ensemble.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ensemble.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "o6bneh5vhHUK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hr1RSLyG6WaH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 104
        },
        "outputId": "c3b2df27-3b77-4300-c8d0-60979e306629"
      },
      "source": [
        "values = np.random.randint(0,100,size=(100, 4))                           \n",
        "values[:5]"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[86, 69, 53, 25],\n",
              "       [14,  5, 77, 92],\n",
              "       [31, 72, 68, 41],\n",
              "       [46,  9, 14, 64],\n",
              "       [33,  3, 29, 20]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QMbK2PmvhL8R",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# synthetic data\n",
        "df = pd.DataFrame(values, columns=['y', 'model1', 'model2', 'model3'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e3TugxE4i_4v",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 202
        },
        "outputId": "894280cd-52a3-443d-9989-2a6925ceb3bd"
      },
      "source": [
        "df.head()"
      ],
      "execution_count": 18,
      "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>y</th>\n",
              "      <th>model1</th>\n",
              "      <th>model2</th>\n",
              "      <th>model3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>86</td>\n",
              "      <td>69</td>\n",
              "      <td>53</td>\n",
              "      <td>25</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>14</td>\n",
              "      <td>5</td>\n",
              "      <td>77</td>\n",
              "      <td>92</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>31</td>\n",
              "      <td>72</td>\n",
              "      <td>68</td>\n",
              "      <td>41</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>46</td>\n",
              "      <td>9</td>\n",
              "      <td>14</td>\n",
              "      <td>64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>33</td>\n",
              "      <td>3</td>\n",
              "      <td>29</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    y  model1  model2  model3\n",
              "0  86      69      53      25\n",
              "1  14       5      77      92\n",
              "2  31      72      68      41\n",
              "3  46       9      14      64\n",
              "4  33       3      29      20"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DZLejX4QhNUZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# ensemble with Linear Regression\n",
        "reg = LinearRegression().fit(df[['model1', 'model2', 'model3']][:80], df.y[:80])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cI7-WeL1-L4M",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "1af7d7d5-7304-4dd7-da94-4acd37138a47"
      },
      "source": [
        "reg.score(df[['model1', 'model2', 'model3']][:80], df.y[:80])"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.04249738515104062"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tMnK8ER6-UtC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "a9859cae-163b-44b6-8c2b-001e25e99c1c"
      },
      "source": [
        "reg.intercept_ "
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "45.16038792734951"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ub8jyp-fANwU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        },
        "outputId": "fa04d5e7-bdb7-4b29-89bf-eb2cd8437ecc"
      },
      "source": [
        "pred = reg.predict(df[['model1', 'model2', 'model3']][80:])\n",
        "pred"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([47.76702089, 37.9047255 , 44.24743597, 57.88857926, 42.55380266,\n",
              "       47.38652815, 42.14776895, 50.53976587, 49.77533985, 53.78614526,\n",
              "       54.7249244 , 54.37153023, 51.2789424 , 39.45280971, 54.57786191,\n",
              "       48.38158648, 45.56975304, 46.52607345, 50.07226277, 52.72517322])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ezMUY-5XCZrG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 381
        },
        "outputId": "a5f40e40-7953-424e-ac51-a0a1b4abf1b2"
      },
      "source": [
        "df.y[80:]"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "80    21\n",
              "81    79\n",
              "82    96\n",
              "83    56\n",
              "84    13\n",
              "85    13\n",
              "86    51\n",
              "87    93\n",
              "88    34\n",
              "89    49\n",
              "90    61\n",
              "91    70\n",
              "92    55\n",
              "93    90\n",
              "94    38\n",
              "95    91\n",
              "96     2\n",
              "97    97\n",
              "98    60\n",
              "99    50\n",
              "Name: y, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e7ai0kkbCv_W",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        },
        "outputId": "a230ccd0-7b04-4ac2-a5da-854160f1cd7d"
      },
      "source": [
        "plt.plot(pred)"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f9ef65e4a90>]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3QAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gAJJl6zrC_Mu",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        },
        "outputId": "5434f63c-f43d-4866-b679-df4359be4b15"
      },
      "source": [
        "plt.plot(df.y[80:])"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f9ef3d76518>]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhM4hoScIgtA5/x85juA49tk/OgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yUdglXcOARcC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "6e6b4cc9-bdf2-4b28-c1dc-67635a41e58d"
      },
      "source": [
        "reg.coef_"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 0.17618763,  0.06110402, -0.06658208])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mvgBS1RZBs10",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}