Dataaspirant-XGBoost-Iris-classification.ipynb

dataaspirant-xgboost-iris-classification.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Dataaspirant-XGBoost-Iris-classification.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/saimadhu-polamuri/a2c1fd3bd76c77328994af3c8298698a/dataaspirant-xgboost-iris-classification.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "M4qSfmQkMUFL"
      },
      "source": [
        "## Using XGBoost for Classification Problem Overiew in Python 3.x\n",
        "Pipeline: \n",
        "1. Import the libraries/modules needed\n",
        "2. Import data\n",
        "3. Data cleaning and pre-processing\n",
        "4. Train-test split\n",
        "5. XGBoost training and prediction\n",
        "6. Model Evaluation"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HbjZySLAMsot"
      },
      "source": [
        "###1. Import the libraries/modules needed"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SCWBNAT5qVgk"
      },
      "source": [
        "# import the libraries needed\n",
        "import xgboost as xgb\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import sklearn"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3JVGZMJsM4LG"
      },
      "source": [
        "###2. Import data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sXDE2yiZqaiv"
      },
      "source": [
        "from sklearn.datasets import load_iris\n",
        "iris = load_iris()"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9oQbiqT0rP4I",
        "outputId": "efcb1ec0-1324-40a4-ea25-dfcc19b0fdb0",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "data = pd.DataFrame(iris.data)\n",
        "data.columns = iris.feature_names\n",
        "print(data.sample(10))"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "     sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)\n",
            "19                 5.1               3.8                1.5               0.3\n",
            "86                 6.7               3.1                4.7               1.5\n",
            "36                 5.5               3.5                1.3               0.2\n",
            "27                 5.2               3.5                1.5               0.2\n",
            "83                 6.0               2.7                5.1               1.6\n",
            "74                 6.4               2.9                4.3               1.3\n",
            "129                7.2               3.0                5.8               1.6\n",
            "68                 6.2               2.2                4.5               1.5\n",
            "131                7.9               3.8                6.4               2.0\n",
            "140                6.7               3.1                5.6               2.4\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pfFXkLomNRMP"
      },
      "source": [
        "###3. Data pre-processing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AvyyajbGrZH5"
      },
      "source": [
        "## Extract the data to the required variables\n",
        "X = iris.data\n",
        "y = iris.target"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ubDPk4btup1p",
        "outputId": "d942befd-f06a-4843-e78e-0aaba3e71bac",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "iris.data.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(150, 4)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2Vw_tj50NVp3"
      },
      "source": [
        "###4. Train-test split"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NaAAbL96re8P"
      },
      "source": [
        "## split the data into train and test set. The test size here is 30% of the data\n",
        "from sklearn.model_selection import train_test_split\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d4ATnBCXrfLI"
      },
      "source": [
        "dtrain = xgb.DMatrix(X_train, label=y_train)\n",
        "dtest = xgb.DMatrix(X_test, label=y_test)"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PkfbyllWu2at",
        "outputId": "c5498079-611f-4a92-9085-c37b195f200b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "y_test.shape"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(45,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oMwwxZtSaQis"
      },
      "source": [
        "## Hyperparameter Tuning"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZchkWNeLsjKS"
      },
      "source": [
        "# some hyperparameter tuning\n",
        "param = {\n",
        "    'max_depth': 3,  # the maximum depth of each tree\n",
        "    'eta': 0.3,  # the training step for each iteration\n",
        "    'silent': 1,  # logging mode - quiet\n",
        "    'objective': 'multi:softprob',  # error evaluation for multiclass training\n",
        "    'num_class': 3}  # the number of classes that exist in this datset\n",
        "num_round = 20  # the number of training iterations"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ePU05kkxMeCV"
      },
      "source": [
        "###5. XGBoost training and prediction"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lMjkxUAns9ha"
      },
      "source": [
        "bst = xgb.train(param, dtrain, num_round)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Qbz51HPxs9k0",
        "outputId": "736c044b-56f9-435d-ccb9-ec46d2c51bdf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "preds = bst.predict(dtest)\n",
        "preds.shape"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(45, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mxHFARIssju3",
        "outputId": "aa09a27e-6584-4b94-c4c3-50a2b54686e0",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "import numpy as np\n",
        "best_preds = np.asarray([np.argmax(line) for line in preds])\n",
        "print(best_preds)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[1 0 2 1 1 0 1 2 1 1 2 0 0 0 0 1 2 1 1 2 0 2 0 2 2 2 2 2 0 0 0 0 1 0 0 2 1\n",
            " 0 0 0 2 1 1 0 0]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YMztb0avMeZy",
        "outputId": "4bc71cad-5eb8-42cb-af0d-ead9115aaa23",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "best_preds.shape"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(45,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oQfMfCbyMhnW"
      },
      "source": [
        "##6. Model Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ey7UNw30MUZI",
        "outputId": "bd6dc269-e49f-4a8f-9039-635d1f8e09ad",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "from sklearn.metrics import precision_score, f1_score\n",
        "print(precision_score(y_test, best_preds, average='macro'))\n",
        "print(f1_score(y_test, best_preds, average='weighted'))\n"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1.0\n",
            "1.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BoZY_M9qaYdf"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}