " 'random_state': 42,\n",
            " 'verbose': 0,\n",
            " 'warm_start': False}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DTJSRHT2qa4c",
        "outputId": "059433ef-ed71-4d55-c6cb-87d9e51b04b7"
      },
      "source": [
        "from sklearn.model_selection import RandomizedSearchCV\n",
        "# Number of trees in random forest\n",
        "n_estimators = [int(x) for x in np.linspace(start = 100, stop = 2000, num = 10)]\n",
        "# Number of features to consider at every split\n",
        "max_features = ['auto', 'sqrt']\n",
        "# Maximum number of levels in tree\n",
        "max_depth = [int(x) for x in np.linspace(10, 110, num = 11)]\n",
        "max_depth.append(None)\n",
        "# Minimum number of samples required to split a node\n",
        "min_samples_split = [2, 5, 10]\n",
        "# Minimum number of samples required at each leaf node\n",
        "min_samples_leaf = [1, 2, 4]\n",
        "# Method of selecting samples for training each tree\n",
        "bootstrap = [True, False]\n",
        "# Create the random grid\n",
        "random_grid = {'n_estimators': n_estimators,\n",
        "               'max_features': max_features,\n",
        "               'max_depth': max_depth,\n",
        "               'min_samples_split': min_samples_split,\n",
        "               'min_samples_leaf': min_samples_leaf,\n",
        "               'bootstrap': bootstrap}\n",
        "pprint(random_grid)\n"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'bootstrap': [True, False],\n",
            " 'max_depth': [10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, None],\n",
            " 'max_features': ['auto', 'sqrt'],\n",
            " 'min_samples_leaf': [1, 2, 4],\n",
            " 'min_samples_split': [2, 5, 10],\n",
            " 'n_estimators': [100, 311, 522, 733, 944, 1155, 1366, 1577, 1788, 2000]}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GXKMetapqbKg",
        "outputId": "31af7756-5768-43c1-d953-c548a8b60782"
      },
      "source": [
        "# Use the random grid to search for best hyperparameters\n",
        "# First create the base model to tune\n",
        "rf = RandomForestRegressor()\n",
        "# Random search of parameters, using 3 fold cross validation, \n",
        "# search across 100 different combinations, and use all available cores\n",
        "rf_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, n_iter = 100, cv = 3, verbose=2, random_state=42, n_jobs = -1)\n",
        "# Fit the random search model\n",
        "rf_random.fit(X, y)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 3 folds for each of 100 candidates, totalling 300 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KcfP9PFMqba4"
      },
      "source": [
        "rf_random.best_params_"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wbW5iF5xqb9S"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tgjJXJCTtzMo"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jsIsU8Wwqbo6"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wnpRmcc5jAn3"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}