dm.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "X19bglnG4lwh"
      },
      "outputs": [],
      "source": [
        "ASSETS_DIR = \"/content/drive/MyDrive/Colab Notebooks/assets\""
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from __future__ import annotations\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "from dataclasses import dataclass, field"
      ],
      "metadata": {
        "id": "5qySRMVg47Ae"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "@dataclass\n",
        "class Node:\n",
        "  indexes: list[int] = field(default_factory=list)\n",
        "  childs: list[Node] = field(default_factory=list)\n",
        "  branch: str = \"\"\n",
        "  key: str = \"\"\n",
        "\n",
        "  def __show(self, indexes, branch, childs, key, r=0, c=0, x=0):\n",
        "    print(c * '\\t', indexes, end=\" \")\n",
        "    print(\"#\" + str(branch) if branch else \"\", end=\" \")\n",
        "    print(\"@\" + str(key) if key else \"\", end=\"\\n\")\n",
        "\n",
        "    if len(childs) == 0:\n",
        "      return\n",
        "    else:\n",
        "      c += 1\n",
        "\n",
        "    for i in childs:\n",
        "      r += 1\n",
        "      x = c * r + 1\n",
        "      self.__show(i.indexes, i.branch, i.childs, i.key, r, c, x)\n",
        "\n",
        "  @property\n",
        "  def show(self):\n",
        "    self.__show(self.indexes, self.branch, self.childs, self.key)"
      ],
      "metadata": {
        "id": "-rb-Z5k3TgA7"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def exclude_from(df, exclude_list):\n",
        "  if len(exclude_list) > 0:\n",
        "    return df.loc[:, ~df.columns.isin(exclude_list)]\n",
        "  return df"
      ],
      "metadata": {
        "id": "pKmVRG9w6YG_"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "# ID3"
      ],
      "metadata": {
        "id": "-08JjTW-TOPp"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def entropy(s):\n",
        "    p = -s.value_counts() \\\n",
        "          .div(s.count()) \\\n",
        "          .agg(lambda x: x*np.log2(x)) \\\n",
        "          .sum()\n",
        "\n",
        "    return p"
      ],
      "metadata": {
        "id": "r-9tlCx04-Mb"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def gain(s, col, target):\n",
        "    p_target = entropy(s[target])\n",
        "    keys = s[col].drop_duplicates()\n",
        "    entropies = keys.apply(lambda _: (_,\n",
        "                                      entropy(s[target][s[col][s[col] == _].index])))\n",
        "    ratios = s[col].value_counts().div(len(s))\n",
        "    res = np.subtract(p_target,\n",
        "                      entropies.apply(lambda _: ratios.loc[_[0]] * _[1])\n",
        "                               .sum())\n",
        "\n",
        "    return col, res"
      ],
      "metadata": {
        "id": "qbiUZf6U5BkW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def id3(s, *, df=None, target, exclude, verbose=False):\n",
        "    node = Node()\n",
        "\n",
        "    # stopping case\n",
        "    if df is not None and len(df) == 0:\n",
        "        return\n",
        "\n",
        "    # store first dataframe\n",
        "    if df is None:\n",
        "        df = s.copy()\n",
        "\n",
        "    node.indexes = list(s.index)\n",
        "\n",
        "    # remove excluded columns from dataframe\n",
        "    s = exclude_from(s, exclude)\n",
        "    #if len(exclude) > 0:\n",
        "    #    s = s.loc[:, ~s.columns.isin(exclude)]\n",
        "\n",
        "    columns = s.columns.drop(labels=[target])\n",
        "\n",
        "    gains = columns.map(lambda x: gain(s, x, target)) \\\n",
        "                   .to_frame(index=False) \\\n",
        "                   .set_index(0)\n",
        "\n",
        "    if verbose:\n",
        "      print(\"\\nSET:\\n\\n\", df.loc[s.index])\n",
        "      print(\"\\nGAINS: \\n\", gains)\n",
        "\n",
        "    if gains.empty or len(df[target].value_counts()) == 1:\n",
        "        if verbose:\n",
        "          print(\"\\n\\nLEAF<<<\\n\\n\")\n",
        "        node.branch = df[target].values[0]\n",
        "        return node\n",
        "\n",
        "    next_branch = gains.idxmax().values[0]\n",
        "    node.branch = next_branch\n",
        "\n",
        "    if verbose:\n",
        "      print(\"\\nNEXT BRANCH: \", next_branch)\n",
        "\n",
        "    next_S = df[next_branch].drop_duplicates() \\\n",
        "                            .apply(lambda _: df.loc[df[next_branch][df[next_branch] == _] \\\n",
        "                                               .index])\n",
        "\n",
        "    for k, v in enumerate(next_S):\n",
        "      if verbose:\n",
        "        print(f\"epoch @{k}\")\n",
        "      #print(v[next_branch].values[0])\n",
        "      n = id3(v, df=df.loc[v.index], target=target, exclude=[*exclude, next_branch])\n",
        "      n.key = v[next_branch].values[0]\n",
        "      node.childs.append(n)\n",
        "\n",
        "    return node\n"
      ],
      "metadata": {
        "id": "ZOGTkc8u5O2Q"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv(ASSETS_DIR + \"/weather.csv\")\n",
        "target = \"play\" #@param {type: \"string\"}\n",
        "exclude = [\"id\"] #@param\n",
        "\n",
        "# 1st epoch\n",
        "tree = id3(df, exclude=exclude, target=target)\n",
        "tree.show"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dNP3Jtvj5aSF",
        "outputId": "eaf67df2-962b-4bbb-fb83-a440331828c9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13] #outlook \n",
            "\t [0, 1, 7, 8, 10] #humidity @sunny\n",
            "\t\t [0, 1, 7]  @high\n",
            "\t\t [8, 10]  @normal\n",
            "\t [2, 6, 11, 12]  @overcast\n",
            "\t [3, 4, 5, 9, 13] #windy @rainy\n",
            "\t\t [3, 4, 9]  \n",
            "\t\t [5, 13]  @True\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---"
      ],
      "metadata": {
        "id": "vNRkb8aBsIuN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv(ASSETS_DIR + \"/job.csv\")\n",
        "target = \"risk\" #@param {type: \"string\"}\n",
        "exclude = [\"customer\"] #@param\n",
        "\n",
        "# 1st epoch\n",
        "tree = id3(df, exclude=exclude, target=target, verbose=False)\n",
        "tree.show"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VtwQZc5OHw2I",
        "outputId": "8d1dca22-9de6-4ca9-f4eb-c5a40f020d5b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] #debt \n",
            "\t [0, 1, 2]  @High\n",
            "\t [3, 4, 5, 6, 7, 8, 9] #revenue @Low\n",
            "\t\t [3, 4, 7, 8] #status @Low\n",
            "\t\t\t [3, 7]  @Employee\n",
            "\t\t\t [4, 8]  @Employeer\n",
            "\t\t [5, 6, 9]  @High\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---"
      ],
      "metadata": {
        "id": "w28IHvUzsLZP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv(ASSETS_DIR + \"/tennis.csv\")\n",
        "target = \"play\" #@param {type: \"string\"}\n",
        "exclude = [] #@param\n",
        "\n",
        "# 1st epoch\n",
        "tree = id3(df, exclude=exclude, target=target)\n",
        "tree.show"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0ae4e985-1b64-40c6-b6d5-782de6eb4982",
        "id": "7x3pDYwfRypB"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13] #outlook \n",
            "\t [0, 1, 7, 8, 10] #humidity @sunny\n",
            "\t\t [0, 1, 7]  @high\n",
            "\t\t [8, 10]  @normal\n",
            "\t [2, 6, 11, 12]  @overcast\n",
            "\t [3, 4, 5, 9, 13] #windy @rainy\n",
            "\t\t [3, 4, 9]  @weak\n",
            "\t\t [5, 13]  @strong\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---"
      ],
      "metadata": {
        "id": "abdi8AKhsNzo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv(ASSETS_DIR + \"/monks.csv\")\n",
        "target = \"has_tie\" #@param {type: \"string\"}\n",
        "exclude = [] #@param\n",
        "\n",
        "# 1st epoch\n",
        "tree = id3(df, exclude=exclude, target=target)\n",
        "tree.show"
      ],
      "metadata": {
        "id": "eVf4zscDKzYs"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data = {\n",
        "    \"hava\": [\"Gunesli\", \"Gunesli\", \"Ruzgarli\", \"Yagmurlu\", \"Yagmurlu\", \"Yagmurlu\", \"Ruzgarli\", \"Ruzgarli\", \"Ruzgarli\", \"Gunesli\"],\n",
        "    \"aile\": [\"Evet\", \"Hayir\", \"Evet\", \"Evet\", \"Hayir\", \"Evet\", \"Hayir\", \"Hayir\", \"Evet\", \"Hayir\"],\n",
        "    \"para\": [\"Z\", \"Z\", \"Z\", \"F\", \"Z\", \"F\", \"F\", \"Z\", \"Z\", \"Z\"],\n",
        "    \"karar\": [\"sinema\", \"tenis\", \"sinema\", \"sinema\", \"ev\", \"sinema\", \"sinema\", \"alisveris\", \"sinema\", \"tenis\"]\n",
        "}\n",
        "\n",
        "df = pd.DataFrame(data)\n",
        "\n",
        "target = \"karar\" #@param {type: \"string\"}\n",
        "exclude = [] #@param\n",
        "\n",
        "# 1st epoch\n",
        "tree = id3(df, exclude=exclude, target=target)\n",
        "tree.show"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yX35PsljFdhp",
        "outputId": "cca2703d-c19d-424c-94d6-5ddfb0d4a326"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] #hava \n",
            "\t [0, 1, 9] #aile @Gunesli\n",
            "\t\t [0] #sinema @Evet\n",
            "\t\t [1, 9] #tenis @Hayir\n",
            "\t [2, 6, 7, 8] #aile @Ruzgarli\n",
            "\t\t [2, 8] #sinema @Evet\n",
            "\t\t [6, 7] #para @Hayir\n",
            "\t\t\t [6] #sinema @F\n",
            "\t\t\t [7] #alisveris @Z\n",
            "\t [3, 4, 5] #aile @Yagmurlu\n",
            "\t\t [3, 5] #sinema @Evet\n",
            "\t\t [4] #ev @Hayir\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---"
      ],
      "metadata": {
        "id": "4fLlu9yYsPAg"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "# C4.5"
      ],
      "metadata": {
        "id": "Gn0BvhEds904"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "data = {\n",
        "    \"A1\": ['a', 'a', 'a', 'a', 'a', 'b', 'b', 'b', 'b', 'c', 'c', 'c', 'c', 'c'],\n",
        "    \"A2\": [70, 90, 85, 95, 70, 90, 78, 65, 75, 80, 70, 80, 70, 96],\n",
        "    \"A3\": [True, True, False, False, False, True, False, True, False, True, True, False, False, False],\n",
        "    \"class\": [\"class1\", \"class2\", \"class2\", \"class2\", \"class1\", \"class1\",\"class1\",\"class1\",\"class1\", \"class2\", \"class2\", \"class1\", \"class1\",\"class1\",]\n",
        "}\n",
        "\n",
        "df = pd.DataFrame(data)\n",
        "threshold = df['A2'].drop_duplicates() \\\n",
        "                    .sort_values() \\\n",
        "                    .median()\n",
        "df['A2'] = df['A2'].apply(lambda x: 'lte' if x <= threshold else 'gt')\n",
        "df[:3]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "JLBhrwd-rTLh",
        "outputId": "0b8d2553-a251-445c-b735-d58fb689761b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  A1   A2     A3   class\n",
              "0  a  lte   True  class1\n",
              "1  a   gt   True  class2\n",
              "2  a   gt  False  class2"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-aa4f5c98-9177-4b3b-9119-df784909744e\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>A1</th>\n",
              "      <th>A2</th>\n",
              "      <th>A3</th>\n",
              "      <th>class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>a</td>\n",
              "      <td>lte</td>\n",
              "      <td>True</td>\n",
              "      <td>class1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>a</td>\n",
              "      <td>gt</td>\n",
              "      <td>True</td>\n",
              "      <td>class2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>a</td>\n",
              "      <td>gt</td>\n",
              "      <td>False</td>\n",
              "      <td>class2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-aa4f5c98-9177-4b3b-9119-df784909744e')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-aa4f5c98-9177-4b3b-9119-df784909744e button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-aa4f5c98-9177-4b3b-9119-df784909744e');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "target = \"class\" #@param {type: \"string\"}\n",
        "exclude = [] #@param\n",
        "\n",
        "# 1st epoch\n",
        "tree = id3(df, exclude=exclude, target=target, verbose=False)\n",
        "tree.show"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yYIdUpDjuCto",
        "outputId": "e487a27c-9a0b-45d0-89ef-3ff2097a13bf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13] #A1 \n",
            "\t [0, 1, 2, 3, 4] #A2 @a\n",
            "\t\t [0, 4]  @lte\n",
            "\t\t [1, 2, 3]  @gt\n",
            "\t [5, 6, 7, 8]  @b\n",
            "\t [9, 10, 11, 12, 13] #A3 @c\n",
            "\t\t [9, 10]  @True\n",
            "\t\t [11, 12, 13]  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "# TWOING"
      ],
      "metadata": {
        "id": "RwhDFNZK05GU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "a = {\n",
        "    \"salary\": [\"N\", \"H\", \"L\", \"H\", \"L\", \"H\", \"L\", \"H\", \"L\", \"H\", \"L\"],\n",
        "     \"exp\": [\"M\", \"N\", \"N\", \"M\", \"M\", \"G\", \"G\", \"M\", \"M\", \"G\", \"G\"],\n",
        "     \"major\": [\"expert\", \"expert\", \"manager\", \"manager\",\"manager\",\"manager\",\"manager\", \"expert\", \"expert\",\"expert\",\"expert\"],\n",
        "     \"satisfied\": [\"Y\", \"Y\", \"Y\", \"Y\", \"Y\", \"Y\", \"Y\", \"N\", \"N\",\"N\",\"N\"]\n",
        "}\n",
        "\n",
        "target = \"satisfied\"\n",
        "exclude = []\n",
        "\n",
        "df = pd.DataFrame(a)\n",
        "df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 394
        },
        "id": "2CRwrp6106Mo",
        "outputId": "c9f710de-4ef6-4aea-c209-d5742b9313ea"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   salary exp    major satisfied\n",
              "0       N   M   expert         Y\n",
              "1       H   N   expert         Y\n",
              "2       L   N  manager         Y\n",
              "3       H   M  manager         Y\n",
              "4       L   M  manager         Y\n",
              "5       H   G  manager         Y\n",
              "6       L   G  manager         Y\n",
              "7       H   M   expert         N\n",
              "8       L   M   expert         N\n",
              "9       H   G   expert         N\n",
              "10      L   G   expert         N"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5ca364ba-0200-4757-b0fb-94726629d780\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>salary</th>\n",
              "      <th>exp</th>\n",
              "      <th>major</th>\n",
              "      <th>satisfied</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>N</td>\n",
              "      <td>M</td>\n",
              "      <td>expert</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>H</td>\n",
              "      <td>N</td>\n",
              "      <td>expert</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>L</td>\n",
              "      <td>N</td>\n",
              "      <td>manager</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>H</td>\n",
              "      <td>M</td>\n",
              "      <td>manager</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>L</td>\n",
              "      <td>M</td>\n",
              "      <td>manager</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>H</td>\n",
              "      <td>G</td>\n",
              "      <td>manager</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>L</td>\n",
              "      <td>G</td>\n",
              "      <td>manager</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>H</td>\n",
              "      <td>M</td>\n",
              "      <td>expert</td>\n",
              "      <td>N</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>L</td>\n",
              "      <td>M</td>\n",
              "      <td>expert</td>\n",
              "      <td>N</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>H</td>\n",
              "      <td>G</td>\n",
              "      <td>expert</td>\n",
              "      <td>N</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>L</td>\n",
              "      <td>G</td>\n",
              "      <td>expert</td>\n",
              "      <td>N</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5ca364ba-0200-4757-b0fb-94726629d780')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-5ca364ba-0200-4757-b0fb-94726629d780 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-5ca364ba-0200-4757-b0fb-94726629d780');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import itertools as it\n",
        "\n",
        "def get_permutations(df):\n",
        "  permutations = {}\n",
        "  for _ in df.iloc[:, :-1]:\n",
        "      p = pd.DataFrame(it.permutations(df[_].unique())) \\\n",
        "            .drop_duplicates(subset=[0]) \\\n",
        "            .reset_index(drop=True)\n",
        "      permutations[_] = p\n",
        "\n",
        "  return permutations"
      ],
      "metadata": {
        "id": "dus-i_U2aS4u"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import collections\n",
        "\n",
        "def twoing(df, *, target, exclude=[]):\n",
        "  def solve(p, k):\n",
        "    def P(r):\n",
        "      l = df[df[k].isin(r)]\n",
        "      ll = len(l)\n",
        "      pl = np.divide(ll, len(df))\n",
        "      pr = l.groupby(target) \\\n",
        "            .count()[k] \\\n",
        "            .apply(lambda x: np.divide(x, ll))\n",
        "  \n",
        "      #pr.index = pr.index.set_names([''])\n",
        "      #pr = pd.DataFrame(pr)\n",
        "      #pr['p'] = pr[k].apply(lambda x: np.divide(x, ll))\n",
        "  \n",
        "      return ll, pl, pr\n",
        "\n",
        "    gains = []\n",
        "    for i in p:\n",
        "      L, R = p.iloc[i, :1], p.iloc[i, 1:]\n",
        "      bl, lpl, lpr = P(L)\n",
        "      br, rpl, rpr = P(R)\n",
        "  \n",
        "      #print(f\"P-LEFT={lpl:.2f} \\t P-RIGHT={rpl:.2f}\")\n",
        "  \n",
        "      val = (2 * lpl * rpl * lpr.sub(rpr, fill_value=0.0)\n",
        "                                .abs() \n",
        "                                .sum())\n",
        "      \n",
        "      gains.append(val)\n",
        "      #print(f\"{val=:.2f}\")\n",
        "      #print('\\n\\n' + 32*'-')\n",
        "\n",
        "    return gains\n",
        "\n",
        "  # stopping case\n",
        "  if df is not None and len(df) == 0:\n",
        "      return\n",
        "\n",
        "  node = Node()\n",
        "  node.indexes = list(df.index)\n",
        "\n",
        "  excluded_df = exclude_from(df, exclude)\n",
        "  #print(excluded_df)\n",
        "\n",
        "  permutations = get_permutations(excluded_df)\n",
        "  gains = collections.defaultdict(list)\n",
        "  for k, v in permutations.items():\n",
        "    e = solve(v, k)\n",
        "    gains[k] = e\n",
        "\n",
        "  if len(gains) == 0 or len(df[target].value_counts()) == 1:\n",
        "      node.key = df[target].values[0]\n",
        "      return node\n",
        "  \n",
        "  next_branch = max(gains, key=gains.get)\n",
        "  next_branch_index = np.argmax(gains[next_branch])\n",
        "  #print(next_branch_index)\n",
        "  node.branch = next_branch\n",
        "\n",
        "  #print(next_branch)\n",
        "  t = permutations[next_branch]\n",
        "  #print(t)\n",
        "  L, R = (t.iloc[next_branch_index, :1], \n",
        "          t.iloc[next_branch_index, 1:])\n",
        "  #print(L, R)\n",
        "  #input(\">>>\")\n",
        "\n",
        "  exclude = [*exclude, next_branch]\n",
        "  ln = twoing(df[df[next_branch].isin(L)], target=target, exclude=exclude)\n",
        "  rn = twoing(df[df[next_branch].isin(R)], target=target, exclude=exclude)\n",
        "\n",
        "  node.childs.append(ln) \n",
        "  node.childs.append(rn) \n",
        "\n",
        "  #for i in permutations[next_branch][0]:\n",
        "  #  n = twoing(df[df[next_branch] == i], target=target, exclude=[*exclude, next_branch])\n",
        "  #  node.childs.append(n)\n",
        "\n",
        "  return node"
      ],
      "metadata": {
        "id": "vG-_3_NHn5-F"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tree = twoing(df, target=\"satisfied\")\n",
        "tree.show"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "R0KHxK1rmNEb",
        "outputId": "3993b74c-c691-4404-f086-9ff20c848c56"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] #major \n",
            "\t [0, 1, 7, 8, 9, 10] #salary \n",
            "\t\t [0]  @Y\n",
            "\t\t [1, 7, 8, 9, 10] #exp \n",
            "\t\t\t [1]  @Y\n",
            "\t\t\t [7, 8, 9, 10]  @N\n",
            "\t [2, 3, 4, 5, 6]  @Y\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_excel(ASSETS_DIR + '/twoing.xls')\n",
        "target = \"Memnun\"\n",
        "exclude = [\"Müşteri\"]\n",
        "\n",
        "tree = twoing(df, target=target, exclude=exclude)\n",
        "tree.show"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HLYYVIRMJ2bs",
        "outputId": "8190fa3a-c33f-4292-9355-0fa3f469ccd7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] #Sektör \n",
            "\t [0, 1, 7, 8, 9, 10] #Gelir \n",
            "\t\t [0]  @Evet\n",
            "\t\t [1, 7, 8, 9, 10] #Eğitim \n",
            "\t\t\t [1]  @Evet\n",
            "\t\t\t [7, 8, 9, 10]  @Hayır\n",
            "\t [2, 3, 4, 5, 6]  @Evet\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "# K-Nearest Neighbors "
      ],
      "metadata": {
        "id": "rBkok15reHsm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def knn(df, *, p, k, target):\n",
        "  excluded_df = df[p.columns]\n",
        "\n",
        "  normalize = (lambda x: (x - excluded_df.min()) \n",
        "                      / (excluded_df.max() - excluded_df.min()))\n",
        "  \n",
        "  normalized_df = normalize(excluded_df)\n",
        "  normalized_p = normalize(p)\n",
        "\n",
        "  df['d'] = (\n",
        "    df[p.columns].sub(p.squeeze())\n",
        "                 .agg(np.square)\n",
        "                 .sum(axis=1)\n",
        "                 .agg(np.sqrt)\n",
        "  )\n",
        "\n",
        "  neighbors = df.sort_values(by='d')[:k]\n",
        "  belongs = (\n",
        "      neighbors.groupby(target)\n",
        "               .count()\n",
        "               .idxmax()[0]\n",
        "  )\n",
        "\n",
        "  return belongs"
      ],
      "metadata": {
        "id": "WQpoKnGXhMGO"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data = {\n",
        "    \"x1\": [2, 3, 3, 4, 5, 6, 7, 9, 11, 10],\n",
        "    \"x2\": [4, 6, 4, 10, 8, 3, 9, 7, 7, 2],\n",
        "    \"y\": ['Bad', 'Good', 'Good', 'Bad', 'Bad', 'Good', 'Good', 'Bad', 'Bad', 'Bad']\n",
        "}\n",
        "\n",
        "p = pd.DataFrame({\n",
        "    'x1': [8],\n",
        "    'x2': [4],\n",
        "})\n",
        "\n",
        "df = pd.DataFrame(data)\n",
        "knn(df, p=p, k=4, target='y')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "No0X6BJieHan",
        "outputId": "e57b395d-f412-43f7-f6b1-ee3f856faa25"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Bad'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data = {\n",
        "    \"x1\": [10, 8, 18, 12, 3],\n",
        "    \"x2\": [5, 2, 16, 15, 15],\n",
        "    \"x3\": [19, 4, 6, 8, 15],\n",
        "    \"y\": ['Yes', 'No', 'No', 'Yes', 'Yes']\n",
        "}\n",
        "\n",
        "p = pd.DataFrame({\n",
        "    'x1': [7],\n",
        "    'x2': [8],\n",
        "    'x3': [5],\n",
        "})\n",
        "\n",
        "df = pd.DataFrame(data)\n",
        "knn(df, p=p, k=3, target='y')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "SouEd1Lh2DGc",
        "outputId": "332df0f9-dd92-4489-d5a6-b758a63b1943"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Yes'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "# K-Means"
      ],
      "metadata": {
        "id": "cmnjWTOsYRB6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def kmean(df, *, k):\n",
        "  columns = df.columns[1:]\n",
        "  clusters = [i for i in range(0, k)]\n",
        "  p = [1/len(clusters)]*len(clusters)\n",
        "\n",
        "  df['cluster'] = np.random.choice(clusters,\n",
        "                                  len(df),\n",
        "                                  p=p)\n",
        "\n",
        "\n",
        "  def solve(cpdf, least_mse=None):\n",
        "    centers = cpdf.groupby('cluster').mean(numeric_only=True)\n",
        "\n",
        "    mse = cpdf.groupby('cluster') \\\n",
        "              .apply(lambda x: np.square(x - x.mean(numeric_only=True)).sum()) \\\n",
        "              .sum(axis=1) \\\n",
        "              .sum()\n",
        "\n",
        "    if least_mse and mse >= least_mse:\n",
        "      return cpdf\n",
        "  \n",
        "    distances = cpdf.apply(lambda x: np.square(centers - x)\n",
        "                                        .sum(axis=1)**0.5, axis=1)\n",
        "\n",
        "    cpdf['cluster'] = distances.T.idxmin()\n",
        "\n",
        "    return solve(cpdf, mse)\n",
        "\n",
        "  cpdf = df.copy()\n",
        "  return solve(cpdf)\n",
        "\n",
        "data = {\n",
        "    \"x\": [4, 8, 5, 10, 11, 4, 12, 2, 3, 20, 19, 19, 15, 6, 1],\n",
        "    \"y\": [2, 4, 1, 6, 8, 3, 8, 2, 3, 20, 18, 21, 35, 32, 45],\n",
        "}\n",
        "\n",
        "df = pd.DataFrame(data)\n",
        "\n",
        "#df = pd.read_csv(ASSETS_DIR + '/housing.csv')\n",
        "#df = load_iris(as_frame=True).data\n",
        "res_df = kmean(df.copy(), k=3)\n",
        "res_df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 520
        },
        "id": "XfExDl8H8LVf",
        "outputId": "86752255-3007-41a6-b37f-2b3a26efffb0"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     x   y  cluster\n",
              "0    4   2        1\n",
              "1    8   4        1\n",
              "2    5   1        1\n",
              "3   10   6        1\n",
              "4   11   8        1\n",
              "5    4   3        1\n",
              "6   12   8        1\n",
              "7    2   2        1\n",
              "8    3   3        1\n",
              "9   20  20        0\n",
              "10  19  18        0\n",
              "11  19  21        0\n",
              "12  15  35        2\n",
              "13   6  32        2\n",
              "14   1  45        2"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-98902a6c-b152-4841-a24d-72eabdfaae5e\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>x</th>\n",
              "      <th>y</th>\n",
              "      <th>cluster</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>8</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>10</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>11</td>\n",
              "      <td>8</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>12</td>\n",
              "      <td>8</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>20</td>\n",
              "      <td>20</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>19</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>19</td>\n",
              "      <td>21</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>15</td>\n",
              "      <td>35</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>6</td>\n",
              "      <td>32</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>1</td>\n",
              "      <td>45</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-98902a6c-b152-4841-a24d-72eabdfaae5e')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-98902a6c-b152-4841-a24d-72eabdfaae5e button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-98902a6c-b152-4841-a24d-72eabdfaae5e');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.plot.scatter(x='x', y='y', colormap='rainbow')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "id": "nAI6zotpmAxi",
        "outputId": "0ddd545f-ac96-48c7-9241-0725e0ec6f03"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f662e094e20>"
            ]
          },
          "metadata": {},
          "execution_count": 22
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "res_df.plot.scatter(x='x', y='y', c='cluster', colormap='rainbow')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 275
        },
        "id": "-WPSjuMVi09j",
        "outputId": "1b3a4e4c-67df-47f9-f92d-226f4f3896b0"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f9c84c95220>"
            ]
          },
          "metadata": {},
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}