quentinchap/linearregression_trainingcolab1.ipynb

## linearregression_trainingcolab1.ipynb
{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "simpleLinearRegression.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.6"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/quentinchap/c34fa5d6e3e8b48c6661b489957d0840/linearregression_trainingcolab1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xHOxn-g4bZrk",
        "colab_type": "text"
      },
      "source": [
        "# Machine learning - Simple Linear Regression in python with sklearn"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a1e9orvMMhH4",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "\n",
        "## Purpose\n",
        "On this post I'll focus on supervised problem with continuous valued input. In other words, I'll trained an algorithm with a dataset compose of known input and output.\n",
        "\n",
        "In this first post dedicated to machine learning I'll try to predict covid evolution thank to simple linear regression algorithm. The goal is to move forward step by step to well understand how to improve the efficiency of our prédictions.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hVFBGkCuTNa6",
        "colab_type": "text"
      },
      "source": [
        "## Choose the data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "350MD2r6TSUt",
        "colab_type": "text"
      },
      "source": [
        "I'll use data from https://www.data.gouv.fr/fr/datasets/coronavirus-covid19-evolution-par-pays-et-dans-le-monde-maj-quotidienne/\n",
        "\n",
        "On this website you'll be able to download files with following data by date and by country:\n",
        "* the number of cases\n",
        "* the number of death\n",
        "* the number of healing\n",
        "* the death rate\n",
        "* the healing rate\n",
        "* the infection rate\n",
        "\n",
        "To begin we'll try to do a simple linear regression only for one country one this file. Later we'll try to add some feature form these data and from other sources like average temperature, average salary, etc.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z-4dO1PxXmTH",
        "colab_type": "text"
      },
      "source": [
        "## Import Lib"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MImk74_xXzBO",
        "colab_type": "text"
      },
      "source": [
        "\n",
        " **Import some libs:**\n",
        "\n",
        "*   Numpy: Basic lib for scientific computing and data science.\n",
        "*   matplot: Lib to draw some graph\n",
        "*   Panda: Load and process input data\n",
        "*   Sklearn: ML lib\n",
        "*   google.colab: Use google drive\n",
        "*   datetime: Date manipulation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PdBqcXGcXvHP",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "from google.colab import drive\n",
        "import datetime"
      ],
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "92hqnaI1YH5R",
        "colab_type": "text"
      },
      "source": [
        "## Load and prepare data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UbfrQwZETGzu",
        "colab_type": "text"
      },
      "source": [
        "I will use some data files hosted on my google drive. The first step is to mount the folder to be able to retreive files."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nFB6ZtFjT9zX",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "20fd4e77-d912-45a9-be79-805872f18132"
      },
      "source": [
        "drive.mount('/content/drive', force_remount=True)"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ic2xxZMvZuRx",
        "colab_type": "text"
      },
      "source": [
        "In this file there are dates. To easly use them I'll convert these date (String) into timestamp (Number)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FI7hLsDe7e55",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "convertDate = lambda x : datetime.datetime.strptime(x, \"%Y-%m-%d\").timestamp()"
      ],
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cLlf9GUqVDw-",
        "colab_type": "text"
      },
      "source": [
        "Now open your csv thank to read_csv. Fill the four parameters:\n",
        "\n",
        "* the path of your file, here it's the path in my google doc\n",
        "* the CSV delimitator, here is \";\"\n",
        "* **error_bad_lines=False** is to avoid to stop the execution in case of bad lines\n",
        "* **converters={'Date':convertDate}** apply your converter on the column Date\n",
        "* **usecols=[0,1,2,3]** allow you to select specific columns. Here I use only the first four."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ggv2SiQHbs1Q",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        },
        "outputId": "6f5f2066-1ad8-4795-b121-75c43a4da9f5"
      },
      "source": [
        "di = pd.read_csv('/content/drive/My Drive/dataML/coronavirus.politologue.com-pays-2020-08-22.csv',\";\", error_bad_lines=False,converters={'Date':convertDate},usecols=[0,1,2,3])\n",
        "print(di)"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "               Date                 Pays  Infections  Deces\n",
            "0      1.598054e+09              Andorre        1045     53\n",
            "1      1.598054e+09  Émirats Arabes Unis       66193    370\n",
            "2      1.598054e+09          Afghanistan       37894   1385\n",
            "3      1.598054e+09   Antigua-et-Barbuda          94      3\n",
            "4      1.598054e+09              Albanie        8119    240\n",
            "...             ...                  ...         ...    ...\n",
            "34337  1.579651e+09                Macao           1      0\n",
            "34338  1.579651e+09         Corée du Sud           1      0\n",
            "34339  1.579651e+09                Japon           2      0\n",
            "34340  1.579651e+09            Hong-Kong           0      0\n",
            "34341  1.579651e+09                Chine         547     17\n",
            "\n",
            "[34342 rows x 4 columns]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TAiYfY2kcc3l",
        "colab_type": "text"
      },
      "source": [
        "Now the file is almost ready but still remain some problem with the \"Pays\" colum. In fact in this column we have categorical data, not numerical data.\n",
        "For this example we'll just keep data from France and remove this column."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AX0HLmofccfM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        },
        "outputId": "989b9998-5f3a-4765-9b02-6ebaf2127796"
      },
      "source": [
        "# Remove all ligne with a value different than \"France\" in column \"Pays\"\n",
        "clean_di = di.drop(di[di.Pays != \"France\"].index)\n",
        "# Remove the column \"Pays\"\n",
        "clean_di = clean_di.drop(axis=1,labels=\"Pays\")\n",
        "print(clean_di)"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "               Date  Infections  Deces\n",
            "60     1.598054e+09      274330  30503\n",
            "339    1.597968e+09      274330  30503\n",
            "460    1.597882e+09      269558  30474\n",
            "738    1.597795e+09      264787  30462\n",
            "861    1.597709e+09      261011  30445\n",
            "...             ...         ...    ...\n",
            "34255  1.580170e+09           4      0\n",
            "34264  1.580083e+09           3      0\n",
            "34289  1.579997e+09           3      0\n",
            "34295  1.579910e+09           3      0\n",
            "34316  1.579824e+09           2      0\n",
            "\n",
            "[212 rows x 3 columns]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oQld_occe1rq",
        "colab_type": "text"
      },
      "source": [
        "## Define the linear regression parameters"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VPnh2sE2eI_w",
        "colab_type": "text"
      },
      "source": [
        "Now I isolate features (input) in a vector x and output data in a vector y. \n",
        "Moreover I create a variable dataset to represent the global dataset. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NwIV-v1fH_QI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "dataset = clean_di;\n",
        "x = clean_di.iloc[:, clean_di.columns == 'Infections'].values\n",
        "y = clean_di.iloc[:, clean_di.columns == 'Deces'].values\n"
      ],
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "AyhQaTwP9RzG"
      },
      "source": [
        "## Splitting the dataset into the Training set and Test set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "bxOOauiN9VpC",
        "colab": {}
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "x_train, x_test, y_train, y_test = train_test_split(x, y, test_size = 1/6, random_state = 0)"
      ],
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "ZijQwFMQ9itx"
      },
      "source": [
        "## Training the Simple Linear Regression model on the Training set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "B4Aj_8YJ9l7J",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "b9b82d0e-409b-4dd0-cc15-29fe7342a957"
      },
      "source": [
        "from sklearn.linear_model import LinearRegression\n",
        "regressor = LinearRegression()\n",
        "regressor.fit(x_train, y_train)"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "-zSoMZ-P9v8t"
      },
      "source": [
        "## Visualising the Training set results"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "IAePn_u-93tI",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "adb6bfb8-1bd0-4ad6-c641-df18c5f0593f"
      },
      "source": [
        "\n",
        "plt.scatter(x_train, y_train, color = 'red')\n",
        "plt.plot(x_train, regressor.predict(x_train), color = 'blue')\n",
        "plt.xlabel('Case')\n",
        "plt.ylabel('Death')\n",
        "plt.show()"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "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": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "EUX1Vhsv97ZT"
      },
      "source": [
        "## Visualising the Test set results"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "Ze9vpBTf-Bol",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "d33b5312-7561-4f61-e930-c990813878e2"
      },
      "source": [
        "plt.scatter(x_test, y_test, color = 'red')\n",
        "plt.plot(x_train, regressor.predict(x_train), color = 'blue')\n",
        "plt.xlabel('Case')\n",
        "plt.ylabel('Death')\n",
        "plt.show()"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "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": "markdown",
      "metadata": {
        "id": "0qGbozQ0aN-u",
        "colab_type": "text"
      },
      "source": [
        "## Conclusion\n",
        "\n",
        "As you can see our prediction seems to be good but the preduction is not good. \n",
        "The Degree of the approximation polynomial function is not sufficient to give us good results. \n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UdDY57odgBCl",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        },
        "outputId": "1887f428-aeb7-4fd1-d335-68756ccf57b7"
      },
      "source": [
        "from sklearn.preprocessing import PolynomialFeatures\n",
        "from sklearn.pipeline import make_pipeline\n",
        "x_seq = np.linspace(x.min(),x.max(),300).reshape(-1,1)\n",
        "\n",
        "regressorDegree5 = make_pipeline(PolynomialFeatures(5), LinearRegression())\n",
        "regressorDegree5.fit(x, y)"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(memory=None,\n",
              "         steps=[('polynomialfeatures',\n",
              "                 PolynomialFeatures(degree=5, include_bias=True,\n",
              "                                    interaction_only=False, order='C')),\n",
              "                ('linearregression',\n",
              "                 LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None,\n",
              "                                  normalize=False))],\n",
              "         verbose=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 56
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ezJQGdargzuW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        },
        "outputId": "fb9d96ed-6b35-4d39-e1dd-60af446c0505"
      },
      "source": [
        "\n",
        "\n",
        "\n",
        "plt.scatter(x_train, y_train, color = 'red')\n",
        "plt.plot(x_seq, regressorDegree5.predict(x_seq), color = 'blue')\n",
        "plt.xlabel('Case')\n",
        "plt.ylabel('Death')\n",
        "plt.show()\n",
        "\n",
        "\n",
        "plt.scatter(x_test, y_test, color = 'red')\n",
        "plt.plot(x_seq, regressorDegree5.predict(x_seq), color = 'blue')\n",
        "plt.xlabel('Case')\n",
        "plt.ylabel('Death')\n",
        "plt.show()"
      ],
      "execution_count": 57,
      "outputs": [
        {
          "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"
          }
        },
        {
          "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": "markdown",
      "metadata": {
        "id": "JlNPaaxzf_b2",
        "colab_type": "text"
      },
      "source": [
        "Really better! \n",
        "It's all for today. Next time we'll try to add some feature to improve our prediction."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MwUz-DURSR9s",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "## Webography\n",
        "* https://www.coursera.org/learn/machine-learning\n",
        "* https://www.udemy.com/course/machinelearning/\n"
      ]
    }
  ]
}