All_About_Ensemble_Learning.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "**Ensemble learning** is a machine learning technique where multiple week models a train on the data and then combine to create a single strong model.\n",
        "\n",
        "**why**:Ensemble learning reduced bias and variance while limiting the risk of overfitting\n",
        "\n",
        "**Bias** is the model tendency to consistently learn the wrong things by not taking into account all the information in the data\n",
        "\n",
        "**High Bias** is a result of the algorithm missing the relevant relationship between features and target outputs\n",
        "\n",
        "**Variance** refers to an algorithm sensitivity to small fluctuations in the training data\n",
        "\n",
        "**High Variance** is the result of algorithms fitting to random noise in the training data\n",
        "\n",
        "**Boosting** is an ensemble method that sequentially train number of weak models (often trees) to create one strong model.\n",
        "\n",
        "1)The sequential nature of training is critical as boosting effectively learned from its mistake with each iteration\n",
        "\n",
        "2)Computationally expensive to use for its sequential nature\n",
        "\n",
        "3)weighted voting\n",
        "\n",
        "4)Boosting address the under fitting problem of the model where high bais and low variance  (reducing the bias.)\n",
        "\n",
        "5)weak and shallow models\n",
        "\n",
        "**Bagging**:\n",
        "\n",
        "1)The main task of bagging is decrease the bias but not variance.\n",
        "\n",
        "2)Each of the model is built here independently.\n",
        "\n",
        "3)It is trying to solve by over fitting problem.\n",
        "4)low bias high varience models(reducing varience)\n",
        "5)weak model but deep models\n",
        "**Stacking:**\n",
        "\n",
        "Stacking, or stacked generalization, is a model ensembling technique that combines multiple base models using a meta-learner. Instead of relying on a single method like voting or averaging, stacking learns how to combine the predictions of the base models to make final predictions. It often involves splitting the training data into multiple folds, where each fold is used to train different base models, and then using another fold to train the meta-learner on the predictions of these base models. Stacking can often result in improved performance compared to individual base models\n",
        "2)Meta model for final prediction\n",
        "\n",
        "3)same as bagging, low bias high varience models(reducing varience)\n",
        "4) each model is trained on full data unlike subsets in other models"
      ],
      "metadata": {
        "id": "8eYSDEUyeuR5"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Example is random forest model by using bagging."
      ],
      "metadata": {
        "id": "0E4X5xCk2X9U"
      }
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "eFJzhF1NetOi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from sklearn import datasets\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.ensemble import RandomForestClassifier  # Import RandomForestClassifier\n",
        "import numpy as np\n",
        "\n",
        "# Load the dataset\n",
        "data = datasets.load_wine(as_frame=True)\n",
        "X = data.data\n",
        "y = data.target\n",
        "\n",
        "# Split the dataset into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=22)\n",
        "\n",
        "# Define a range of estimator values\n",
        "estimator_range = np.arange(2, 21, 2)\n",
        "\n",
        "# Lists to store models and scores\n",
        "models = []\n",
        "scores = []\n",
        "\n",
        "# Iterate over the estimator range\n",
        "for n_estimators in estimator_range:\n",
        "    # Create a Random Forest classifier\n",
        "    clf = RandomForestClassifier(n_estimators=n_estimators, random_state=22)\n",
        "\n",
        "    # Fit the model\n",
        "    clf.fit(X_train, y_train)\n",
        "\n",
        "    # Predict on the test set\n",
        "    y_pred = clf.predict(X_test)\n",
        "\n",
        "    # Calculate the accuracy score\n",
        "    score = accuracy_score(y_true=y_test, y_pred=y_pred)\n",
        "\n",
        "    # Output the number of estimators and the corresponding score\n",
        "    print(f\"Number of estimators: {n_estimators}, Score: {score}\")\n",
        "\n",
        "    # Append the model and score to their respective lists\n",
        "    models.append(clf)\n",
        "    scores.append(score)\n",
        "\n",
        "# Generate the plot of the scores against the number of estimators\n",
        "plt.figure(figsize=(9, 6))\n",
        "plt.plot(estimator_range, scores)\n",
        "\n",
        "# Adjust labels and font (to make them visible)\n",
        "plt.xlabel(\"n_estimators\", fontsize=18)\n",
        "plt.ylabel(\"score\", fontsize=18)\n",
        "plt.tick_params(labelsize=16)\n",
        "\n",
        "# Show the plot\n",
        "plt.show()\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 729
        },
        "id": "QCgC-43bc2fp",
        "outputId": "8538c632-e590-43b4-9762-2d1e9a01e225"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of estimators: 2, Score: 0.9111111111111111\n",
            "Number of estimators: 4, Score: 0.9111111111111111\n",
            "Number of estimators: 6, Score: 0.9333333333333333\n",
            "Number of estimators: 8, Score: 0.9555555555555556\n",
            "Number of estimators: 10, Score: 0.9333333333333333\n",
            "Number of estimators: 12, Score: 0.9333333333333333\n",
            "Number of estimators: 14, Score: 0.9555555555555556\n",
            "Number of estimators: 16, Score: 0.9555555555555556\n",
            "Number of estimators: 18, Score: 0.9555555555555556\n",
            "Number of estimators: 20, Score: 0.9555555555555556\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 900x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "RandomForestClassifier().get_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Gia8WSKm9sLj",
        "outputId": "1bf8c7ac-fd86-4b00-ab43-08f2840e3958"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'bootstrap': True,\n",
              " 'ccp_alpha': 0.0,\n",
              " 'class_weight': None,\n",
              " 'criterion': 'gini',\n",
              " 'max_depth': None,\n",
              " 'max_features': 'sqrt',\n",
              " 'max_leaf_nodes': None,\n",
              " 'max_samples': None,\n",
              " 'min_impurity_decrease': 0.0,\n",
              " 'min_samples_leaf': 1,\n",
              " 'min_samples_split': 2,\n",
              " 'min_weight_fraction_leaf': 0.0,\n",
              " 'n_estimators': 100,\n",
              " 'n_jobs': None,\n",
              " 'oob_score': False,\n",
              " 'random_state': None,\n",
              " 'verbose': 0,\n",
              " 'warm_start': False}"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# AdaBoost using the boosting technique."
      ],
      "metadata": {
        "id": "sY2B0pz43DvI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from sklearn import datasets\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.ensemble import AdaBoostClassifier  # Import AdaBoostClassifier\n",
        "import numpy as np\n",
        "\n",
        "# Load the dataset\n",
        "data = datasets.load_wine(as_frame=True)\n",
        "X = data.data\n",
        "y = data.target\n",
        "\n",
        "# Split the dataset into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=22)\n",
        "\n",
        "# Define a range of estimator values\n",
        "estimator_range = np.arange(2, 21, 2)\n",
        "\n",
        "# Lists to store models and scores\n",
        "models = []\n",
        "scores = []\n",
        "\n",
        "# Iterate over the estimator range\n",
        "for n_estimators in estimator_range:\n",
        "    # Create an AdaBoost classifier\n",
        "    clf = AdaBoostClassifier(n_estimators=n_estimators, random_state=22)\n",
        "\n",
        "    # Fit the model\n",
        "    clf.fit(X_train, y_train)\n",
        "\n",
        "    # Predict on the test set\n",
        "    y_pred = clf.predict(X_test)\n",
        "\n",
        "    # Calculate the accuracy score\n",
        "    score = accuracy_score(y_true=y_test, y_pred=y_pred)\n",
        "\n",
        "    # Output the number of estimators and the corresponding score\n",
        "    print(f\"Number of estimators: {n_estimators}, Score: {score}\")\n",
        "\n",
        "    # Append the model and score to their respective lists\n",
        "    models.append(clf)\n",
        "    scores.append(score)\n",
        "\n",
        "# Generate the plot of the scores against the number of estimators\n",
        "plt.figure(figsize=(9, 6))\n",
        "plt.plot(estimator_range, scores)\n",
        "\n",
        "# Adjust labels and font (to make them visible)\n",
        "plt.xlabel(\"n_estimators\", fontsize=18)\n",
        "plt.ylabel(\"score\", fontsize=18)\n",
        "plt.tick_params(labelsize=16)\n",
        "\n",
        "# Show the plot\n",
        "plt.show()\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 729
        },
        "id": "U1_0Ebp_2_5l",
        "outputId": "7d2ed5d3-ffa8-4741-d436-e7e915d9e552"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of estimators: 2, Score: 0.8222222222222222\n",
            "Number of estimators: 4, Score: 0.9111111111111111\n",
            "Number of estimators: 6, Score: 0.9111111111111111\n",
            "Number of estimators: 8, Score: 0.9111111111111111\n",
            "Number of estimators: 10, Score: 0.9111111111111111\n",
            "Number of estimators: 12, Score: 0.9111111111111111\n",
            "Number of estimators: 14, Score: 0.9111111111111111\n",
            "Number of estimators: 16, Score: 0.9111111111111111\n",
            "Number of estimators: 18, Score: 0.9111111111111111\n",
            "Number of estimators: 20, Score: 0.9111111111111111\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 900x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "AdaBoostClassifier().get_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HxoQyaQj4iM1",
        "outputId": "7e44d99e-25bc-41c6-8ffc-5c7c9c191d3e"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'algorithm': 'SAMME.R',\n",
              " 'base_estimator': 'deprecated',\n",
              " 'estimator': None,\n",
              " 'learning_rate': 1.0,\n",
              " 'n_estimators': 50,\n",
              " 'random_state': None}"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# GradientBoostingClassifier for boosting"
      ],
      "metadata": {
        "id": "hRe7ZRwZ6kh_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from sklearn import datasets\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.ensemble import GradientBoostingClassifier  # Import GradientBoostingClassifier\n",
        "import numpy as np\n",
        "\n",
        "# Load the dataset\n",
        "data = datasets.load_wine(as_frame=True)\n",
        "X = data.data\n",
        "y = data.target\n",
        "\n",
        "# Split the dataset into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=22)\n",
        "\n",
        "# Define a range of estimator values\n",
        "estimator_range = np.arange(2, 21, 2)\n",
        "\n",
        "# Lists to store models and scores\n",
        "models = []\n",
        "scores = []\n",
        "\n",
        "# Iterate over the estimator range\n",
        "for n_estimators in estimator_range:\n",
        "    # Create a Gradient Boosting classifier\n",
        "    clf = GradientBoostingClassifier(n_estimators=n_estimators, random_state=22)\n",
        "\n",
        "    # Fit the model\n",
        "    clf.fit(X_train, y_train)\n",
        "\n",
        "    # Predict on the test set\n",
        "    y_pred = clf.predict(X_test)\n",
        "\n",
        "    # Calculate the accuracy score\n",
        "    score = accuracy_score(y_true=y_test, y_pred=y_pred)\n",
        "\n",
        "    # Output the number of estimators and the corresponding score\n",
        "    print(f\"Number of estimators: {n_estimators}, Score: {score}\")\n",
        "\n",
        "    # Append the model and score to their respective lists\n",
        "    models.append(clf)\n",
        "    scores.append(score)\n",
        "\n",
        "# Generate the plot of the scores against the number of estimators\n",
        "plt.figure(figsize=(9, 6))\n",
        "plt.plot(estimator_range, scores)\n",
        "\n",
        "# Adjust labels and font (to make them visible)\n",
        "plt.xlabel(\"n_estimators\", fontsize=18)\n",
        "plt.ylabel(\"score\", fontsize=18)\n",
        "plt.tick_params(labelsize=16)\n",
        "\n",
        "# Show the plot\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 729
        },
        "id": "UY13X3eU6fZm",
        "outputId": "16ceeaaf-31e3-463c-a48f-9edf1b732158"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of estimators: 2, Score: 0.8444444444444444\n",
            "Number of estimators: 4, Score: 0.8666666666666667\n",
            "Number of estimators: 6, Score: 0.8666666666666667\n",
            "Number of estimators: 8, Score: 0.8666666666666667\n",
            "Number of estimators: 10, Score: 0.8444444444444444\n",
            "Number of estimators: 12, Score: 0.8666666666666667\n",
            "Number of estimators: 14, Score: 0.8666666666666667\n",
            "Number of estimators: 16, Score: 0.8666666666666667\n",
            "Number of estimators: 18, Score: 0.8666666666666667\n",
            "Number of estimators: 20, Score: 0.8888888888888888\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 900x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "GradientBoostingClassifier().get_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HFDh0HpS-Chu",
        "outputId": "273eb942-3bc8-47e7-dab4-cc12f1a8bb62"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'ccp_alpha': 0.0,\n",
              " 'criterion': 'friedman_mse',\n",
              " 'init': None,\n",
              " 'learning_rate': 0.1,\n",
              " 'loss': 'log_loss',\n",
              " 'max_depth': 3,\n",
              " 'max_features': None,\n",
              " 'max_leaf_nodes': None,\n",
              " 'min_impurity_decrease': 0.0,\n",
              " 'min_samples_leaf': 1,\n",
              " 'min_samples_split': 2,\n",
              " 'min_weight_fraction_leaf': 0.0,\n",
              " 'n_estimators': 100,\n",
              " 'n_iter_no_change': None,\n",
              " 'random_state': None,\n",
              " 'subsample': 1.0,\n",
              " 'tol': 0.0001,\n",
              " 'validation_fraction': 0.1,\n",
              " 'verbose': 0,\n",
              " 'warm_start': False}"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Stacking Implementation"
      ],
      "metadata": {
        "id": "Up5FaV5ZBViZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from sklearn import datasets\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.ensemble import StackingClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.ensemble import GradientBoostingClassifier\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "import numpy as np\n",
        "\n",
        "# Load the dataset\n",
        "data = datasets.load_wine(as_frame=True)\n",
        "X = data.data\n",
        "y = data.target\n",
        "\n",
        "# Split the dataset into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=22)\n",
        "\n",
        "# Define base classifiers\n",
        "base_classifiers = [\n",
        "    ('random_forest', RandomForestClassifier(n_estimators=10, random_state=22)),\n",
        "    ('gradient_boosting', GradientBoostingClassifier(n_estimators=10, random_state=22))\n",
        "]\n",
        "\n",
        "# Define the meta-learner\n",
        "meta_learner = DecisionTreeClassifier(random_state=22)\n",
        "\n",
        "# Create the stacking classifier\n",
        "clf = StackingClassifier(estimators=base_classifiers, final_estimator=meta_learner)\n",
        "\n",
        "# Fit the model\n",
        "clf.fit(X_train, y_train)\n",
        "\n",
        "# Predict on the test set\n",
        "y_pred = clf.predict(X_test)\n",
        "\n",
        "# Calculate the accuracy score\n",
        "score = accuracy_score(y_true=y_test, y_pred=y_pred)\n",
        "print(\"Stacking Classifier Accuracy:\", score)\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DYPcfxx2BUMo",
        "outputId": "19d9676d-58fc-46e9-9179-b31be40711dd"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Stacking Classifier Accuracy: 0.8888888888888888\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Create the stacking classifier\n",
        "stacking_clf = StackingClassifier(estimators=base_classifiers)\n",
        "\n",
        "# Now you can access the parameters using get_params()\n",
        "params = stacking_clf.get_params()\n",
        "print(params)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8uWe96ISC0qr",
        "outputId": "5111ba5c-c7c9-41aa-de0d-a83018cc4fbb"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'cv': None, 'estimators': [('random_forest', RandomForestClassifier(n_estimators=10, random_state=22)), ('gradient_boosting', GradientBoostingClassifier(n_estimators=10, random_state=22))], 'final_estimator': None, 'n_jobs': None, 'passthrough': False, 'stack_method': 'auto', 'verbose': 0, 'random_forest': RandomForestClassifier(n_estimators=10, random_state=22), 'gradient_boosting': GradientBoostingClassifier(n_estimators=10, random_state=22), 'random_forest__bootstrap': True, 'random_forest__ccp_alpha': 0.0, 'random_forest__class_weight': None, 'random_forest__criterion': 'gini', 'random_forest__max_depth': None, 'random_forest__max_features': 'sqrt', 'random_forest__max_leaf_nodes': None, 'random_forest__max_samples': None, 'random_forest__min_impurity_decrease': 0.0, 'random_forest__min_samples_leaf': 1, 'random_forest__min_samples_split': 2, 'random_forest__min_weight_fraction_leaf': 0.0, 'random_forest__n_estimators': 10, 'random_forest__n_jobs': None, 'random_forest__oob_score': False, 'random_forest__random_state': 22, 'random_forest__verbose': 0, 'random_forest__warm_start': False, 'gradient_boosting__ccp_alpha': 0.0, 'gradient_boosting__criterion': 'friedman_mse', 'gradient_boosting__init': None, 'gradient_boosting__learning_rate': 0.1, 'gradient_boosting__loss': 'log_loss', 'gradient_boosting__max_depth': 3, 'gradient_boosting__max_features': None, 'gradient_boosting__max_leaf_nodes': None, 'gradient_boosting__min_impurity_decrease': 0.0, 'gradient_boosting__min_samples_leaf': 1, 'gradient_boosting__min_samples_split': 2, 'gradient_boosting__min_weight_fraction_leaf': 0.0, 'gradient_boosting__n_estimators': 10, 'gradient_boosting__n_iter_no_change': None, 'gradient_boosting__random_state': 22, 'gradient_boosting__subsample': 1.0, 'gradient_boosting__tol': 0.0001, 'gradient_boosting__validation_fraction': 0.1, 'gradient_boosting__verbose': 0, 'gradient_boosting__warm_start': False}\n"
          ]
        }
      ]
    }
  ]
}