day65-random-forest/rf_learning_tool.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "rf-learning-tool.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "Pt3njPwTYgw3"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "from sklearn.datasets import make_classification"
      ],
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7BsWDgjaYncT"
      },
      "source": [
        "X,y = make_classification(n_features=5, n_redundant=0, n_informative=5,n_clusters_per_class=1)"
      ],
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 223
        },
        "id": "OksCU9vaY5-L",
        "outputId": "803427a6-cedd-49a3-a5be-d92df2cab2a0"
      },
      "source": [
        "df = pd.DataFrame(X,columns=['col1','col2','col3','col4','col5'])\n",
        "df['target'] = y\n",
        "print(df.shape)\n",
        "df.head()"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(100, 6)\n"
          ],
          "name": "stdout"
        },
        {
          "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>col1</th>\n",
              "      <th>col2</th>\n",
              "      <th>col3</th>\n",
              "      <th>col4</th>\n",
              "      <th>col5</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.551380</td>\n",
              "      <td>-1.876815</td>\n",
              "      <td>-1.807618</td>\n",
              "      <td>-1.557932</td>\n",
              "      <td>-0.039809</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-0.148084</td>\n",
              "      <td>-3.129695</td>\n",
              "      <td>0.387900</td>\n",
              "      <td>-1.322573</td>\n",
              "      <td>-1.448057</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>-3.874607</td>\n",
              "      <td>-1.727474</td>\n",
              "      <td>-0.943985</td>\n",
              "      <td>-2.895265</td>\n",
              "      <td>0.601189</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1.709298</td>\n",
              "      <td>-0.825889</td>\n",
              "      <td>2.392230</td>\n",
              "      <td>-1.987155</td>\n",
              "      <td>1.819194</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1.802057</td>\n",
              "      <td>-0.911493</td>\n",
              "      <td>0.022074</td>\n",
              "      <td>-2.545045</td>\n",
              "      <td>2.194047</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       col1      col2      col3      col4      col5  target\n",
              "0  0.551380 -1.876815 -1.807618 -1.557932 -0.039809       0\n",
              "1 -0.148084 -3.129695  0.387900 -1.322573 -1.448057       0\n",
              "2 -3.874607 -1.727474 -0.943985 -2.895265  0.601189       1\n",
              "3  1.709298 -0.825889  2.392230 -1.987155  1.819194       0\n",
              "4  1.802057 -0.911493  0.022074 -2.545045  2.194047       0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bmPSHQpAaZZw"
      },
      "source": [
        "# function for row sampling\n",
        "\n",
        "def sample_rows(df,percent):\n",
        "  return df.sample(int(percent*df.shape[0]),replace=True)"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f-ziftsvbJ6j"
      },
      "source": [
        "# function for feature sampling\n",
        "def sample_features(df,percent):\n",
        "  cols = random.sample(df.columns.tolist()[:-1],int(percent*(df.shape[1]-1)))\n",
        "  new_df = df[cols]\n",
        "  new_df['target'] = df['target']\n",
        "  return new_df"
      ],
      "execution_count": 111,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yoPV7Lkc50tL"
      },
      "source": [
        "# function for combined sampling\n",
        "\n",
        "def combined_sampling(df,row_percent,col_percent):\n",
        "  new_df = sample_rows(df,row_percent)\n",
        "  return sample_features(new_df,col_percent)"
      ],
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tAi9pPs49VsF",
        "outputId": "97589529-cfce-48d2-fb6d-03c3c6dfcf09"
      },
      "source": [
        "df1 = combined_sampling(df,0.5,0.5)"
      ],
      "execution_count": 132,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \"\"\"\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xaq3G-P19Vy6",
        "outputId": "8cd80f49-6579-4031-812e-f2199d1afb0b"
      },
      "source": [
        "df2 = combined_sampling(df,0.5,0.5)"
      ],
      "execution_count": 133,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \"\"\"\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WgPzMsz-9V5z",
        "outputId": "129228a4-037e-4c73-e81a-47ae83ebd218"
      },
      "source": [
        "df3 = combined_sampling(df,0.5,0.5)"
      ],
      "execution_count": 134,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \"\"\"\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6XMaiuf59WHb",
        "outputId": "7536fcf6-f0bb-4821-d5e5-592ff3efc8e9"
      },
      "source": [
        "print(df1.columns)\n",
        "print(df2.columns)\n",
        "print(df3.columns)"
      ],
      "execution_count": 135,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Index(['col5', 'col1', 'target'], dtype='object')\n",
            "Index(['col5', 'col1', 'target'], dtype='object')\n",
            "Index(['col3', 'col2', 'target'], dtype='object')\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "zIq1H_X99WNj",
        "outputId": "11810499-c52c-4f11-f9cd-15a5c77afb8b"
      },
      "source": [
        "df3"
      ],
      "execution_count": 141,
      "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>col3</th>\n",
              "      <th>col2</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>-0.943985</td>\n",
              "      <td>-1.727474</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>53</th>\n",
              "      <td>2.879522</td>\n",
              "      <td>-1.179645</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>-0.601854</td>\n",
              "      <td>0.240279</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>-1.209724</td>\n",
              "      <td>0.077615</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>-0.147061</td>\n",
              "      <td>-0.352877</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>-1.029536</td>\n",
              "      <td>-0.770751</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>94</th>\n",
              "      <td>-0.930578</td>\n",
              "      <td>-1.265605</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.022074</td>\n",
              "      <td>-0.911493</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>2.538559</td>\n",
              "      <td>-1.042250</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>0.595038</td>\n",
              "      <td>-0.162759</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99</th>\n",
              "      <td>1.558165</td>\n",
              "      <td>-2.026716</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>-0.256852</td>\n",
              "      <td>-3.206283</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>91</th>\n",
              "      <td>-1.182347</td>\n",
              "      <td>-1.528637</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>-1.209724</td>\n",
              "      <td>0.077615</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>-0.679621</td>\n",
              "      <td>-1.275311</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>-1.343188</td>\n",
              "      <td>-0.141246</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>54</th>\n",
              "      <td>-0.384640</td>\n",
              "      <td>-0.270707</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>79</th>\n",
              "      <td>-0.558081</td>\n",
              "      <td>-1.985678</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>70</th>\n",
              "      <td>-0.438380</td>\n",
              "      <td>-1.497126</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>89</th>\n",
              "      <td>1.305171</td>\n",
              "      <td>-0.374577</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>-0.513609</td>\n",
              "      <td>-1.946675</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>-1.774446</td>\n",
              "      <td>-0.503487</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>87</th>\n",
              "      <td>-1.347947</td>\n",
              "      <td>-0.196919</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>96</th>\n",
              "      <td>-1.153825</td>\n",
              "      <td>-0.494453</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>-1.039782</td>\n",
              "      <td>-0.106427</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>84</th>\n",
              "      <td>1.986095</td>\n",
              "      <td>-1.284953</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49</th>\n",
              "      <td>0.578898</td>\n",
              "      <td>-1.649036</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>74</th>\n",
              "      <td>1.125979</td>\n",
              "      <td>-2.884012</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.387900</td>\n",
              "      <td>-3.129695</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>73</th>\n",
              "      <td>-0.580873</td>\n",
              "      <td>-0.734898</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>86</th>\n",
              "      <td>1.411655</td>\n",
              "      <td>1.288265</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>66</th>\n",
              "      <td>0.885978</td>\n",
              "      <td>-3.087130</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>0.595038</td>\n",
              "      <td>-0.162759</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>91</th>\n",
              "      <td>-1.182347</td>\n",
              "      <td>-1.528637</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48</th>\n",
              "      <td>-1.042498</td>\n",
              "      <td>-0.064323</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>92</th>\n",
              "      <td>-1.406179</td>\n",
              "      <td>-1.957413</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>-0.147061</td>\n",
              "      <td>-0.352877</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>89</th>\n",
              "      <td>1.305171</td>\n",
              "      <td>-0.374577</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50</th>\n",
              "      <td>0.534164</td>\n",
              "      <td>-1.161223</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98</th>\n",
              "      <td>-0.158598</td>\n",
              "      <td>-1.798861</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>83</th>\n",
              "      <td>-1.550002</td>\n",
              "      <td>-1.126191</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51</th>\n",
              "      <td>-0.986968</td>\n",
              "      <td>-2.296564</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>54</th>\n",
              "      <td>-0.384640</td>\n",
              "      <td>-0.270707</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>56</th>\n",
              "      <td>-1.061683</td>\n",
              "      <td>-0.685227</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>91</th>\n",
              "      <td>-1.182347</td>\n",
              "      <td>-1.528637</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>94</th>\n",
              "      <td>-0.930578</td>\n",
              "      <td>-1.265605</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49</th>\n",
              "      <td>0.578898</td>\n",
              "      <td>-1.649036</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>91</th>\n",
              "      <td>-1.182347</td>\n",
              "      <td>-1.528637</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>-0.943985</td>\n",
              "      <td>-1.727474</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>78</th>\n",
              "      <td>-1.013512</td>\n",
              "      <td>-1.587964</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        col3      col2  target\n",
              "2  -0.943985 -1.727474       1\n",
              "53  2.879522 -1.179645       0\n",
              "17 -0.601854  0.240279       1\n",
              "12 -1.209724  0.077615       1\n",
              "8  -0.147061 -0.352877       0\n",
              "6  -1.029536 -0.770751       1\n",
              "94 -0.930578 -1.265605       1\n",
              "4   0.022074 -0.911493       0\n",
              "38  2.538559 -1.042250       0\n",
              "40  0.595038 -0.162759       0\n",
              "99  1.558165 -2.026716       0\n",
              "25 -0.256852 -3.206283       1\n",
              "91 -1.182347 -1.528637       0\n",
              "12 -1.209724  0.077615       1\n",
              "11 -0.679621 -1.275311       1\n",
              "32 -1.343188 -0.141246       1\n",
              "54 -0.384640 -0.270707       1\n",
              "79 -0.558081 -1.985678       0\n",
              "70 -0.438380 -1.497126       1\n",
              "89  1.305171 -0.374577       0\n",
              "10 -0.513609 -1.946675       1\n",
              "22 -1.774446 -0.503487       1\n",
              "87 -1.347947 -0.196919       1\n",
              "96 -1.153825 -0.494453       1\n",
              "14 -1.039782 -0.106427       1\n",
              "84  1.986095 -1.284953       0\n",
              "49  0.578898 -1.649036       0\n",
              "74  1.125979 -2.884012       0\n",
              "1   0.387900 -3.129695       0\n",
              "73 -0.580873 -0.734898       1\n",
              "86  1.411655  1.288265       0\n",
              "66  0.885978 -3.087130       0\n",
              "40  0.595038 -0.162759       0\n",
              "91 -1.182347 -1.528637       0\n",
              "48 -1.042498 -0.064323       1\n",
              "92 -1.406179 -1.957413       1\n",
              "8  -0.147061 -0.352877       0\n",
              "89  1.305171 -0.374577       0\n",
              "50  0.534164 -1.161223       0\n",
              "98 -0.158598 -1.798861       0\n",
              "83 -1.550002 -1.126191       1\n",
              "51 -0.986968 -2.296564       1\n",
              "54 -0.384640 -0.270707       1\n",
              "56 -1.061683 -0.685227       1\n",
              "91 -1.182347 -1.528637       0\n",
              "94 -0.930578 -1.265605       1\n",
              "49  0.578898 -1.649036       0\n",
              "91 -1.182347 -1.528637       0\n",
              "2  -0.943985 -1.727474       1\n",
              "78 -1.013512 -1.587964       1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 141
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ee9eCA0tbdlo"
      },
      "source": [
        "from sklearn.tree import DecisionTreeClassifier\n",
        "clf1 = DecisionTreeClassifier()\n",
        "clf2 = DecisionTreeClassifier()\n",
        "clf3 = DecisionTreeClassifier()"
      ],
      "execution_count": 136,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FsbRwvPGbskE",
        "outputId": "ea00a6c2-c5c9-413b-ec38-071670643e3e"
      },
      "source": [
        "clf1.fit(df1.iloc[:,0:2],df1.iloc[:,-1])\n",
        "clf2.fit(df2.iloc[:,0:2],df2.iloc[:,-1])\n",
        "clf3.fit(df3.iloc[:,0:2],df3.iloc[:,-1])"
      ],
      "execution_count": 137,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DecisionTreeClassifier(ccp_alpha=0.0, class_weight=None, criterion='gini',\n",
              "                       max_depth=None, max_features=None, max_leaf_nodes=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, presort='deprecated',\n",
              "                       random_state=None, splitter='best')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 137
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1TgG9KBcb75I"
      },
      "source": [
        "from sklearn.tree import plot_tree"
      ],
      "execution_count": 81,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "ezaWbbKlcDo8",
        "outputId": "eccea661-a3d3-40f6-f6de-2e90ffd2be78"
      },
      "source": [
        "plot_tree(clf1)"
      ],
      "execution_count": 138,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Text(167.4, 199.32, 'X[1] <= 0.417\\ngini = 0.487\\nsamples = 50\\nvalue = [21, 29]'),\n",
              " Text(55.800000000000004, 163.07999999999998, 'X[0] <= -1.244\\ngini = 0.198\\nsamples = 27\\nvalue = [3, 24]'),\n",
              " Text(27.900000000000002, 126.83999999999999, 'gini = 0.0\\nsamples = 2\\nvalue = [2, 0]'),\n",
              " Text(83.7, 126.83999999999999, 'X[1] <= 0.149\\ngini = 0.077\\nsamples = 25\\nvalue = [1, 24]'),\n",
              " Text(55.800000000000004, 90.6, 'gini = 0.0\\nsamples = 22\\nvalue = [0, 22]'),\n",
              " Text(111.60000000000001, 90.6, 'X[0] <= 1.129\\ngini = 0.444\\nsamples = 3\\nvalue = [1, 2]'),\n",
              " Text(83.7, 54.359999999999985, 'gini = 0.0\\nsamples = 1\\nvalue = [1, 0]'),\n",
              " Text(139.5, 54.359999999999985, 'gini = 0.0\\nsamples = 2\\nvalue = [0, 2]'),\n",
              " Text(279.0, 163.07999999999998, 'X[1] <= 2.099\\ngini = 0.34\\nsamples = 23\\nvalue = [18, 5]'),\n",
              " Text(251.10000000000002, 126.83999999999999, 'X[1] <= 1.129\\ngini = 0.245\\nsamples = 21\\nvalue = [18, 3]'),\n",
              " Text(223.20000000000002, 90.6, 'X[1] <= 0.901\\ngini = 0.397\\nsamples = 11\\nvalue = [8, 3]'),\n",
              " Text(195.3, 54.359999999999985, 'X[0] <= 1.885\\ngini = 0.198\\nsamples = 9\\nvalue = [8, 1]'),\n",
              " Text(167.4, 18.119999999999976, 'gini = 0.0\\nsamples = 8\\nvalue = [8, 0]'),\n",
              " Text(223.20000000000002, 18.119999999999976, 'gini = 0.0\\nsamples = 1\\nvalue = [0, 1]'),\n",
              " Text(251.10000000000002, 54.359999999999985, 'gini = 0.0\\nsamples = 2\\nvalue = [0, 2]'),\n",
              " Text(279.0, 90.6, 'gini = 0.0\\nsamples = 10\\nvalue = [10, 0]'),\n",
              " Text(306.90000000000003, 126.83999999999999, 'gini = 0.0\\nsamples = 2\\nvalue = [0, 2]')]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 138
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 473
        },
        "id": "hYpTwnxOcFgA",
        "outputId": "1191ef5b-7551-402d-8ccb-593eb90ad126"
      },
      "source": [
        "plot_tree(clf2)"
      ],
      "execution_count": 139,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Text(209.25, 201.90857142857143, 'X[1] <= -0.391\\ngini = 0.5\\nsamples = 50\\nvalue = [25, 25]'),\n",
              " Text(167.4, 170.84571428571428, 'gini = 0.0\\nsamples = 17\\nvalue = [0, 17]'),\n",
              " Text(251.10000000000002, 170.84571428571428, 'X[0] <= 3.267\\ngini = 0.367\\nsamples = 33\\nvalue = [25, 8]'),\n",
              " Text(209.25, 139.78285714285715, 'X[1] <= 0.588\\ngini = 0.278\\nsamples = 30\\nvalue = [25, 5]'),\n",
              " Text(167.4, 108.72, 'X[1] <= 0.357\\ngini = 0.496\\nsamples = 11\\nvalue = [6, 5]'),\n",
              " Text(125.55000000000001, 77.65714285714284, 'X[1] <= -0.15\\ngini = 0.245\\nsamples = 7\\nvalue = [6, 1]'),\n",
              " Text(83.7, 46.59428571428572, 'X[1] <= -0.162\\ngini = 0.5\\nsamples = 2\\nvalue = [1, 1]'),\n",
              " Text(41.85, 15.531428571428563, 'gini = 0.0\\nsamples = 1\\nvalue = [1, 0]'),\n",
              " Text(125.55000000000001, 15.531428571428563, 'gini = 0.0\\nsamples = 1\\nvalue = [0, 1]'),\n",
              " Text(167.4, 46.59428571428572, 'gini = 0.0\\nsamples = 5\\nvalue = [5, 0]'),\n",
              " Text(209.25, 77.65714285714284, 'gini = 0.0\\nsamples = 4\\nvalue = [0, 4]'),\n",
              " Text(251.10000000000002, 108.72, 'gini = 0.0\\nsamples = 19\\nvalue = [19, 0]'),\n",
              " Text(292.95, 139.78285714285715, 'gini = 0.0\\nsamples = 3\\nvalue = [0, 3]')]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 139
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 438
        },
        "id": "hoW0vMXScRMk",
        "outputId": "2e9d4596-08e9-46dc-c440-1584ebd9f95a"
      },
      "source": [
        "plot_tree(clf3)"
      ],
      "execution_count": 140,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Text(251.10000000000002, 199.32, 'X[0] <= -0.208\\ngini = 0.499\\nsamples = 50\\nvalue = [24, 26]'),\n",
              " Text(209.25, 163.07999999999998, 'X[1] <= -1.513\\ngini = 0.271\\nsamples = 31\\nvalue = [5, 26]'),\n",
              " Text(167.4, 126.83999999999999, 'X[1] <= -1.558\\ngini = 0.486\\nsamples = 12\\nvalue = [5, 7]'),\n",
              " Text(125.55000000000001, 90.6, 'X[1] <= -1.972\\ngini = 0.219\\nsamples = 8\\nvalue = [1, 7]'),\n",
              " Text(83.7, 54.359999999999985, 'X[1] <= -2.141\\ngini = 0.444\\nsamples = 3\\nvalue = [1, 2]'),\n",
              " Text(41.85, 18.119999999999976, 'gini = 0.0\\nsamples = 2\\nvalue = [0, 2]'),\n",
              " Text(125.55000000000001, 18.119999999999976, 'gini = 0.0\\nsamples = 1\\nvalue = [1, 0]'),\n",
              " Text(167.4, 54.359999999999985, 'gini = 0.0\\nsamples = 5\\nvalue = [0, 5]'),\n",
              " Text(209.25, 90.6, 'gini = 0.0\\nsamples = 4\\nvalue = [4, 0]'),\n",
              " Text(251.10000000000002, 126.83999999999999, 'gini = 0.0\\nsamples = 19\\nvalue = [0, 19]'),\n",
              " Text(292.95, 163.07999999999998, 'gini = 0.0\\nsamples = 19\\nvalue = [19, 0]')]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 140
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "I6TF3rdncVt0",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c55af97d-a9fd-4f7d-c740-101551f4fc3d"
      },
      "source": [
        "clf1.predict(np.array([-1.042498,\t-0.064323]).reshape(1,2))"
      ],
      "execution_count": 143,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 143
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VM7HvDCqiMo0",
        "outputId": "26db8422-8897-48fa-dff6-cf21e9aef41b"
      },
      "source": [
        "clf2.predict(np.array([-1.042498,\t-0.064323]).reshape(1,2))"
      ],
      "execution_count": 144,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 144
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ppIM05J2iTYg",
        "outputId": "7b181b3f-3ab5-46bc-da57-28b3be2402f0"
      },
      "source": [
        "clf3.predict(np.array([-1.042498,\t-0.064323]).reshape(1,2))"
      ],
      "execution_count": 145,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 145
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 465
        },
        "id": "nIrui56hiU7X",
        "outputId": "64d6f3ff-47dd-43ea-b57f-d94bf14cec1e"
      },
      "source": [
        "df.sample(14,replace=True)"
      ],
      "execution_count": null,
      "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>humidity</th>\n",
              "      <th>wind</th>\n",
              "      <th>play</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    humidity  wind  play\n",
              "0          0     1     0\n",
              "0          0     1     0\n",
              "0          0     1     0\n",
              "10         1     0     1\n",
              "13         0     0     0\n",
              "4          1     1     1\n",
              "7          0     1     0\n",
              "6          1     0     1\n",
              "4          1     1     1\n",
              "3          0     1     1\n",
              "9          1     1     1\n",
              "8          1     1     1\n",
              "10         1     0     1\n",
              "2          0     1     1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LCVUEl5okDtB"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}